{"paper_id":"43d39237-bbe0-40da-b5bd-ddbff019f3b2","body_text":"Incidence Rate of Occult Lymph Node Metastasis in Clinical T 1-2 N 0 M 0 Small Cell Lung Cancer Patients and Radiomic Prediction Based on Contrast-enhanced CT Imaging: A Multicentre Study | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Incidence Rate of Occult Lymph Node Metastasis in Clinical T 1-2 N 0 M 0 Small Cell Lung Cancer Patients and Radiomic Prediction Based on Contrast-enhanced CT Imaging: A Multicentre Study Xu Jiang, Chao Luo, Xin Peng, Jing Zhang, Lin Yang, Li-Zhi Liu, and 8 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-3832084/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 9 You are reading this latest preprint version Abstract Background This study aimed to explore the incidence of occult lymph node metastasis (OLM) in clinical T 1 − 2 N 0 M 0 (cT 1 − 2 N 0 M 0 ) small cell lung cancer (SCLC) patients and develop machine learning prediction models using preoperative intratumoral and peritumoral contrast-enhanced CT-based radiomic data. Methods By conducting a retrospective analysis involving 242 eligible patients from 4 centres, we determined the incidence of OLM in cT 1 − 2 N 0 M 0 SCLC patients. For each lesion, two ROIs were defined using the gross tumour volume (GTV) and peritumoral volume 15 mm around the tumour (PTV). By extracting a comprehensive set of 1595 enhanced CT-based radiomic features individually from the GTV and PTV, we rigorously evaluated the model performance using various metrics, including the area under the curve (AUC), accuracy, sensitivity, specificity, calibration curve, and decision curve analysis (DCA). For enhanced clinical applicability, we formulated a nomogram that integrates clinical parameters and the rad_score (GTV and PTV). Results The initial investigation revealed a 33.9% OLM positivity rate in cT 1 − 2 N 0 M 0 SCLC patients. Our combined model, which incorporates three radiomic features from the GTV and PTV, along with two clinical parameters (smoking status and shape), exhibited robust predictive capabilities. With a peak AUC value of 0.772 in the external validation cohort, the model outperformed the alternative models. The nomogram significantly enhanced diagnostic precision for radiologists and added substantial value to the clinical decision-making process for T 1 − 2 N 0 M 0 SCLC patients. Conclusions The incidence of OLM in SCLC patients surpassed that in non-small cell lung cancer patients. The combined model demonstrated a notable generalization effect, effectively distinguishing between positive and negative OLMs in a noninvasive manner, thereby guiding individualized clinical decisions for patients with cT 1 − 2 N 0 M 0 SCLC. Small-cell lung cancer Occult lymph node metastases Contrast-enhanced computed tomography Prediction model Radiomics Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 1. Introduction Lung cancer is the most common cause of cancer-related death worldwide and accounts for approximately 18.0% of all such deaths ( 1 ). Lung cancer can be divided into small cell lung cancer (SCLC) and non-small cell lung cancer (NSCLC) based on histological subtype, and SCLC accounts for approximately 15% of all lung cancer cases ( 2 ). SCLC is a high-grade neuroendocrine carcinoma with an exceptionally poor prognosis and an overall 5-year survival rate of only 7% ( 3 ). Previous studies have shown that concurrent chemoradiotherapy (CCRT) has been the standard treatment for SCLC since the early 1990s ( 4 , 5 ). With the widespread use of CT, the number of early peripheral SCLC tumours has increased ( 6 – 8 ). Recent attention has shifted towards surgical intervention, revealing promising 5-year survival rates of up to 50% for pathological T 1 − 2 N 0 M 0 SCLC patients ( 9 – 11 ). Hence, the National Comprehensive Cancer Network guidelines recommend surgery as the primary treatment modality for pathological T 1 − 2 N 0 M 0 SCLC ( 12 , 13 ). However, in clinical practice, while imaging is effective in determining the T 1 − 2 stage, defining N 0 is challenging because surgical lymph node dissection often yields positive results when imaging does not indicate lymph node metastases ( 14 – 16 ). Occult lymph node metastasis (OLM) refers to the situation in which lymph node metastasis is not detected by presurgical imaging (mainly CT) but is confirmed by postoperative pathology ( 17 – 19 ). Preoperative imaging examinations rely mainly on CT to diagnose lymph node metastasis, but many OLMs are missed, resulting in ineffective surgery. For cT 1 − 2 N 0 M 0 SCLC patients, the presence or absence of OLM determines whether the patient is able to undergo surgery. Thoracoscopic biopsy is the \"gold standard\" for detecting the status of chest lymph nodes, but this is an invasive examination method that may lead to a series of complications, such as bleeding, infection, and pneumothorax. Therefore, identifying new and valuable noninvasive imaging methods for predicting OLM in cT 1 − 2 N 0 M 0 SCLC patients is necessary. In recent years, radiomics has emerged as a prominent area of research, allowing for the high-throughput extraction of extensive data from medical images ( 20 ). This approach enables the analysis of high-level and quantitative image features, providing a profound reflection of the spatial heterogeneity within tumour tissues ( 21 ). Previous studies have successfully developed models for predicting OLM in NSCLC patients based on radiomic features of primary lesions, demonstrating robust predictive performance ( 22 – 24 ). Additionally, peritumoral radiomics has proven equally predictive ( 25 , 26 ). Furthermore, the literature has focused predominantly on OLM in NSCLC ( 26 – 29 ), with limited studies exploring the incidence rate of OLM in SCLC. Consequently, this study focused on cT 1 − 2 N 0 M 0 SCLC patients to investigate the incidence rate of OLM in this clinical population and develop predictive models for OLM that integrate clinical parameters and intratumoral and peritumoral contrast-enhanced CT radiomics. 2. Materials and Methods 2.1. Patient Selection The institutional review boards approved this retrospective study, and the requirement for written informed consent was waived. The histopathology of the tumours was defined according to the 2015 World Health Organization definition ( 30 ), and clinical and pathological staging was based on the 8th edition of the TNM classification ( 31 ). This study retrospectively reviewed 242 patients with SCLC confirmed by postoperative pathology from four centres between January 2014 and September 2022. The inclusion criteria were as follows: ( 1 ) underwent resection of the primary lesion and systematic lymph node dissection; ( 2 ) underwent preoperative enhanced CT; and ( 3 ) had a clinical stage before surgery of T 1 − 2 N 0 M 0 . Additionally, all patients had solitary pulmonary nodules in clinical stage T 1 ~ 2 based on enhanced CT imaging and no enlarged lymph nodes (i.e., short diameter of LN ≤ 10.0 mm on CT imaging). The exclusion criteria were as follows: ( 1 ) patients who received radiotherapy, chemotherapy, or other treatments for SCLC before surgery; ( 2 ) had an interval between CT examination and surgery of more than 2 weeks; ( 3 ) had thin-layer images (with a slice thickness less than or equal to 1.25 mm) missing; and ( 4 ) had severe CT artefacts and poor image quality. For patients with multiple lesions, only SCLC lesions with conclusive pathological results were included. The patient's lymph node metastasis was obtained from the postoperative pathology report and reconfirmed by a senior pathology professor in the Department of Pathology. All patients from centre 1 were allocated to the training cohort, and patients from centres 2, 3 and 4 composed the external validation cohort (Fig. 1 ). 1). 2.2. CT Scanning and Semantic Features All enrolled patients in the four hospitals underwent a similar scan setup but with different systems and parameters (Appendix E1). The definitions and evaluation criteria for clinical parameters are described in Appendix E2. Two radiologists, each with 2 years of experience in lung imaging and blinded to the clinical and pathologic results, evaluated semantic features in the lung window setting (level, -550 HU; width, 1500 HU) and the mediastinal window setting (level, 40 HU; width, 400 HU). Any disagreements regarding the description of semantic features were resolved through consensus reading, and the results were subsequently confirmed by a chief radiologist specializing in chest imaging. 2.3. CT image acquisition and lesion segmentation Enhanced DICOM CT images were anonymized, and regions of interest (ROIs) were delineated by ITK-SANP software (version 3.8.0; https://www.itksnap.org ). According to previous studies ( 25 , 32 ), the gross tumour volume (GTV) was dilated 15 mm in three dimensions and uniformly served as the GTV + PTV (peritumoral volume). The boundaries of the lung nodules were checked by a radiologist and manually adjusted if necessary. Notably, the parts that cross the interlobar pleura, chest wall and mediastinum should be removed ( 33 ). We obtained the PTV by subtracting the two values. To ensure that the PTV did not contain any GTV components, we specifically added 1 mm to the region of the GTV (PTV = GPTV-(GTV + 1 mm)). To assess the robustness of the intratumoral and peritumoral segmentation methods, 30 patients were randomly selected, and two junior radiologists performed segmentation on their ROIs twice, with a 2-month interval between sessions, to obtain intraclass correlation coefficients (ICCs) (Appendix E3). 2.4. Radiomic features The images were resampled using linear interpolation to achieve a uniform voxel size of 1×1×1 in all three anatomical directions ( 34 ), and the image grayscale was discretized to 25 grayscales. We utilized PyRadiomics to extract features from segmented GPVs and PTVs ( 35 ). For each region, 14 shape features (3D), 18 first-order features, 24 grey level cooccurrence matrix (GLCM) features, 16 grey level run length matrix (GLRLM) features, 16 grey level size zone matrix (GLSZM) features, 14 grey level dependence matrix (GLDM) features, and 5 neighbouring grey-tone difference matrix (NGTDM) features were obtained. For each GTV region and PTV, 1595 radiomic features were extracted from the original images. A detailed list of the extracted features and the parameters used in CT image preprocessing and feature extraction is provided in Appendix E4. 2.5. Feature Selection and Modelling Before radiomic feature selection, only reproducible radiomic features with an ICC ≥ 0.8 were included in the analysis ( 36 ). Univariate analysis was subsequently performed, and features with a significance level of P < 0.01 were retained in the model. Additionally, features with a correlation coefficient exceeding 0.9 were removed from the model. The least absolute shrinkage and selection operator (LASSO) method, which compresses some independent variables with little or no influence on 0, was used to select the most robust and nonredundant radiomic features from the extracted features ( 37 ). Clinical parameters were evaluated in combination with selected radiomic features in the multivariable logistic regression model for predicting the presence of OLM (Appendix E5). The radiomics model's output scores (Rad_score) were merged with the clinical features to construct the nomogram. This comprehensive model effectively integrated both radiomic and clinical parameters, enhancing the overall predictive power and potential clinical utility of the model. 2.6. Statistical analysis Continuous variables were compared using two-sample t tests, whereas categorical variables were assessed through chi-square tests and Fisher’s exact tests. The GTV, PTV, GTV + PTV, clinical, and combined models were established and verified by using R (version 4.1.0, https://www , rproject.org). To assess the model's performance, the area under the ROC curve (AUC) was utilized, with the optimal cut-off value determined using the derived Youden index. Additionally, the model's accuracy, sensitivity, specificity, negative predictive value, and positive predictive value were computed. Decision curve analysis (DCA) was performed according to the methods of a previous study ( 38 ). The Delong test was used to compare different AUC values ( 39 ). A two-tailed p value of less than 0.05 was considered to indicate statistical significance. 3. Results 3.1. Patient characteristics A total of 242 patients (186 men, 56 women) with 242 lesions (OLM-negative, 160; OLM-positive, 82) were included after the application of the exclusion criteria (Fig. 1 ). The rate of OLM positivity in all patients was 33.9% (82/242). The characteristics of all the patients are detailed in Table 1 . Table 1 The Parameters in the Development of the Clinical Model Training cohort External validation cohort OLM (-) OLM (+) p value OLM (-) OLM (+) p value N = 98 N = 60 N = 62 N = 22 Gender 0.052 0.546 Female 18 (18.367%) 20 (33.333%) 12 (19.355%) 6 (27.273%) Male 80 (81.633%) 40 (66.667%) 50 (80.645%) 16 (72.727%) Age 63.000 [57.000;68.000] 61.500 [55.000;66.500] 0.204 62.500 [56.000;67.750] 63.000 [57.000;67.000] 0.579 Smoke 0.002* 0.012 No 24 (24.490%) 30 (50.000%) 14 (22.581%) 12 (54.545%) Yes 74 (75.510%) 30 (50.000%) 48 (77.419%) 10 (45.455%) Family history 0.655 0.371 No 81 (82.653%) 52 (86.667%) 58 (93.548%) 19 (86.364%) Yes 17 (17.347%) 8 (13.333%) 4 (6.452%) 3 (13.636%) Lobe 0.753 0.472 RUL 27 (27.551%) 12 (20.000%) 17 (27.419%) 4 (18.182%) RML 6 (6.122%) 3 (5.000%) 1 (1.613%) 1 (4.545%) RLL 19 (19.388%) 16 (26.667%) 13 (20.968%) 4 (18.182%) LUL 27 (27.551%) 16 (26.667%) 15 (24.194%) 9 (40.909%) LLL 19 (19.388%) 13 (21.667%) 16 (25.806%) 4 (18.182%) Location 0.023* 0.071 Center 20 (20.408%) 23 (38.333%) 6 (9.677%) 6 (27.273%) Peripheral 78 (79.592%) 37 (61.667%) 56 (90.323%) 16 (72.727%) Clinical stage T 0.051 0.085 1 61 (62.245%) 27 (45.000%) 33 (53.226%) 17 (77.273%) 2 37 (37.755%) 33 (55.000%) 29 (46.774%) 5 (22.727%) Shape 0.033* 0.677 Irregular 67 (68.367%) 30 (50.000%) 47 (75.806%) 15 (68.182%) Round or oval 31 (31.633%) 30 (50.000%) 15 (24.194%) 7 (31.818%) Branching 0.423 0.280 No 83 (84.694%) 47 (78.333%) 60 (96.774%) 20 (90.909%) Yes 15 (15.306%) 13 (21.667%) 2 (3.226%) 2 (9.091%) Lobulation 0.581 0.506 No 9 (9.184%) 8 (13.333%) 9 (14.516%) 5 (22.727%) Yes 89 (90.816%) 52 (86.667%) 53 (85.484%) 17 (77.273%) Spiculation sign 0.501 1.000 No 69 (70.408%) 46 (76.667%) 45 (72.581%) 16 (72.727%) Yes 29 (29.592%) 14 (23.333%) 17 (27.419%) 6 (27.273%) Calcification 0.635 1.000 No 96 (97.959%) 58 (96.667%) 60 (96.774%) 22 (100.000%) Yes 2 (2.041%) 2 (3.333%) 2 (3.226%) 0 (0.000%) Concavity 1.000 . No 97 (98.980%) 59 (98.333%) 62 (100.000%) 22 (100.000%) Yes 1 (1.020%) 1 (1.667%) 0 (0.000%) 0 (0.000%) Carcinoma 0.527 1.000 No 41 (41.837%) 29 (48.333%) 27 (43.548%) 10 (45.455%) Yes 57 (58.163%) 31 (51.667%) 35 (56.452%) 12 (54.545%) Bronchial 0.053 0.345 No 64 (65.306%) 29 (48.333%) 40 (64.516%) 11 (50.000%) Yes 34 (34.694%) 31 (51.667%) 22 (35.484%) 11 (50.000%) Air-bronchogram 0.302 . No 94 (95.918%) 55 (91.667%) 62 (100.000%) 22 (100.000%) Yes 4 (4.082%) 5 (8.333%) 0 (0.000%) 0 (0.000%) Obstructive: 0.022* 0.770 No 74 (75.510%) 34 (56.667%) 48 (77.419%) 18 (81.818%) Yes 24 (24.490%) 26 (43.333%) 14 (22.581%) 4 (18.182%) Enhancement Heterogeneity 0.015* 0.112 homogeneous 27 (27.551%) 6 (10.000%) 9 (14.516%) 7 (31.818%) not homogeneous 71 (72.449%) 54 (90.000%) 53 (85.484%) 15 (68.182%) BVB 0.951 0.761 No 57 (58.163%) 36 (60.000%) 41 (66.129%) 16 (72.727%) Yes 41 (41.837%) 24 (40.000%) 21 (33.871%) 6 (27.273%) Pleural Retraction 0.800 0.720 No 86 (87.755%) 51 (85.000%) 53 (85.484%) 20 (90.909%) Yes 12 (12.245%) 9 (15.000%) 9 (14.516%) 2 (9.091%) Pleural Attachment 0.069 1.000 No 75 (76.531%) 37 (61.667%) 46 (74.194%) 17 (77.273%) Yes 23 (23.469%) 23 (38.333%) 16 (25.806%) 5 (22.727%) Peripheral Emphysema 0.124 0.469 No 57 (58.163%) 43 (71.667%) 32 (51.613%) 14 (63.636%) Yes 41 (41.837%) 17 (28.333%) 30 (48.387%) 8 (36.364%) Interstitial Pneumonia 0.749 0.053 No 92 (93.878%) 55 (91.667%) 61 (98.387%) 19 (86.364%) Yes 6 (6.122%) 5 (8.333%) 1 (1.613%) 3 (13.636%) *Significant difference (p < 0.05). RUL, right upper lung; RML, right middle lung; RLL, right lower lung; LUL, left upper lung; LLL, left lower lung; BVB, bronchovascular bundle thickening Clinical parameters that were significant at p ≤ 0.05 in univariate analysis were subsequently entered into multivariate analysis. Furthermore, smoking status and shape (P < 0.05) were also included in the multivariate analysis. 3.2. Feature Selection and Model Construction Figure 2 shows the workflow of the radiomic feature analysis. The radiomic features were selected by using the ICC, univariate analysis, multivariate analysis, correlation analysis, LASSO regression, and multivariable logistic regression. Finally, two radiomic features were selected and utilized to construct the GTV model, and three radiomic features were used to construct the PTV model. After the five selected features were integrated, correlation analysis and multivariate stepwise regression were performed, resulting in the final selection of three features—i.e., the MCC from the GTV, median and IDN from the PTV—for use in constructing the GTV + PTV model. The combined model was established by incorporating one GTV radiomic feature (MCC), two PTV radiomic features (median and IDN), and two clinical parameters. The parameters of the five models are detailed in Fig. 3 and Appendix E6. 3.3. Performance and Comparison of the Three Models for All Patients All five models have some predictive power. The AUC values of the combined model were 0.774 and 0.772 in the training and external testing cohorts, respectively, performing better than any other models in our study. All the results regarding predictive performance are enumerated in Table 2 , and the ROC curves are shown in Fig. 4 . The correlation analysis of clinical and radiomic features is indicated in Appendix E5. With respect to the training cohort, the DeLong test revealed significant differences in the area under the curve (AUC) (p < 0.05) between the GTV model and the combined model, between the PTV model and combined model, between the GTV + PTV model and combined model, and between the clinical model and combined model. With respect to the external testing cohort, the Delong test revealed that there were significant differences in the area under the curve (AUC) between the GTV + PTV model and the combined model (p < 0.05) (Appendix E7). The DCAs (Fig. 5 ) revealed that when the probability of the threshold was between approximately 10 ~ 80%, the net benefits of the combined model and the GTV + PTV model for the prediction of OLM were greater than those of any other type of model. The calibration plot revealed good predictive accuracy between the actual probability and the predicted probability of the GTV + PTV model and the combined model (Fig. 5 ). Table 2 Performance of the Five Models AUC (95% CI) ACC SEN SPE PPV NPV Training cohort (n = 158) GTV 0.657(0.571–0.744) 0.633 0.685 0.606 0.474 0.788 PTV 0.687(0.599–0.776) 0.646 0.667 0.635 0.486 0.786 GTV + PTV 0.697(0.610–0.785) 0.665 0.722 0.635 0.506 0.815 Clinical 0.689(0.605–0.773) 0.627 0.741 0.567 0.471 0.808 Combined 0.774(0.696–0.853) 0.759 0.63 0.827 0.654 0.811 External test cohort (n = 84) GTV 0.663(0.535–0.792) 0.607 0.607 0.607 0.436 0.756 PTV 0.673(0.541–0.804) 0.607 0.536 0.643 0.429 0.735 GTV + PTV 0.703(0.575–0.831) 0.667 0.607 0.696 0.5 0.78 Clinical 0.675(0.557–0.792) 0.595 0.786 0.5 0.44 0.824 Combined 0.772(0.656–0.887) 0.762 0.679 0.804 0.633 0.833 4. Discussion In this multicentre study, for the first time, we revealed a 33.9% positivity rate for OLM among patients with cT 1 − 2 N 0 M 0 SCLC. This observation suggested that OLM in cT 1 − 2 N 0 M 0 SCLC surpasses the prevalence observed in NSCLC, where it ranges from 16–29% ( 26 – 29 , 40 – 45 ). In addition, we addressed a crucial challenge in managing cT 1 ~ 2 N 0 M 0 SCLC by developing and validating predictive models for OLM. Our combined model consistently outperformed the other models in our study, as evidenced by the higher area under the curve (AUC) values in both the training cohort (0.774) and the validation cohort (0.772). According to the model, patients identified as having a higher risk of OLM in cT 1 − 2 N 0 M 0 SCLC could avoid unnecessary surgeries. Conversely, individuals assessed as having a lower risk might be more confidently considered for surgical resection, with the potential for significant improvements in survival. This study offers a promising approach for accurately identifying OLM in cT 1 − 2 N 0 M 0 SCLC patients, guiding personalized treatment decisions. In terms of clinical parameters and conventional CT features, smoking status and tumour shape exhibited noteworthy differences in predicting OLM status, while the remaining features showed no significant distinctions. Our study revealed a tendency for patients with OLM to be smokers, a well-established association with the occurrence and progression of SCLC ( 12 , 46 , 47 ). In contrast, nonsmokers were more inclined to have OLM in the context of NSCLC ( 40 ). Additionally, we reported for the first time that round and oval tumour shapes hold notable significance, suggesting that lesions with regular shapes may be at a greater risk of OLM positivity. Previous studies on risk factors for NSCLC have been abundant, but uniform results have been lacking, implicating factors such as female sex, adenocarcinoma, and a small tumour size ( 40 , 44 , 48 , 49 ). Thus, there are significant differences in the risk factors for OLM between these two distinct pathological types of lung cancer, enhancing our understanding of OLM in lung cancer patients. However, further studies with larger patient cohorts are necessary to validate our findings. Following our in-depth analysis, three radiomic features were ultimately selected: one was the MCC according to intratumoral imaging, and the other two were the median and IDN according to peritumoral imaging. The MCC and IDN are obtained from the grey level co-occurrence matrix (GLCM), which is a texture analysis method that describes spatial relationships between neighbouring pixels to reflect the internal texture of tumours, such as the complexity and heterogeneity of the tumour regions and peritumor regions ( 50 ). The first-order median is expressed as the median grey-level intensity of all pixels in the ROI, which can reflect the textural characteristics of regions around the lesions. Based on our results, even if the imaging findings may be similar in both groups, the MCC, median, and IDN in OLM may serve as noninvasive predictive biomarkers and provide additional information from both intra- and peritumor radiomic data. Surgical intervention has emerged as a highly impactful therapeutic modality for cT 1 − 2 N 0 M 0 SCLC, emphasizing the crucial role of promoting this approach in clinical practice ( 51 ). A critical consideration is determining whether patients lacking observable lymph node enlargement on routine imaging harbour OLMs. In clinical practice, CT serves as the primary method for preoperative lymph node staging in patients with lung cancer, commonly using a short-axis diameter greater than 1 cm as a threshold( 40 ). However, OLM cannot be assessed. PET-CT supplements this assessment but has inherent limitations, including false positives and negatives ( 52 ), and its high cost makes widespread clinical application challenging ( 40 ). This study presented a comprehensive noninvasive model that demonstrated good performance across all dimensions, boasting a specificity of 82.7% and a sensitivity of 63% on the training dataset. The model exhibited robustness during external validation. A comparison of the five models using DCA curves revealed that our integrated model outperformed the others within the 10%-80% probability threshold range. Within this clinically relevant range, both the specificity and sensitivity were considered acceptable. The effectiveness of all the models underscores the inadequacy of traditional methods for evaluating our study's specific objectives, establishing a significant advantage in efficacy for our research in the field. These findings hold substantial clinical relevance for cT 1 − 2 N 0 M 0 SCLC patients identified with preoperative negative OLMs, emphasizing the potential impact of timely surgical intervention. The comparison of AUC values between radiomic models utilized the pairwise DeLong test, with corresponding p values provided in Supplement E7. In the training cohort, four p values derived from the DeLong test were less than 0.05, suggesting that the combination of the GTV and PTV radiomic features with clinical parameters may surpass the performance of a single radiomic feature. Notably, clinical parameters play a pivotal role in predicting OLM in patients with cT 1 − 2 N 0 M 0 SCLC. In the validation cohort, only the comparison between the GTV model and the combined model yielded a p value less than 0.05, implying that the combined model has a greater predictive ability than the solitary GTV model. Additionally, for the first time, we employed radiomics for prediction, revealing its pioneering significance. Future enhancements with increased data volume will further boost the model's efficacy. Moreover, our research offers notable advantages. First, this study pioneers the application of radiomic techniques for OLM prediction in cT 1 − 2 N 0 M 0 SCLC patients, advancing clinical diagnostic proficiency and facilitating precise decision-making and tailored treatment. Second, as a multicentre study, this study included a substantial sample size within the realm of enhanced CT-based radiomic research. Third, the combined model consistently demonstrated stable and commendable performance across both the internal training and external validation datasets, while the nomogram provided visualization and served as a valuable clinical tool for predicting OLM in presurgical cT 1 − 2 N 0 M 0 SCLC patients. Our study has certain limitations. First, selection bias is inherent in retrospective studies and is exacerbated by a modest sample size. Second, diverse machine parameters across different hospitals may introduce variations. Nevertheless, this variability contributes to the robustness of the models we trained. Third, compared to traditional radiomic methods, deep learning enhances the prediction model's performance to some extent. Emerging machine learning technologies such as convolutional neural networks are particularly suitable for classification tasks. Our future studies will prioritize data from larger sample sizes and incorporate deep learning applications to further enhance the robustness and performance of our models. In conclusion, OLM is not rare and has a greater incidence than NSCLC. Our combined model, which incorporates both intra- and peritumoral radiomic features based on contrast-enhanced CT imaging, serves as a valuable tool for discerning OLM in cT 1 − 2 N 0 M 0 SCLC patients, guiding individualized clinical decisions. Abbreviations SCLC Small cell lung cancer NSCLC Non-small cell lung cancer OLM Occult lymph node metastases CT Computed tomography GTV Gross tumour volume PTV Peritumoural volume GPTV Gross tumour volume ICC Intraclass correlation coefficient LASSO Least absolute shrinkage and selection operator AUC Area under curve Declarations Funding This study received funding from the Chinese Academy of Medical Science Innovation Fund for Medical Sciences (CIFMS) (2021-I2M-C&T-B-061) and the Beijing Hope Run Special Fund of Cancer Foundation of China (LC2022A22). Ethical approval The study was approved by the Cancer Hospital, Chinese Academy of Medical Sciences Ethics Commission (NCC2021C-213). Acknowledgements We thank all the study participants and referring technicians for their participation in this study. Availability of data and materials The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request. Competing interests The authors of this manuscript declare that they have no competing interests. Consent for publication Not applicable. Consent for publication Informed consent was waived due to the retrospective nature. Authors’ contributions XT.Y., L.Z. and M.L: Study design. JL.R. and MW.L.: data analysis. X.J., C.L. and X.P.: manuscript writing. J.Z., LZ.L. and YF.C.: data collection. MW.L., L.M. and JM.J.: study supervision and manuscript revising. All authors read and approved the final manuscript. References Sung H, Ferlay J, Siegel RL, Laversanne M, Soerjomataram I, Jemal A, et al. Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2021;71(3):209–49. Thai AA, Solomon BJ, Sequist LV, Gainor JF, Heist RS. Lung cancer. Lancet. 2021;398(10299):535–54. Siegel RL, Miller KD, Fuchs HE, Jemal A. Cancer statistics, 2022. CA Cancer J Clin. 2022;72(1):7–33. Lad T, Piantadosi S, Thomas P, Payne D, Ruckdeschel J, Giaccone G. A prospective randomized trial to determine the benefit of surgical resection of residual disease following response of small cell lung cancer to combination chemotherapy. Chest. 1994;106(6 Suppl):320s–3s. Barnes H, See K, Barnett S, Manser R. Surgery for limited-stage small-cell lung cancer. Cochrane Database Syst Rev. 2017;4(4):Cd011917. Miyauchi E, Motoi N, Ono H, Ninomiya H, Ohyanagi F, Nishio M, et al. Distinct Characteristics of Small Cell Lung Cancer Correlate With Central or Peripheral Origin: Subtyping Based on Location and Expression of Transcription Factor TTF-1. Med (Baltim). 2015;94(51):e2324. Nobashi T, Koyasu S, Nakamoto Y, Kubo T, Ishimori T, Kim YH, et al. Prognostic value of fluorine-18 fludeoxyglucose positron emission tomography parameters differs according to primary tumour location in small-cell lung cancer. Br J Radiol. 2016;89(1059):20150618. Kanaji N, Sakai K, Ueda Y, Miyawaki H, Ishii T, Watanabe N, et al. Peripheral-type small cell lung cancer is associated with better survival and higher frequency of interstitial lung disease. Lung Cancer. 2017;108:126–33. Uprety D, Arjyal L, Vallatharasu Y, Bista A, Borgert A, Fitzsimmons AJ, et al. Utilization of Surgery and Its Impact on Survival in Patients With Early Stage Small-cell Lung Cancer in the United States. Clin Lung Cancer. 2020;21(2):186–93e2. Yang CJ, Chan DY, Shah SA, Yerokun BA, Wang XF, D'Amico TA, et al. Long-term survival after surgery compared with concurrent chemoradiation for node-negative small cell lung cancer. Ann Surg. 2018;268(6):1105–12. Yang CF, Chan DY, Speicher PJ, Gulack BC, Wang X, Hartwig MG, et al. Role of adjuvant therapy in a population-based cohort of patients with early-stage small-cell lung cancer. J Clin Oncol. 2016;34(10):1057–64. Ganti AKP, Loo BW, Bassetti M, Blakely C, Chiang A, D'Amico TA, et al. Small Cell Lung Cancer, Version 2.2022, NCCN Clinical Practice Guidelines in Oncology. J Natl Compr Canc Netw. 2021;19(12):1441–64. Jett JR, Schild SE, Kesler KA, Kalemkerian GP. Treatment of small cell lung cancer: Diagnosis and management of lung cancer, 3rd ed: American college of chest physicians evidence-based clinical practice guidelines. Chest. 2013;143(5 Suppl):e400S-e19S. Moon Y, Kim KS, Lee KY, Sung SW, Kim YK, Park JK. Clinicopathologic factors associated with occult lymph node metastasis in patients with clinically diagnosed no lung adenocarcinoma. Ann Thorac Surg. 2016;101(5):1928–35. Li L, Ren S, Zhang Y, Guan Y, Zhao J, Liu J, et al. Risk factors for predicting the occult nodal metastasis in T1-2N0M0 NSCLC patients staged by PET/CT: Potential value in the clinic. Lung Cancer. 2013;81(2):213–7. Stamatis G. Staging of lung cancer: The role of noninvasive, minimally invasive and invasive techniques. Eur Respir J. 2015;46(2):521–31. Li M, Wu N, Liu Y, Zheng R, Liang Y, Zhang W, et al. Regional nodal staging with 18F-FDG PET-CT in non-small cell lung cancer: Additional diagnostic value of CT attenuation and dual-time-point imaging. Eur J Radiol. 2012;81(8):1886–90. Schmidt-Hansen M, Baldwin DR, Hasler E, Zamora J, Abraira V, Roqué IFM. PET-CT for assessing mediastinal lymph node involvement in patients with suspected resectable non-small cell lung cancer. Cochrane Database Syst Rev. 2014;2014(11):Cd009519. Kayawake H, Chen-Yoshikawa TF, Oda H, Motoyama H, Hamaji M, Menju T, et al. Complications of endobronchial ultrasound-guided transbronchial needle aspiration. Ann Thorac Surg. 2017;104(5):e363–e5. Gillies RJ, Kinahan PE, Hricak H. Radiomics: Images Are More than Pictures, They Are Data. Radiology. 2016;278(2):563–77. Gillies RJ, Kinahan PE, Hricak H, Radiomics. Images are more than pictures, they are data. Radiology. 2016;278(2):563–77. Zhang R, Zhang R, Luan T, Liu B, Zhang Y, Xu Y, et al. A Radiomics Nomogram for Preoperative Prediction of Clinical Occult Lymph Node Metastasis in cT1-2N0M0 Solid Lung Adenocarcinoma. Cancer Manag Res. 2021;13:8157–67. Sha X, Gong G, Qiu Q, Duan J, Li D, Yin Y. Discrimination of mediastinal metastatic lymph nodes in NSCLC based on radiomic features in different phases of CT imaging. BMC Med Imaging. 2020;20(1):12. Wang L, Li T, Hong J, Zhang M, Ouyang M, Zheng X, et al. (18)F-FDG PET-based radiomics model for predicting occult lymph node metastasis in clinical N0 solid lung adenocarcinoma. Quant Imaging Med Surg. 2021;11(1):215–25. Wang X, Zhao X, Li Q, Xia W, Peng Z, Zhang R, et al. Can peritumoral radiomics increase the efficiency of the prediction for lymph node metastasis in clinical stage T1 lung adenocarcinoma on CT? Eur Radiol. 2019;29(11):6049–58. Tunali I, Hall LO, Napel S, Cherezov D, Guvenis A, Gillies RJ, et al. Stability and reproducibility of computed tomography radiomic features extracted from peritumoral regions of lung cancer lesions. Med Phys. 2019;46(11):5075–85. Ishida T, Yano T, Maeda K, Kaneko S, Tateishi M, Sugimachi K. Strategy for lymphadenectomy in lung cancer three centimeters or less in diameter. Ann Thorac Surg. 1990;50(5):708–13. Inoue M, Minami M, Shiono H, Sawabata N, Ideguchi K, Okumura M. Clinicopathologic study of resected, peripheral, small-sized, non-small cell lung cancer tumors of 2 cm or less in diameter: Pleural invasion and increase of serum carcinoembryonic antigen level as predictors of nodal involvement. J Thorac Cardiovasc Surg. 2006;131(5):988–93. Bao F, Yuan P, Yuan X, Lv X, Wang Z, Hu J. Predictive risk factors for lymph node metastasis in patients with small size non-small cell lung cancer. J Thorac Dis. 2014;6(12):1697–703. Travis WD, Brambilla E, Nicholson AG, Yatabe Y, Austin JHM, Beasley MB, et al. The 2015 World Health Organization classification of lung tumors: Impact of genetic, clinical and radiologic advances since the 2004 classification. J Thorac Oncol. 2015;10(9):1243–60. Rami-Porta R, Asamura H, Travis WD, Rusch VW. Lung cancer - major changes in the american joint committee on cancer eighth edition cancer staging manual. CA Cancer J Clin. 2017;67(2):138–55. Wang T, She Y, Yang Y, Liu X, Chen S, Zhong Y, et al. Radiomics for survival risk stratification of clinical and pathologic stage IA pure-solid non-small cell lung cancer. Radiology. 2022;302(2):425–34. Zhuo Y, Feng M, Yang S, Zhou L, Ge D, Lu S, et al. Radiomics nomograms of tumors and peritumoral regions for the preoperative prediction of spread through air spaces in lung adenocarcinoma. Transl Oncol. 2020;13(10):100820. Shafiq-Ul-Hassan M, Zhang GG, Latifi K, Ullah G, Hunt DC, Balagurunathan Y, et al. Intrinsic dependencies of CT radiomic features on voxel size and number of gray levels. Med Phys. 2017;44(3):1050–62. van Griethuysen JJM, Fedorov A, Parmar C, Hosny A, Aucoin N, Narayan V, et al. Computational Radiomics System to Decode the Radiographic Phenotype. Cancer Res. 2017;77(21):e104–e7. Choe J, Lee SM, Do KH, Lee G, Lee JG, Lee SM, et al. Deep Learning-based Image Conversion of CT Reconstruction Kernels Improves Radiomics Reproducibility for Pulmonary Nodules or Masses. Radiology. 2019;292(2):365–73. Vasquez MM, Hu C, Roe DJ, Halonen M, Guerra S. Measurement error correction in the least absolute shrinkage and selection operator model when validation data are available. Stat Methods Med Res. 2019;28(3):670–80. Vickers AJ, Elkin EB. Decision curve analysis: a novel method for evaluating prediction models. Med Decis Making. 2006;26(6):565–74. DeLong ER, DeLong DM, Clarke-Pearson DL. Comparing the areas under two or more correlated receiver operating characteristic curves: A nonparametric approach. Biometrics. 1988;44(3):837–45. Cai JS, Yang F, Wang X. Occult lymph node metastasis is not a favorable factor for resected NSCLC patients. BMC Cancer. 2023;23(1):822. Beyaz F, Verhoeven RLJ, Schuurbiers OCJ, Verhagen A, van der Heijden E. Occult lymph node metastases in clinical N0/N1 NSCLC; A single center in-depth analysis. Lung Cancer. 2020;150:186–94. Gwóźdź P, Pasieka-Lis M, Kołodziej K, Pankowski J, Banaś R, Wiłkojć M, et al. Prognosis of Patients With Stages I and II Non-Small Cell Lung Cancer With Nodal Micrometastases. Ann Thorac Surg. 2018;105(5):1551–7. Haque W, Singh A, Park HS, Teh BS, Butler EB, Zeng M, et al. Quantifying the rate and predictors of occult lymph node involvement in patients with clinically node-negative non-small cell lung cancer. Acta Oncol. 2022;61(4):403–8. Deng J, Zhong Y, Wang T, Yang M, Ma M, Song Y, et al. Lung cancer with PET/CT-defined occult nodal metastasis yields favourable prognosis and benefits from adjuvant therapy: a multicentre study. Eur J Nucl Med Mol Imaging. 2022;49(7):2414–24. Moon Y, Choi SY, Park JK, Lee KY. Risk Factors for Occult Lymph Node Metastasis in Peripheral Non-Small Cell Lung Cancer with Invasive Component Size 3 cm or Less. World J Surg. 2020;44(5):1658–65. Dingemans AC, Früh M, Ardizzoni A, Besse B, Faivre-Finn C, Hendriks LE, et al. Small-cell lung cancer: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up(☆). Ann Oncol. 2021;32(7):839–53. Rudin CM, Brambilla E, Faivre-Finn C, Sage J. Small-cell lung cancer. Nat Rev Dis Primers. 2021;7(1):3. He XQ, Luo TY, Li X, Huo JW, Gong JW, Li Q. Clinicopathological and computed tomographic features associated with occult lymph node metastasis in patients with peripheral solid non-small cell lung cancer. Eur J Radiol. 2021;144:109981. Gómez-Caro A, Boada M, Cabañas M, Sanchez M, Arguis P, Lomeña F, et al. False-negative rate after positron emission tomography/computer tomography scan for mediastinal staging in cI stage non-small-cell lung cancer. Eur J Cardiothorac Surg. 2012;42(1):93–100. discussion. Kim JY, Park JE, Jo Y, Shim WH, Nam SJ, Kim JH, et al. Incorporating diffusion- and perfusion-weighted MRI into a radiomics model improves diagnostic performance for pseudoprogression in glioblastoma patients. Neuro Oncol. 2019;21(3):404–14. Lim K, Hsin MKY, Commentary. Resection for small cell lung cancer should be offered more often, and preferably anatomical. J Thorac Cardiovasc Surg. 2021;161(4):1495–6. Navani N, Spiro SG. PET scanning is important in lung cancer; but it has its limitations. Respirology. 2010;15(8):1149–51. Additional Declarations No competing interests reported. Supplementary Files SupplementaryMaterials.docx Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 01 Feb, 2024 Reviews received at journal 01 Feb, 2024 Reviewers agreed at journal 24 Jan, 2024 Reviews received at journal 20 Jan, 2024 Reviewers agreed at journal 10 Jan, 2024 Reviewers invited by journal 07 Jan, 2024 Editor assigned by journal 04 Jan, 2024 Submission checks completed at journal 03 Jan, 2024 First submitted to journal 03 Jan, 2024 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {\"props\":{\"pageProps\":{\"initialData\":{\"identity\":\"rs-3832084\",\"acceptedTermsAndConditions\":true,\"allowDirectSubmit\":false,\"archivedVersions\":[],\"articleType\":\"Research Article\",\"associatedPublications\":[],\"authors\":[{\"id\":265294746,\"identity\":\"1dbb3dc3-ce03-4ae9-8a65-831e1fb1a3f1\",\"order_by\":0,\"name\":\"Xu Jiang\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"National Cancer Center, Chinese Academy of Medical Sciences and Peking Union Medical College\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Xu\",\"middleName\":\"\",\"lastName\":\"Jiang\",\"suffix\":\"\"},{\"id\":265294747,\"identity\":\"6a1b9cd2-2221-4e0c-a337-652dfe3d39f6\",\"order_by\":1,\"name\":\"Chao Luo\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Sun Yat-sen University Cancer Center\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Chao\",\"middleName\":\"\",\"lastName\":\"Luo\",\"suffix\":\"\"},{\"id\":265294748,\"identity\":\"f62312e9-cd4e-4120-bb54-594b7d1f147a\",\"order_by\":2,\"name\":\"Xin Peng\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"The Third People's Hospital of Chengdu\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Xin\",\"middleName\":\"\",\"lastName\":\"Peng\",\"suffix\":\"\"},{\"id\":265294749,\"identity\":\"a3df9781-ec19-4d39-affc-2abc6605380e\",\"order_by\":3,\"name\":\"Jing Zhang\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Shanxi Cancer Hospital, Shanxi Medical University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Jing\",\"middleName\":\"\",\"lastName\":\"Zhang\",\"suffix\":\"\"},{\"id\":265294750,\"identity\":\"abdb5e31-43cb-454c-ba2f-8f2b2204e74d\",\"order_by\":4,\"name\":\"Lin Yang\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"National Cancer Center, Chinese Academy of Medical Sciences and Peking Union Medical College\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Lin\",\"middleName\":\"\",\"lastName\":\"Yang\",\"suffix\":\"\"},{\"id\":265294751,\"identity\":\"af72b18a-bdf1-4d4c-81bf-ffe3740f54e1\",\"order_by\":5,\"name\":\"Li-Zhi Liu\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Sun Yat-sen University Cancer Center\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Li-Zhi\",\"middleName\":\"\",\"lastName\":\"Liu\",\"suffix\":\"\"},{\"id\":265294752,\"identity\":\"1130cfee-54d0-436a-8ba0-b9a5d96fcb33\",\"order_by\":6,\"name\":\"Yan-Fen Cui\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Shanxi Cancer Hospital, Shanxi Medical University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Yan-Fen\",\"middleName\":\"\",\"lastName\":\"Cui\",\"suffix\":\"\"},{\"id\":265294753,\"identity\":\"5e70b843-4110-4497-b4dc-7aa0d6923e45\",\"order_by\":7,\"name\":\"Meng-Wen Liu\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"National Cancer Center, Chinese Academy of Medical Sciences and Peking Union Medical College\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Meng-Wen\",\"middleName\":\"\",\"lastName\":\"Liu\",\"suffix\":\"\"},{\"id\":265294754,\"identity\":\"7cbc0a14-12cd-437c-95e9-b3db82375a44\",\"order_by\":8,\"name\":\"Lei Miao\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"National Cancer Center, Chinese Academy of Medical Sciences and Peking Union Medical College\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Lei\",\"middleName\":\"\",\"lastName\":\"Miao\",\"suffix\":\"\"},{\"id\":265294755,\"identity\":\"9f2ab703-120a-45d7-ad30-a45779bcc052\",\"order_by\":9,\"name\":\"Jiu-Ming Jiang\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"National Cancer Center, Chinese Academy of Medical Sciences and Peking Union Medical College\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Jiu-Ming\",\"middleName\":\"\",\"lastName\":\"Jiang\",\"suffix\":\"\"},{\"id\":265294756,\"identity\":\"ef8dbf6d-4ac2-42d1-8b57-adb1e420c67c\",\"order_by\":10,\"name\":\"Jia-Liang Ren\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"GE HealthCare\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Jia-Liang\",\"middleName\":\"\",\"lastName\":\"Ren\",\"suffix\":\"\"},{\"id\":265294757,\"identity\":\"1afada55-0523-4a4d-acae-34373da49268\",\"order_by\":11,\"name\":\"Xiao-Tang Yang\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Shanxi Cancer Hospital, Shanxi Medical University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Xiao-Tang\",\"middleName\":\"\",\"lastName\":\"Yang\",\"suffix\":\"\"},{\"id\":265294758,\"identity\":\"bab71317-c95f-47b9-8c70-bed8bc7a9fb6\",\"order_by\":12,\"name\":\"Meng Li\",\"email\":\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAoElEQVRIiWNgGAWjYBACPghlw8PP30CkFjYIlSYjOeMAaVoO2xg0JBCrRezww083/pznMWA4wPjhYw4xWqTTjKVz227zmDM3MEvO3EaUlhwG6dyG2zyWDQfYmHmJ1ML8O+fPOR6DAwnEawGhAyRpSTOzzm1L5pGccbCZOL/wSyc/vp3zx86en7/54IePxGhBAowNpKkfBaNgFIyCUYAbAAAkoi+Jmj2b6AAAAABJRU5ErkJggg==\",\"orcid\":\"\",\"institution\":\"National Cancer Center, Chinese Academy of Medical Sciences and Peking Union Medical College\",\"correspondingAuthor\":true,\"prefix\":\"\",\"firstName\":\"Meng\",\"middleName\":\"\",\"lastName\":\"Li\",\"suffix\":\"\"},{\"id\":265294759,\"identity\":\"ce988426-a9bd-4568-bf93-72525f5b6088\",\"order_by\":13,\"name\":\"Li Zhang\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"National Cancer Center, Chinese Academy of Medical Sciences and Peking Union Medical College\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Li\",\"middleName\":\"\",\"lastName\":\"Zhang\",\"suffix\":\"\"}],\"badges\":[],\"createdAt\":\"2024-01-03 14:44:13\",\"currentVersionCode\":1,\"declarations\":\"\",\"doi\":\"10.21203/rs.3.rs-3832084/v1\",\"doiUrl\":\"https://doi.org/10.21203/rs.3.rs-3832084/v1\",\"draftVersion\":[],\"editorialEvents\":[],\"editorialNote\":\"\",\"failedWorkflow\":false,\"files\":[{\"id\":49323292,\"identity\":\"bd2d7f88-1b6c-4cb1-96a6-9fd0b97fb834\",\"added_by\":\"auto\",\"created_at\":\"2024-01-08 17:10:12\",\"extension\":\"png\",\"order_by\":1,\"title\":\"Figure 1\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":672873,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003eFlow diagrams showing the pathways associated with patient inclusion and exclusion. SCLC = small-cell lung cancer. DICOM=Digital Imaging and Communications in Medicine.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"Fig.1.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-3832084/v1/032d614bfe19c8648eef7e3e.png\"},{\"id\":49323294,\"identity\":\"7599d493-452d-49cf-8db3-3890ef205dad\",\"added_by\":\"auto\",\"created_at\":\"2024-01-08 17:10:12\",\"extension\":\"png\",\"order_by\":2,\"title\":\"Figure 2\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":592054,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003eWorkflow of radiomic analysis.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"Fig.2.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-3832084/v1/4a41421e7b294ea4fb2bca78.png\"},{\"id\":49324427,\"identity\":\"c2a81824-d803-43fb-b9c3-cc7d3291dea2\",\"added_by\":\"auto\",\"created_at\":\"2024-01-08 17:18:12\",\"extension\":\"png\",\"order_by\":3,\"title\":\"Figure 3\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":87202,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003eThe resulting features in the GTV model (a), PTV model (b), GTV+PTV model (c), clinical model (d), and combined model (e). The y-axis indicates the selected features, and the x-axis represents the coefficient of features.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"Fig.3.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-3832084/v1/bf8ca688a114109afb379414.png\"},{\"id\":49323297,\"identity\":\"b38ad780-80b5-45ef-bdf1-ac5aacbec923\",\"added_by\":\"auto\",\"created_at\":\"2024-01-08 17:10:13\",\"extension\":\"png\",\"order_by\":4,\"title\":\"Figure 4\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":803076,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003eDemonstration of the radiomic nomogram and ROC curves.\\u003c/p\\u003e\\n\\u003cp\\u003e(a) A radiomic nomogram incorporating clinical parameters, GTV, and PTV features was constructed.\\u003c/p\\u003e\\n\\u003cp\\u003e(b, c) ROC curves showing the performance of the GTV model, PTV model, GTV+PTV model, clinical model, and combined model for the prediction of OLM in the training (b) and external validation (c) cohorts.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"Fig.4.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-3832084/v1/140637ebad0cfebff76bbb0e.png\"},{\"id\":49323295,\"identity\":\"68c6b190-9c74-4cc2-94c4-21f4b92cc22a\",\"added_by\":\"auto\",\"created_at\":\"2024-01-08 17:10:12\",\"extension\":\"png\",\"order_by\":5,\"title\":\"Figure 5\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":850845,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003eDecision curve analysis of the training cohort (a) and external validation cohort (b). The calibrations of the GTV+PTV model (c) and combined model (d).\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"Fig.5.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-3832084/v1/ee6a9d2f42851c3907cd9425.png\"},{\"id\":49325587,\"identity\":\"f4d4cbc6-488b-48b7-9198-655f3be536de\",\"added_by\":\"auto\",\"created_at\":\"2024-01-08 17:26:13\",\"extension\":\"pdf\",\"order_by\":0,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"manuscript-pdf\",\"size\":1498673,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"manuscript.pdf\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-3832084/v1/5d246b40-c159-4096-b6cf-8f3ba97afa48.pdf\"},{\"id\":49323296,\"identity\":\"0e166f56-2f01-4c70-8f14-14f185f00f87\",\"added_by\":\"auto\",\"created_at\":\"2024-01-08 17:10:12\",\"extension\":\"docx\",\"order_by\":9,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"supplement\",\"size\":28807,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"SupplementaryMaterials.docx\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-3832084/v1/97aa614c1bd09c890ade1be6.docx\"}],\"financialInterests\":\"No competing interests reported.\",\"formattedTitle\":\"Incidence Rate of Occult Lymph Node Metastasis in Clinical T 1-2 N 0 M 0 Small Cell Lung Cancer Patients and Radiomic Prediction Based on Contrast-enhanced CT Imaging: A Multicentre Study\",\"fulltext\":[{\"header\":\"1. Introduction\",\"content\":\"\\u003cp\\u003eLung cancer is the most common cause of cancer-related death worldwide and accounts for approximately 18.0% of all such deaths (\\u003cspan citationid=\\\"CR1\\\" class=\\\"CitationRef\\\"\\u003e1\\u003c/span\\u003e). Lung cancer can be divided into small cell lung cancer (SCLC) and non-small cell lung cancer (NSCLC) based on histological subtype, and SCLC accounts for approximately 15% of all lung cancer cases (\\u003cspan citationid=\\\"CR2\\\" class=\\\"CitationRef\\\"\\u003e2\\u003c/span\\u003e). SCLC is a high-grade neuroendocrine carcinoma with an exceptionally poor prognosis and an overall 5-year survival rate of only 7% (\\u003cspan citationid=\\\"CR3\\\" class=\\\"CitationRef\\\"\\u003e3\\u003c/span\\u003e). Previous studies have shown that concurrent chemoradiotherapy (CCRT) has been the standard treatment for SCLC since the early 1990s (\\u003cspan citationid=\\\"CR4\\\" class=\\\"CitationRef\\\"\\u003e4\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR5\\\" class=\\\"CitationRef\\\"\\u003e5\\u003c/span\\u003e). With the widespread use of CT, the number of early peripheral SCLC tumours has increased (\\u003cspan additionalcitationids=\\\"CR7\\\" citationid=\\\"CR6\\\" class=\\\"CitationRef\\\"\\u003e6\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR8\\\" class=\\\"CitationRef\\\"\\u003e8\\u003c/span\\u003e). Recent attention has shifted towards surgical intervention, revealing promising 5-year survival rates of up to 50% for pathological T\\u003csub\\u003e1\\u0026thinsp;\\u0026minus;\\u0026thinsp;2\\u003c/sub\\u003eN\\u003csub\\u003e0\\u003c/sub\\u003eM\\u003csub\\u003e0\\u003c/sub\\u003e SCLC patients (\\u003cspan additionalcitationids=\\\"CR10\\\" citationid=\\\"CR9\\\" class=\\\"CitationRef\\\"\\u003e9\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR11\\\" class=\\\"CitationRef\\\"\\u003e11\\u003c/span\\u003e). Hence, the National Comprehensive Cancer Network guidelines recommend surgery as the primary treatment modality for pathological T\\u003csub\\u003e1\\u0026thinsp;\\u0026minus;\\u0026thinsp;2\\u003c/sub\\u003eN\\u003csub\\u003e0\\u003c/sub\\u003eM\\u003csub\\u003e0\\u003c/sub\\u003e SCLC (\\u003cspan citationid=\\\"CR12\\\" class=\\\"CitationRef\\\"\\u003e12\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR13\\\" class=\\\"CitationRef\\\"\\u003e13\\u003c/span\\u003e).\\u003c/p\\u003e \\u003cp\\u003eHowever, in clinical practice, while imaging is effective in determining the T\\u003csub\\u003e1\\u0026thinsp;\\u0026minus;\\u0026thinsp;2\\u003c/sub\\u003e stage, defining N\\u003csub\\u003e0\\u003c/sub\\u003e is challenging because surgical lymph node dissection often yields positive results when imaging does not indicate lymph node metastases (\\u003cspan additionalcitationids=\\\"CR15\\\" citationid=\\\"CR14\\\" class=\\\"CitationRef\\\"\\u003e14\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR16\\\" class=\\\"CitationRef\\\"\\u003e16\\u003c/span\\u003e). Occult lymph node metastasis (OLM) refers to the situation in which lymph node metastasis is not detected by presurgical imaging (mainly CT) but is confirmed by postoperative pathology (\\u003cspan additionalcitationids=\\\"CR18\\\" citationid=\\\"CR17\\\" class=\\\"CitationRef\\\"\\u003e17\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR19\\\" class=\\\"CitationRef\\\"\\u003e19\\u003c/span\\u003e). Preoperative imaging examinations rely mainly on CT to diagnose lymph node metastasis, but many OLMs are missed, resulting in ineffective surgery. For cT\\u003csub\\u003e1\\u0026thinsp;\\u0026minus;\\u0026thinsp;2\\u003c/sub\\u003eN\\u003csub\\u003e0\\u003c/sub\\u003eM\\u003csub\\u003e0\\u003c/sub\\u003e SCLC patients, the presence or absence of OLM determines whether the patient is able to undergo surgery. Thoracoscopic biopsy is the \\\"gold standard\\\" for detecting the status of chest lymph nodes, but this is an invasive examination method that may lead to a series of complications, such as bleeding, infection, and pneumothorax. Therefore, identifying new and valuable noninvasive imaging methods for predicting OLM in cT\\u003csub\\u003e1\\u0026thinsp;\\u0026minus;\\u0026thinsp;2\\u003c/sub\\u003eN\\u003csub\\u003e0\\u003c/sub\\u003eM\\u003csub\\u003e0\\u003c/sub\\u003e SCLC patients is necessary.\\u003c/p\\u003e \\u003cp\\u003eIn recent years, radiomics has emerged as a prominent area of research, allowing for the high-throughput extraction of extensive data from medical images (\\u003cspan citationid=\\\"CR20\\\" class=\\\"CitationRef\\\"\\u003e20\\u003c/span\\u003e). This approach enables the analysis of high-level and quantitative image features, providing a profound reflection of the spatial heterogeneity within tumour tissues (\\u003cspan citationid=\\\"CR21\\\" class=\\\"CitationRef\\\"\\u003e21\\u003c/span\\u003e). Previous studies have successfully developed models for predicting OLM in NSCLC patients based on radiomic features of primary lesions, demonstrating robust predictive performance (\\u003cspan additionalcitationids=\\\"CR23\\\" citationid=\\\"CR22\\\" class=\\\"CitationRef\\\"\\u003e22\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR24\\\" class=\\\"CitationRef\\\"\\u003e24\\u003c/span\\u003e). Additionally, peritumoral radiomics has proven equally predictive (\\u003cspan citationid=\\\"CR25\\\" class=\\\"CitationRef\\\"\\u003e25\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR26\\\" class=\\\"CitationRef\\\"\\u003e26\\u003c/span\\u003e). Furthermore, the literature has focused predominantly on OLM in NSCLC (\\u003cspan additionalcitationids=\\\"CR27 CR28\\\" citationid=\\\"CR26\\\" class=\\\"CitationRef\\\"\\u003e26\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR29\\\" class=\\\"CitationRef\\\"\\u003e29\\u003c/span\\u003e), with limited studies exploring the incidence rate of OLM in SCLC.\\u003c/p\\u003e \\u003cp\\u003eConsequently, this study focused on cT\\u003csub\\u003e1\\u0026thinsp;\\u0026minus;\\u0026thinsp;2\\u003c/sub\\u003eN\\u003csub\\u003e0\\u003c/sub\\u003eM\\u003csub\\u003e0\\u003c/sub\\u003e SCLC patients to investigate the incidence rate of OLM in this clinical population and develop predictive models for OLM that integrate clinical parameters and intratumoral and peritumoral contrast-enhanced CT radiomics.\\u003c/p\\u003e\"},{\"header\":\"2. Materials and Methods\",\"content\":\"\\u003cdiv id=\\\"Sec3\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003e2.1. Patient Selection\\u003c/h2\\u003e \\u003cp\\u003e The institutional review boards approved this retrospective study, and the requirement for written informed consent was waived. The histopathology of the tumours was defined according to the 2015 World Health Organization definition (\\u003cspan citationid=\\\"CR30\\\" class=\\\"CitationRef\\\"\\u003e30\\u003c/span\\u003e), and clinical and pathological staging was based on the 8th edition of the TNM classification (\\u003cspan citationid=\\\"CR31\\\" class=\\\"CitationRef\\\"\\u003e31\\u003c/span\\u003e). This study retrospectively reviewed 242 patients with SCLC confirmed by postoperative pathology from four centres between January 2014 and September 2022. The inclusion criteria were as follows: (\\u003cspan citationid=\\\"CR1\\\" class=\\\"CitationRef\\\"\\u003e1\\u003c/span\\u003e) underwent resection of the primary lesion and systematic lymph node dissection; (\\u003cspan citationid=\\\"CR2\\\" class=\\\"CitationRef\\\"\\u003e2\\u003c/span\\u003e) underwent preoperative enhanced CT; and (\\u003cspan citationid=\\\"CR3\\\" class=\\\"CitationRef\\\"\\u003e3\\u003c/span\\u003e) had a clinical stage before surgery of T\\u003csub\\u003e1\\u0026thinsp;\\u0026minus;\\u0026thinsp;2\\u003c/sub\\u003eN\\u003csub\\u003e0\\u003c/sub\\u003eM\\u003csub\\u003e0\\u003c/sub\\u003e. Additionally, all patients had solitary pulmonary nodules in clinical stage T\\u003csub\\u003e1\\u0026thinsp;~\\u0026thinsp;2\\u003c/sub\\u003e based on enhanced CT imaging and no enlarged lymph nodes (i.e., short diameter of LN\\u0026thinsp;\\u0026le;\\u0026thinsp;10.0 mm on CT imaging). The exclusion criteria were as follows: (\\u003cspan citationid=\\\"CR1\\\" class=\\\"CitationRef\\\"\\u003e1\\u003c/span\\u003e) patients who received radiotherapy, chemotherapy, or other treatments for SCLC before surgery; (\\u003cspan citationid=\\\"CR2\\\" class=\\\"CitationRef\\\"\\u003e2\\u003c/span\\u003e) had an interval between CT examination and surgery of more than 2 weeks; (\\u003cspan citationid=\\\"CR3\\\" class=\\\"CitationRef\\\"\\u003e3\\u003c/span\\u003e) had thin-layer images (with a slice thickness less than or equal to 1.25 mm) missing; and (\\u003cspan citationid=\\\"CR4\\\" class=\\\"CitationRef\\\"\\u003e4\\u003c/span\\u003e) had severe CT artefacts and poor image quality. For patients with multiple lesions, only SCLC lesions with conclusive pathological results were included. The patient's lymph node metastasis was obtained from the postoperative pathology report and reconfirmed by a senior pathology professor in the Department of Pathology.\\u003c/p\\u003e \\u003cp\\u003eAll patients from centre 1 were allocated to the training cohort, and patients from centres 2, 3 and 4 composed the external validation cohort (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig1\\\" class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003e). 1).\\u003c/p\\u003e \\u003cp\\u003e \\u003c/p\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec4\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003e2.2. CT Scanning and Semantic Features\\u003c/h2\\u003e \\u003cp\\u003eAll enrolled patients in the four hospitals underwent a similar scan setup but with different systems and parameters (Appendix E1). The definitions and evaluation criteria for clinical parameters are described in Appendix E2. Two radiologists, each with 2 years of experience in lung imaging and blinded to the clinical and pathologic results, evaluated semantic features in the lung window setting (level, -550 HU; width, 1500 HU) and the mediastinal window setting (level, 40 HU; width, 400 HU). Any disagreements regarding the description of semantic features were resolved through consensus reading, and the results were subsequently confirmed by a chief radiologist specializing in chest imaging.\\u003c/p\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec5\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003e2.3. CT image acquisition and lesion segmentation\\u003c/h2\\u003e \\u003cp\\u003eEnhanced DICOM CT images were anonymized, and regions of interest (ROIs) were delineated by ITK-SANP software (version 3.8.0; \\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003ehttps://www.itksnap.org\\u003c/span\\u003e\\u003cspan address=\\\"https://www.itksnap.org\\\" targettype=\\\"URL\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e). According to previous studies (\\u003cspan citationid=\\\"CR25\\\" class=\\\"CitationRef\\\"\\u003e25\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR32\\\" class=\\\"CitationRef\\\"\\u003e32\\u003c/span\\u003e), the gross tumour volume (GTV) was dilated 15 mm in three dimensions and uniformly served as the GTV\\u0026thinsp;+\\u0026thinsp;PTV (peritumoral volume). The boundaries of the lung nodules were checked by a radiologist and manually adjusted if necessary. Notably, the parts that cross the interlobar pleura, chest wall and mediastinum should be removed (\\u003cspan citationid=\\\"CR33\\\" class=\\\"CitationRef\\\"\\u003e33\\u003c/span\\u003e). We obtained the PTV by subtracting the two values. To ensure that the PTV did not contain any GTV components, we specifically added 1 mm to the region of the GTV (PTV\\u0026thinsp;=\\u0026thinsp;GPTV-(GTV\\u0026thinsp;+\\u0026thinsp;1 mm)).\\u003c/p\\u003e \\u003cp\\u003eTo assess the robustness of the intratumoral and peritumoral segmentation methods, 30 patients were randomly selected, and two junior radiologists performed segmentation on their ROIs twice, with a 2-month interval between sessions, to obtain intraclass correlation coefficients (ICCs) (Appendix E3).\\u003c/p\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec6\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003e2.4. Radiomic features\\u003c/h2\\u003e \\u003cp\\u003eThe images were resampled using linear interpolation to achieve a uniform voxel size of 1\\u0026times;1\\u0026times;1 in all three anatomical directions (\\u003cspan citationid=\\\"CR34\\\" class=\\\"CitationRef\\\"\\u003e34\\u003c/span\\u003e), and the image grayscale was discretized to 25 grayscales. We utilized \\u003cem\\u003ePyRadiomics\\u003c/em\\u003e to extract features from segmented GPVs and PTVs (\\u003cspan citationid=\\\"CR35\\\" class=\\\"CitationRef\\\"\\u003e35\\u003c/span\\u003e). For each region, 14 shape features (3D), 18 first-order features, 24 grey level cooccurrence matrix (GLCM) features, 16 grey level run length matrix (GLRLM) features, 16 grey level size zone matrix (GLSZM) features, 14 grey level dependence matrix (GLDM) features, and 5 neighbouring grey-tone difference matrix (NGTDM) features were obtained. For each GTV region and PTV, 1595 radiomic features were extracted from the original images. A detailed list of the extracted features and the parameters used in CT image preprocessing and feature extraction is provided in Appendix E4.\\u003c/p\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec7\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003e2.5. Feature Selection and Modelling\\u003c/h2\\u003e \\u003cp\\u003eBefore radiomic feature selection, only reproducible radiomic features with an ICC\\u0026thinsp;\\u0026ge;\\u0026thinsp;0.8 were included in the analysis (\\u003cspan citationid=\\\"CR36\\\" class=\\\"CitationRef\\\"\\u003e36\\u003c/span\\u003e). Univariate analysis was subsequently performed, and features with a significance level of P\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.01 were retained in the model. Additionally, features with a correlation coefficient exceeding 0.9 were removed from the model. The least absolute shrinkage and selection operator (LASSO) method, which compresses some independent variables with little or no influence on 0, was used to select the most robust and nonredundant radiomic features from the extracted features (\\u003cspan citationid=\\\"CR37\\\" class=\\\"CitationRef\\\"\\u003e37\\u003c/span\\u003e).\\u003c/p\\u003e \\u003cp\\u003eClinical parameters were evaluated in combination with selected radiomic features in the multivariable logistic regression model for predicting the presence of OLM (Appendix E5). The radiomics model's output scores (Rad_score) were merged with the clinical features to construct the nomogram. This comprehensive model effectively integrated both radiomic and clinical parameters, enhancing the overall predictive power and potential clinical utility of the model.\\u003c/p\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec8\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003e2.6. Statistical analysis\\u003c/h2\\u003e \\u003cp\\u003eContinuous variables were compared using two-sample t tests, whereas categorical variables were assessed through chi-square tests and Fisher\\u0026rsquo;s exact tests. The GTV, PTV, GTV\\u0026thinsp;+\\u0026thinsp;PTV, clinical, and combined models were established and verified by using R (version 4.1.0, \\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003ehttps://www\\u003c/span\\u003e\\u003cspan address=\\\"https://www\\\" targettype=\\\"URL\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e, rproject.org). To assess the model's performance, the area under the ROC curve (AUC) was utilized, with the optimal cut-off value determined using the derived Youden index. Additionally, the model's accuracy, sensitivity, specificity, negative predictive value, and positive predictive value were computed. Decision curve analysis (DCA) was performed according to the methods of a previous study (\\u003cspan citationid=\\\"CR38\\\" class=\\\"CitationRef\\\"\\u003e38\\u003c/span\\u003e). The Delong test was used to compare different AUC values (\\u003cspan citationid=\\\"CR39\\\" class=\\\"CitationRef\\\"\\u003e39\\u003c/span\\u003e). A two-tailed p value of less than 0.05 was considered to indicate statistical significance.\\u003c/p\\u003e \\u003c/div\\u003e\"},{\"header\":\"3. Results\",\"content\":\"\\u003cdiv id=\\\"Sec10\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003e3.1. Patient characteristics\\u003c/h2\\u003e \\u003cp\\u003eA total of 242 patients (186 men, 56 women) with 242 lesions (OLM-negative, 160; OLM-positive, 82) were included after the application of the exclusion criteria (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig1\\\" class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003e). The rate of OLM positivity in all patients was 33.9% (82/242). The characteristics of all the patients are detailed in Table\\u0026nbsp;\\u003cspan refid=\\\"Tab1\\\" class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003e.\\u003c/p\\u003e \\u003cp\\u003e \\u003cdiv class=\\\"gridtable\\\"\\u003e\\u003ctable float=\\\"Yes\\\" id=\\\"Tab1\\\" border=\\\"1\\\"\\u003e \\u003ccaption language=\\\"En\\\"\\u003e \\u003cdiv class=\\\"CaptionNumber\\\"\\u003eTable 1\\u003c/div\\u003e \\u003cdiv class=\\\"CaptionContent\\\"\\u003e \\u003cp\\u003eThe Parameters in the Development of the Clinical Model\\u003c/p\\u003e \\u003c/div\\u003e \\u003c/caption\\u003e \\u003ccolgroup cols=\\\"7\\\"\\u003e \\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c1\\\" colnum=\\\"1\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c2\\\" colnum=\\\"2\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c3\\\" colnum=\\\"3\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c4\\\" colnum=\\\"4\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c5\\\" colnum=\\\"5\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c6\\\" colnum=\\\"6\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c7\\\" colnum=\\\"7\\\"\\u003e\\u003c/div\\u003e \\u003cthead\\u003e \\u003ctr\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c1\\\"\\u003e\\u0026nbsp;\\u003c/th\\u003e \\u003cth align=\\\"left\\\" colspan=\\\"3\\\" nameend=\\\"c4\\\" namest=\\\"c2\\\"\\u003e \\u003cp\\u003eTraining cohort\\u003c/p\\u003e \\u003c/th\\u003e \\u003cth align=\\\"left\\\" colspan=\\\"3\\\" nameend=\\\"c7\\\" namest=\\\"c5\\\"\\u003e \\u003cp\\u003eExternal validation cohort\\u003c/p\\u003e \\u003c/th\\u003e \\u003c/tr\\u003e \\u003c/thead\\u003e \\u003ctbody\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003eOLM (-)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003eOLM (+)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003ep value\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003eOLM (-)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003eOLM (+)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e \\u003cp\\u003ep value\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003eN\\u0026thinsp;=\\u0026thinsp;98\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003eN\\u0026thinsp;=\\u0026thinsp;60\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003eN\\u0026thinsp;=\\u0026thinsp;62\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003eN\\u0026thinsp;=\\u0026thinsp;22\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eGender\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.052\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e \\u003cp\\u003e0.546\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eFemale\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e18 (18.367%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e20 (33.333%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e12 (19.355%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e6 (27.273%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eMale\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e80 (81.633%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e40 (66.667%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e50 (80.645%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e16 (72.727%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eAge\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e63.000 [57.000;68.000]\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e61.500 [55.000;66.500]\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.204\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e62.500 [56.000;67.750]\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e63.000 [57.000;67.000]\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e \\u003cp\\u003e0.579\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eSmoke\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.002*\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e \\u003cp\\u003e0.012\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eNo\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e24 (24.490%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e30 (50.000%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e14 (22.581%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e12 (54.545%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eYes\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e74 (75.510%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e30 (50.000%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e48 (77.419%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e10 (45.455%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eFamily history\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.655\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e \\u003cp\\u003e0.371\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eNo\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e81 (82.653%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e52 (86.667%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e58 (93.548%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e19 (86.364%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eYes\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e17 (17.347%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e8 (13.333%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e4 (6.452%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e3 (13.636%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eLobe\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.753\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e \\u003cp\\u003e0.472\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eRUL\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e27 (27.551%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e12 (20.000%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e17 (27.419%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e4 (18.182%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eRML\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e6 (6.122%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e3 (5.000%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e1 (1.613%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e1 (4.545%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eRLL\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e19 (19.388%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e16 (26.667%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e13 (20.968%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e4 (18.182%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eLUL\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e27 (27.551%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e16 (26.667%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e15 (24.194%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e9 (40.909%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eLLL\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e19 (19.388%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e13 (21.667%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e16 (25.806%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e4 (18.182%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eLocation\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.023*\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e \\u003cp\\u003e0.071\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eCenter\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e20 (20.408%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e23 (38.333%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e6 (9.677%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e6 (27.273%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003ePeripheral\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e78 (79.592%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e37 (61.667%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e56 (90.323%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e16 (72.727%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eClinical stage T\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.051\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e \\u003cp\\u003e0.085\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003e1\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e61 (62.245%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e27 (45.000%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e33 (53.226%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e17 (77.273%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003e2\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e37 (37.755%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e33 (55.000%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e29 (46.774%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e5 (22.727%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eShape\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.033*\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e \\u003cp\\u003e0.677\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eIrregular\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e67 (68.367%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e30 (50.000%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e47 (75.806%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e15 (68.182%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eRound or oval\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e31 (31.633%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e30 (50.000%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e15 (24.194%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e7 (31.818%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eBranching\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.423\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e \\u003cp\\u003e0.280\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eNo\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e83 (84.694%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e47 (78.333%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e60 (96.774%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e20 (90.909%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eYes\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e15 (15.306%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e13 (21.667%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e2 (3.226%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e2 (9.091%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eLobulation\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.581\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e \\u003cp\\u003e0.506\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eNo\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e9 (9.184%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e8 (13.333%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e9 (14.516%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e5 (22.727%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eYes\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e89 (90.816%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e52 (86.667%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e53 (85.484%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e17 (77.273%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eSpiculation sign\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.501\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e \\u003cp\\u003e1.000\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eNo\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e69 (70.408%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e46 (76.667%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e45 (72.581%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e16 (72.727%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eYes\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e29 (29.592%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e14 (23.333%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e17 (27.419%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e6 (27.273%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eCalcification\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.635\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e \\u003cp\\u003e1.000\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eNo\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e96 (97.959%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e58 (96.667%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e60 (96.774%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e22 (100.000%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eYes\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e2 (2.041%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e2 (3.333%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e2 (3.226%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e0 (0.000%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eConcavity\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e1.000\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e \\u003cp\\u003e.\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eNo\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e97 (98.980%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e59 (98.333%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e62 (100.000%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e22 (100.000%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eYes\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e1 (1.020%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e1 (1.667%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0 (0.000%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e0 (0.000%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eCarcinoma\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.527\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e \\u003cp\\u003e1.000\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eNo\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e41 (41.837%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e29 (48.333%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e27 (43.548%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e10 (45.455%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eYes\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e57 (58.163%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e31 (51.667%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e35 (56.452%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e12 (54.545%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eBronchial\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.053\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e \\u003cp\\u003e0.345\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eNo\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e64 (65.306%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e29 (48.333%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e40 (64.516%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e11 (50.000%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eYes\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e34 (34.694%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e31 (51.667%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e22 (35.484%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e11 (50.000%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eAir-bronchogram\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.302\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e \\u003cp\\u003e.\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eNo\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e94 (95.918%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e55 (91.667%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e62 (100.000%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e22 (100.000%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eYes\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e4 (4.082%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e5 (8.333%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0 (0.000%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e0 (0.000%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eObstructive:\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.022*\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e \\u003cp\\u003e0.770\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eNo\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e74 (75.510%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e34 (56.667%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e48 (77.419%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e18 (81.818%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eYes\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e24 (24.490%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e26 (43.333%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e14 (22.581%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e4 (18.182%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eEnhancement Heterogeneity\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.015*\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e \\u003cp\\u003e0.112\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003ehomogeneous\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e27 (27.551%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e6 (10.000%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e9 (14.516%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e7 (31.818%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003enot homogeneous\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e71 (72.449%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e54 (90.000%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e53 (85.484%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e15 (68.182%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eBVB\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.951\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e \\u003cp\\u003e0.761\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eNo\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e57 (58.163%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e36 (60.000%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e41 (66.129%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e16 (72.727%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eYes\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e41 (41.837%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e24 (40.000%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e21 (33.871%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e6 (27.273%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003ePleural Retraction\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.800\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e \\u003cp\\u003e0.720\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eNo\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e86 (87.755%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e51 (85.000%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e53 (85.484%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e20 (90.909%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eYes\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e12 (12.245%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e9 (15.000%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e9 (14.516%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e2 (9.091%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003ePleural Attachment\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.069\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e \\u003cp\\u003e1.000\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eNo\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e75 (76.531%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e37 (61.667%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e46 (74.194%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e17 (77.273%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eYes\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e23 (23.469%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e23 (38.333%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e16 (25.806%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e5 (22.727%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003ePeripheral Emphysema\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.124\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e \\u003cp\\u003e0.469\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eNo\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e57 (58.163%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e43 (71.667%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e32 (51.613%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e14 (63.636%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eYes\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e41 (41.837%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e17 (28.333%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e30 (48.387%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e8 (36.364%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eInterstitial Pneumonia\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.749\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e \\u003cp\\u003e0.053\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eNo\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e92 (93.878%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e55 (91.667%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e61 (98.387%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e19 (86.364%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eYes\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e6 (6.122%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e5 (8.333%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e1 (1.613%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e3 (13.636%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003c/tr\\u003e \\u003c/tbody\\u003e \\u003c/colgroup\\u003e \\u003ctfoot\\u003e \\u003ctr\\u003e\\u003ctd colspan=\\\"7\\\"\\u003e*Significant difference (p\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.05). RUL, right upper lung; RML, right middle lung; RLL, right lower lung; LUL, left upper lung; LLL, left lower lung; BVB, bronchovascular bundle thickening\\u003c/td\\u003e\\u003c/tr\\u003e \\u003c/tfoot\\u003e \\u003c/table\\u003e\\u003c/div\\u003e \\u003c/p\\u003e \\u003cp\\u003eClinical parameters that were significant at p\\u0026thinsp;\\u0026le;\\u0026thinsp;0.05 in univariate analysis were subsequently entered into multivariate analysis. Furthermore, smoking status and shape (P\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.05) were also included in the multivariate analysis.\\u003c/p\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec11\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003e3.2. Feature Selection and Model Construction\\u003c/h2\\u003e \\u003cp\\u003eFigure \\u003cspan refid=\\\"Fig2\\\" class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003e shows the workflow of the radiomic feature analysis. The radiomic features were selected by using the ICC, univariate analysis, multivariate analysis, correlation analysis, LASSO regression, and multivariable logistic regression. Finally, two radiomic features were selected and utilized to construct the GTV model, and three radiomic features were used to construct the PTV model. After the five selected features were integrated, correlation analysis and multivariate stepwise regression were performed, resulting in the final selection of three features\\u0026mdash;i.e., the MCC from the GTV, median and IDN from the PTV\\u0026mdash;for use in constructing the GTV\\u0026thinsp;+\\u0026thinsp;PTV model. The combined model was established by incorporating one GTV radiomic feature (MCC), two PTV radiomic features (median and IDN), and two clinical parameters. The parameters of the five models are detailed in Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig3\\\" class=\\\"InternalRef\\\"\\u003e3\\u003c/span\\u003e and Appendix E6.\\u003c/p\\u003e \\u003cp\\u003e \\u003c/p\\u003e \\u003cp\\u003e \\u003c/p\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec12\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003e3.3. Performance and Comparison of the Three Models for All Patients\\u003c/h2\\u003e \\u003cp\\u003eAll five models have some predictive power. The AUC values of the combined model were 0.774 and 0.772 in the training and external testing cohorts, respectively, performing better than any other models in our study.\\u003c/p\\u003e \\u003cp\\u003eAll the results regarding predictive performance are enumerated in Table\\u0026nbsp;\\u003cspan refid=\\\"Tab2\\\" class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003e, and the ROC curves are shown in Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig4\\\" class=\\\"InternalRef\\\"\\u003e4\\u003c/span\\u003e. The correlation analysis of clinical and radiomic features is indicated in Appendix E5. With respect to the training cohort, the DeLong test revealed significant differences in the area under the curve (AUC) (p\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.05) between the GTV model and the combined model, between the PTV model and combined model, between the GTV\\u0026thinsp;+\\u0026thinsp;PTV model and combined model, and between the clinical model and combined model. With respect to the external testing cohort, the Delong test revealed that there were significant differences in the area under the curve (AUC) between the GTV\\u0026thinsp;+\\u0026thinsp;PTV model and the combined model (p\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.05) (Appendix E7). The DCAs (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig5\\\" class=\\\"InternalRef\\\"\\u003e5\\u003c/span\\u003e) revealed that when the probability of the threshold was between approximately 10\\u0026thinsp;~\\u0026thinsp;80%, the net benefits of the combined model and the GTV\\u0026thinsp;+\\u0026thinsp;PTV model for the prediction of OLM were greater than those of any other type of model. The calibration plot revealed good predictive accuracy between the actual probability and the predicted probability of the GTV\\u0026thinsp;+\\u0026thinsp;PTV model and the combined model (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig5\\\" class=\\\"InternalRef\\\"\\u003e5\\u003c/span\\u003e).\\u003c/p\\u003e \\u003cp\\u003e \\u003cdiv class=\\\"gridtable\\\"\\u003e\\u003ctable float=\\\"Yes\\\" id=\\\"Tab2\\\" border=\\\"1\\\"\\u003e \\u003ccaption language=\\\"En\\\"\\u003e \\u003cdiv class=\\\"CaptionNumber\\\"\\u003eTable 2\\u003c/div\\u003e \\u003cdiv class=\\\"CaptionContent\\\"\\u003e \\u003cp\\u003ePerformance of the Five Models\\u003c/p\\u003e \\u003c/div\\u003e \\u003c/caption\\u003e \\u003ccolgroup cols=\\\"7\\\"\\u003e \\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c1\\\" colnum=\\\"1\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"char\\\" char=\\\".\\\" class=\\\"colspec\\\" colname=\\\"c2\\\" colnum=\\\"2\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"char\\\" char=\\\".\\\" class=\\\"colspec\\\" colname=\\\"c3\\\" colnum=\\\"3\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"char\\\" char=\\\".\\\" class=\\\"colspec\\\" colname=\\\"c4\\\" colnum=\\\"4\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"char\\\" char=\\\".\\\" class=\\\"colspec\\\" colname=\\\"c5\\\" colnum=\\\"5\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"char\\\" char=\\\".\\\" class=\\\"colspec\\\" colname=\\\"c6\\\" colnum=\\\"6\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c7\\\" colnum=\\\"7\\\"\\u003e\\u003c/div\\u003e \\u003cthead\\u003e \\u003ctr\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c1\\\"\\u003e\\u0026nbsp;\\u003c/th\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003eAUC (95% CI)\\u003c/p\\u003e \\u003c/th\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003eACC\\u003c/p\\u003e \\u003c/th\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003eSEN\\u003c/p\\u003e \\u003c/th\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003eSPE\\u003c/p\\u003e \\u003c/th\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003ePPV\\u003c/p\\u003e \\u003c/th\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c7\\\"\\u003e \\u003cp\\u003eNPV\\u003c/p\\u003e \\u003c/th\\u003e \\u003c/tr\\u003e \\u003c/thead\\u003e \\u003ctbody\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colspan=\\\"3\\\" nameend=\\\"c3\\\" namest=\\\"c1\\\"\\u003e \\u003cp\\u003eTraining cohort (n\\u0026thinsp;=\\u0026thinsp;158)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eGTV\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e0.657(0.571\\u0026ndash;0.744)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e0.633\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.685\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.606\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e0.474\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e \\u003cp\\u003e0.788\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003ePTV\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e0.687(0.599\\u0026ndash;0.776)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e0.646\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.667\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.635\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e0.486\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e \\u003cp\\u003e0.786\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eGTV\\u0026thinsp;+\\u0026thinsp;PTV\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e0.697(0.610\\u0026ndash;0.785)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e0.665\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.722\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.635\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e0.506\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e \\u003cp\\u003e0.815\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eClinical\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e0.689(0.605\\u0026ndash;0.773)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e0.627\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.741\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.567\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e0.471\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e \\u003cp\\u003e0.808\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eCombined\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e0.774(0.696\\u0026ndash;0.853)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e0.759\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.63\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.827\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e0.654\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e \\u003cp\\u003e0.811\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colspan=\\\"2\\\" nameend=\\\"c2\\\" namest=\\\"c1\\\"\\u003e \\u003cp\\u003eExternal test cohort (n\\u0026thinsp;=\\u0026thinsp;84)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eGTV\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e0.663(0.535\\u0026ndash;0.792)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e0.607\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.607\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.607\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e0.436\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e \\u003cp\\u003e0.756\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003ePTV\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e0.673(0.541\\u0026ndash;0.804)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e0.607\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.536\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.643\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e0.429\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e \\u003cp\\u003e0.735\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eGTV\\u0026thinsp;+\\u0026thinsp;PTV\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e0.703(0.575\\u0026ndash;0.831)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e0.667\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.607\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.696\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e0.5\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e \\u003cp\\u003e0.78\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eClinical\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e0.675(0.557\\u0026ndash;0.792)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e0.595\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.786\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.5\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e0.44\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e \\u003cp\\u003e0.824\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eCombined\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e0.772(0.656\\u0026ndash;0.887)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e0.762\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.679\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.804\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e0.633\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e \\u003cp\\u003e0.833\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003c/tbody\\u003e \\u003c/colgroup\\u003e \\u003c/table\\u003e\\u003c/div\\u003e \\u003c/p\\u003e \\u003cp\\u003e \\u003c/p\\u003e \\u003cp\\u003e \\u003c/p\\u003e \\u003c/div\\u003e\"},{\"header\":\"4. Discussion\",\"content\":\"\\u003cp\\u003eIn this multicentre study, for the first time, we revealed a 33.9% positivity rate for OLM among patients with cT\\u003csub\\u003e1\\u0026thinsp;\\u0026minus;\\u0026thinsp;2\\u003c/sub\\u003eN\\u003csub\\u003e0\\u003c/sub\\u003eM\\u003csub\\u003e0\\u003c/sub\\u003e SCLC. This observation suggested that OLM in cT\\u003csub\\u003e1\\u0026thinsp;\\u0026minus;\\u0026thinsp;2\\u003c/sub\\u003eN\\u003csub\\u003e0\\u003c/sub\\u003eM\\u003csub\\u003e0\\u003c/sub\\u003e SCLC surpasses the prevalence observed in NSCLC, where it ranges from 16\\u0026ndash;29% (\\u003cspan additionalcitationids=\\\"CR27 CR28\\\" citationid=\\\"CR26\\\" class=\\\"CitationRef\\\"\\u003e26\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR29\\\" class=\\\"CitationRef\\\"\\u003e29\\u003c/span\\u003e, \\u003cspan additionalcitationids=\\\"CR41 CR42 CR43 CR44\\\" citationid=\\\"CR40\\\" class=\\\"CitationRef\\\"\\u003e40\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR45\\\" class=\\\"CitationRef\\\"\\u003e45\\u003c/span\\u003e). In addition, we addressed a crucial challenge in managing cT\\u003csub\\u003e1\\u0026thinsp;~\\u0026thinsp;2\\u003c/sub\\u003eN\\u003csub\\u003e0\\u003c/sub\\u003eM\\u003csub\\u003e0\\u003c/sub\\u003e SCLC by developing and validating predictive models for OLM. Our combined model consistently outperformed the other models in our study, as evidenced by the higher area under the curve (AUC) values in both the training cohort (0.774) and the validation cohort (0.772). According to the model, patients identified as having a higher risk of OLM in cT\\u003csub\\u003e1\\u0026thinsp;\\u0026minus;\\u0026thinsp;2\\u003c/sub\\u003eN\\u003csub\\u003e0\\u003c/sub\\u003eM\\u003csub\\u003e0\\u003c/sub\\u003e SCLC could avoid unnecessary surgeries. Conversely, individuals assessed as having a lower risk might be more confidently considered for surgical resection, with the potential for significant improvements in survival. This study offers a promising approach for accurately identifying OLM in cT\\u003csub\\u003e1\\u0026thinsp;\\u0026minus;\\u0026thinsp;2\\u003c/sub\\u003eN\\u003csub\\u003e0\\u003c/sub\\u003eM\\u003csub\\u003e0\\u003c/sub\\u003e SCLC patients, guiding personalized treatment decisions.\\u003c/p\\u003e \\u003cp\\u003eIn terms of clinical parameters and conventional CT features, smoking status and tumour shape exhibited noteworthy differences in predicting OLM status, while the remaining features showed no significant distinctions. Our study revealed a tendency for patients with OLM to be smokers, a well-established association with the occurrence and progression of SCLC (\\u003cspan citationid=\\\"CR12\\\" class=\\\"CitationRef\\\"\\u003e12\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR46\\\" class=\\\"CitationRef\\\"\\u003e46\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR47\\\" class=\\\"CitationRef\\\"\\u003e47\\u003c/span\\u003e). In contrast, nonsmokers were more inclined to have OLM in the context of NSCLC (\\u003cspan citationid=\\\"CR40\\\" class=\\\"CitationRef\\\"\\u003e40\\u003c/span\\u003e). Additionally, we reported for the first time that round and oval tumour shapes hold notable significance, suggesting that lesions with regular shapes may be at a greater risk of OLM positivity. Previous studies on risk factors for NSCLC have been abundant, but uniform results have been lacking, implicating factors such as female sex, adenocarcinoma, and a small tumour size (\\u003cspan citationid=\\\"CR40\\\" class=\\\"CitationRef\\\"\\u003e40\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR44\\\" class=\\\"CitationRef\\\"\\u003e44\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR48\\\" class=\\\"CitationRef\\\"\\u003e48\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR49\\\" class=\\\"CitationRef\\\"\\u003e49\\u003c/span\\u003e). Thus, there are significant differences in the risk factors for OLM between these two distinct pathological types of lung cancer, enhancing our understanding of OLM in lung cancer patients. However, further studies with larger patient cohorts are necessary to validate our findings. Following our in-depth analysis, three radiomic features were ultimately selected: one was the MCC according to intratumoral imaging, and the other two were the median and IDN according to peritumoral imaging. The MCC and IDN are obtained from the grey level co-occurrence matrix (GLCM), which is a texture analysis method that describes spatial relationships between neighbouring pixels to reflect the internal texture of tumours, such as the complexity and heterogeneity of the tumour regions and peritumor regions (\\u003cspan citationid=\\\"CR50\\\" class=\\\"CitationRef\\\"\\u003e50\\u003c/span\\u003e). The first-order median is expressed as the median grey-level intensity of all pixels in the ROI, which can reflect the textural characteristics of regions around the lesions. Based on our results, even if the imaging findings may be similar in both groups, the MCC, median, and IDN in OLM may serve as noninvasive predictive biomarkers and provide additional information from both intra- and peritumor radiomic data.\\u003c/p\\u003e \\u003cp\\u003eSurgical intervention has emerged as a highly impactful therapeutic modality for cT\\u003csub\\u003e1\\u0026thinsp;\\u0026minus;\\u0026thinsp;2\\u003c/sub\\u003eN\\u003csub\\u003e0\\u003c/sub\\u003eM\\u003csub\\u003e0\\u003c/sub\\u003e SCLC, emphasizing the crucial role of promoting this approach in clinical practice (\\u003cspan citationid=\\\"CR51\\\" class=\\\"CitationRef\\\"\\u003e51\\u003c/span\\u003e). A critical consideration is determining whether patients lacking observable lymph node enlargement on routine imaging harbour OLMs. In clinical practice, CT serves as the primary method for preoperative lymph node staging in patients with lung cancer, commonly using a short-axis diameter greater than 1 cm as a threshold(\\u003cspan citationid=\\\"CR40\\\" class=\\\"CitationRef\\\"\\u003e40\\u003c/span\\u003e). However, OLM cannot be assessed. PET-CT supplements this assessment but has inherent limitations, including false positives and negatives (\\u003cspan citationid=\\\"CR52\\\" class=\\\"CitationRef\\\"\\u003e52\\u003c/span\\u003e), and its high cost makes widespread clinical application challenging (\\u003cspan citationid=\\\"CR40\\\" class=\\\"CitationRef\\\"\\u003e40\\u003c/span\\u003e). This study presented a comprehensive noninvasive model that demonstrated good performance across all dimensions, boasting a specificity of 82.7% and a sensitivity of 63% on the training dataset. The model exhibited robustness during external validation. A comparison of the five models using DCA curves revealed that our integrated model outperformed the others within the 10%-80% probability threshold range. Within this clinically relevant range, both the specificity and sensitivity were considered acceptable. The effectiveness of all the models underscores the inadequacy of traditional methods for evaluating our study's specific objectives, establishing a significant advantage in efficacy for our research in the field. These findings hold substantial clinical relevance for cT\\u003csub\\u003e1\\u0026thinsp;\\u0026minus;\\u0026thinsp;2\\u003c/sub\\u003eN\\u003csub\\u003e0\\u003c/sub\\u003eM\\u003csub\\u003e0\\u003c/sub\\u003e SCLC patients identified with preoperative negative OLMs, emphasizing the potential impact of timely surgical intervention. The comparison of AUC values between radiomic models utilized the pairwise DeLong test, with corresponding p values provided in Supplement E7. In the training cohort, four p values derived from the DeLong test were less than 0.05, suggesting that the combination of the GTV and PTV radiomic features with clinical parameters may surpass the performance of a single radiomic feature. Notably, clinical parameters play a pivotal role in predicting OLM in patients with cT\\u003csub\\u003e1\\u0026thinsp;\\u0026minus;\\u0026thinsp;2\\u003c/sub\\u003eN\\u003csub\\u003e0\\u003c/sub\\u003eM\\u003csub\\u003e0\\u003c/sub\\u003e SCLC. In the validation cohort, only the comparison between the GTV model and the combined model yielded a p value less than 0.05, implying that the combined model has a greater predictive ability than the solitary GTV model. Additionally, for the first time, we employed radiomics for prediction, revealing its pioneering significance. Future enhancements with increased data volume will further boost the model's efficacy.\\u003c/p\\u003e \\u003cp\\u003eMoreover, our research offers notable advantages. First, this study pioneers the application of radiomic techniques for OLM prediction in cT\\u003csub\\u003e1\\u0026thinsp;\\u0026minus;\\u0026thinsp;2\\u003c/sub\\u003eN\\u003csub\\u003e0\\u003c/sub\\u003eM\\u003csub\\u003e0\\u003c/sub\\u003e SCLC patients, advancing clinical diagnostic proficiency and facilitating precise decision-making and tailored treatment. Second, as a multicentre study, this study included a substantial sample size within the realm of enhanced CT-based radiomic research. Third, the combined model consistently demonstrated stable and commendable performance across both the internal training and external validation datasets, while the nomogram provided visualization and served as a valuable clinical tool for predicting OLM in presurgical cT\\u003csub\\u003e1\\u0026thinsp;\\u0026minus;\\u0026thinsp;2\\u003c/sub\\u003eN\\u003csub\\u003e0\\u003c/sub\\u003eM\\u003csub\\u003e0\\u003c/sub\\u003e SCLC patients.\\u003c/p\\u003e \\u003cp\\u003eOur study has certain limitations. First, selection bias is inherent in retrospective studies and is exacerbated by a modest sample size. Second, diverse machine parameters across different hospitals may introduce variations. Nevertheless, this variability contributes to the robustness of the models we trained. Third, compared to traditional radiomic methods, deep learning enhances the prediction model's performance to some extent. Emerging machine learning technologies such as convolutional neural networks are particularly suitable for classification tasks. Our future studies will prioritize data from larger sample sizes and incorporate deep learning applications to further enhance the robustness and performance of our models.\\u003c/p\\u003e \\u003cp\\u003eIn conclusion, OLM is not rare and has a greater incidence than NSCLC. Our combined model, which incorporates both intra- and peritumoral radiomic features based on contrast-enhanced CT imaging, serves as a valuable tool for discerning OLM in cT\\u003csub\\u003e1\\u0026thinsp;\\u0026minus;\\u0026thinsp;2\\u003c/sub\\u003eN\\u003csub\\u003e0\\u003c/sub\\u003eM\\u003csub\\u003e0\\u003c/sub\\u003e SCLC patients, guiding individualized clinical decisions.\\u003c/p\\u003e\"},{\"header\":\"Abbreviations\",\"content\":\"\\u003cp\\u003eSCLC \\u0026nbsp; \\u0026nbsp; \\u0026nbsp;Small cell lung cancer\\u003c/p\\u003e\\n\\u003cp\\u003eNSCLC \\u0026nbsp; Non-small cell lung cancer\\u003c/p\\u003e\\n\\u003cp\\u003eOLM \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; Occult lymph node metastases\\u003c/p\\u003e\\n\\u003cp\\u003eCT \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; Computed tomography\\u003c/p\\u003e\\n\\u003cp\\u003eGTV \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp;Gross tumour volume\\u003c/p\\u003e\\n\\u003cp\\u003ePTV \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; Peritumoural volume\\u003c/p\\u003e\\n\\u003cp\\u003eGPTV \\u0026nbsp; \\u0026nbsp; \\u0026nbsp;Gross tumour volume\\u003c/p\\u003e\\n\\u003cp\\u003eICC \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp;Intraclass correlation coefficient\\u003c/p\\u003e\\n\\u003cp\\u003eLASSO \\u0026nbsp; \\u0026nbsp;Least absolute shrinkage and selection operator\\u003c/p\\u003e\\n\\u003cp\\u003eAUC \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp;Area under curve\\u003c/p\\u003e\"},{\"header\":\"Declarations\",\"content\":\"\\u003cp\\u003e\\u003cstrong\\u003eFunding\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eThis study received funding\\u0026nbsp;from\\u0026nbsp;the Chinese Academy of Medical Science Innovation Fund for Medical Sciences (CIFMS) (2021-I2M-C\\u0026amp;T-B-061) and\\u0026nbsp;the Beijing Hope Run Special Fund of Cancer Foundation of China (LC2022A22).\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eEthical approval\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eThe study was approved by the Cancer Hospital, Chinese Academy of Medical Sciences\\u0026nbsp;Ethics Commission\\u0026nbsp;(NCC2021C-213).\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eAcknowledgements\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eWe thank all the study participants and referring technicians for their participation in this study.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eAvailability of data and materials\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eThe datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eCompeting interests\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eThe authors of this manuscript declare that they have no competing interests.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eConsent for publication\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eNot applicable.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eConsent for publication\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eInformed consent was waived due to the retrospective nature.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eAuthors\\u0026rsquo; contributions\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eXT.Y., L.Z. and M.L: Study design. JL.R. and MW.L.: data analysis. X.J., C.L. and X.P.: manuscript writing. J.Z., LZ.L. and YF.C.: data collection. MW.L., L.M. and JM.J.: study supervision and manuscript revising. All authors read and approved the final manuscript.\\u003c/p\\u003e\"},{\"header\":\"References\",\"content\":\"\\u003col\\u003e\\u003cli\\u003e\\u003cspan\\u003eSung H, Ferlay J, Siegel RL, Laversanne M, Soerjomataram I, Jemal A, et al. Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2021;71(3):209\\u0026ndash;49.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eThai AA, Solomon BJ, Sequist LV, Gainor JF, Heist RS. Lung cancer. Lancet. 2021;398(10299):535\\u0026ndash;54.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eSiegel RL, Miller KD, Fuchs HE, Jemal A. Cancer statistics, 2022. CA Cancer J Clin. 2022;72(1):7\\u0026ndash;33.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eLad T, Piantadosi S, Thomas P, Payne D, Ruckdeschel J, Giaccone G. A prospective randomized trial to determine the benefit of surgical resection of residual disease following response of small cell lung cancer to combination chemotherapy. Chest. 1994;106(6 Suppl):320s\\u0026ndash;3s.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eBarnes H, See K, Barnett S, Manser R. Surgery for limited-stage small-cell lung cancer. Cochrane Database Syst Rev. 2017;4(4):Cd011917.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eMiyauchi E, Motoi N, Ono H, Ninomiya H, Ohyanagi F, Nishio M, et al. Distinct Characteristics of Small Cell Lung Cancer Correlate With Central or Peripheral Origin: Subtyping Based on Location and Expression of Transcription Factor TTF-1. Med (Baltim). 2015;94(51):e2324.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eNobashi T, Koyasu S, Nakamoto Y, Kubo T, Ishimori T, Kim YH, et al. Prognostic value of fluorine-18 fludeoxyglucose positron emission tomography parameters differs according to primary tumour location in small-cell lung cancer. Br J Radiol. 2016;89(1059):20150618.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eKanaji N, Sakai K, Ueda Y, Miyawaki H, Ishii T, Watanabe N, et al. Peripheral-type small cell lung cancer is associated with better survival and higher frequency of interstitial lung disease. Lung Cancer. 2017;108:126\\u0026ndash;33.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eUprety D, Arjyal L, Vallatharasu Y, Bista A, Borgert A, Fitzsimmons AJ, et al. Utilization of Surgery and Its Impact on Survival in Patients With Early Stage Small-cell Lung Cancer in the United States. Clin Lung Cancer. 2020;21(2):186\\u0026ndash;93e2.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eYang CJ, Chan DY, Shah SA, Yerokun BA, Wang XF, D'Amico TA, et al. Long-term survival after surgery compared with concurrent chemoradiation for node-negative small cell lung cancer. Ann Surg. 2018;268(6):1105\\u0026ndash;12.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eYang CF, Chan DY, Speicher PJ, Gulack BC, Wang X, Hartwig MG, et al. Role of adjuvant therapy in a population-based cohort of patients with early-stage small-cell lung cancer. J Clin Oncol. 2016;34(10):1057\\u0026ndash;64.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eGanti AKP, Loo BW, Bassetti M, Blakely C, Chiang A, D'Amico TA, et al. Small Cell Lung Cancer, Version 2.2022, NCCN Clinical Practice Guidelines in Oncology. J Natl Compr Canc Netw. 2021;19(12):1441\\u0026ndash;64.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eJett JR, Schild SE, Kesler KA, Kalemkerian GP. Treatment of small cell lung cancer: Diagnosis and management of lung cancer, 3rd ed: American college of chest physicians evidence-based clinical practice guidelines. Chest. 2013;143(5 Suppl):e400S-e19S.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eMoon Y, Kim KS, Lee KY, Sung SW, Kim YK, Park JK. Clinicopathologic factors associated with occult lymph node metastasis in patients with clinically diagnosed no lung adenocarcinoma. Ann Thorac Surg. 2016;101(5):1928\\u0026ndash;35.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eLi L, Ren S, Zhang Y, Guan Y, Zhao J, Liu J, et al. Risk factors for predicting the occult nodal metastasis in T1-2N0M0 NSCLC patients staged by PET/CT: Potential value in the clinic. Lung Cancer. 2013;81(2):213\\u0026ndash;7.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eStamatis G. Staging of lung cancer: The role of noninvasive, minimally invasive and invasive techniques. Eur Respir J. 2015;46(2):521\\u0026ndash;31.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eLi M, Wu N, Liu Y, Zheng R, Liang Y, Zhang W, et al. Regional nodal staging with 18F-FDG PET-CT in non-small cell lung cancer: Additional diagnostic value of CT attenuation and dual-time-point imaging. Eur J Radiol. 2012;81(8):1886\\u0026ndash;90.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eSchmidt-Hansen M, Baldwin DR, Hasler E, Zamora J, Abraira V, Roqu\\u0026eacute; IFM. PET-CT for assessing mediastinal lymph node involvement in patients with suspected resectable non-small cell lung cancer. Cochrane Database Syst Rev. 2014;2014(11):Cd009519.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eKayawake H, Chen-Yoshikawa TF, Oda H, Motoyama H, Hamaji M, Menju T, et al. Complications of endobronchial ultrasound-guided transbronchial needle aspiration. Ann Thorac Surg. 2017;104(5):e363\\u0026ndash;e5.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eGillies RJ, Kinahan PE, Hricak H. Radiomics: Images Are More than Pictures, They Are Data. Radiology. 2016;278(2):563\\u0026ndash;77.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eGillies RJ, Kinahan PE, Hricak H, Radiomics. Images are more than pictures, they are data. Radiology. 2016;278(2):563\\u0026ndash;77.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eZhang R, Zhang R, Luan T, Liu B, Zhang Y, Xu Y, et al. A Radiomics Nomogram for Preoperative Prediction of Clinical Occult Lymph Node Metastasis in cT1-2N0M0 Solid Lung Adenocarcinoma. Cancer Manag Res. 2021;13:8157\\u0026ndash;67.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eSha X, Gong G, Qiu Q, Duan J, Li D, Yin Y. Discrimination of mediastinal metastatic lymph nodes in NSCLC based on radiomic features in different phases of CT imaging. BMC Med Imaging. 2020;20(1):12.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eWang L, Li T, Hong J, Zhang M, Ouyang M, Zheng X, et al. (18)F-FDG PET-based radiomics model for predicting occult lymph node metastasis in clinical N0 solid lung adenocarcinoma. Quant Imaging Med Surg. 2021;11(1):215\\u0026ndash;25.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eWang X, Zhao X, Li Q, Xia W, Peng Z, Zhang R, et al. Can peritumoral radiomics increase the efficiency of the prediction for lymph node metastasis in clinical stage T1 lung adenocarcinoma on CT? Eur Radiol. 2019;29(11):6049\\u0026ndash;58.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eTunali I, Hall LO, Napel S, Cherezov D, Guvenis A, Gillies RJ, et al. Stability and reproducibility of computed tomography radiomic features extracted from peritumoral regions of lung cancer lesions. Med Phys. 2019;46(11):5075\\u0026ndash;85.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eIshida T, Yano T, Maeda K, Kaneko S, Tateishi M, Sugimachi K. Strategy for lymphadenectomy in lung cancer three centimeters or less in diameter. Ann Thorac Surg. 1990;50(5):708\\u0026ndash;13.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eInoue M, Minami M, Shiono H, Sawabata N, Ideguchi K, Okumura M. Clinicopathologic study of resected, peripheral, small-sized, non-small cell lung cancer tumors of 2 cm or less in diameter: Pleural invasion and increase of serum carcinoembryonic antigen level as predictors of nodal involvement. J Thorac Cardiovasc Surg. 2006;131(5):988\\u0026ndash;93.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eBao F, Yuan P, Yuan X, Lv X, Wang Z, Hu J. Predictive risk factors for lymph node metastasis in patients with small size non-small cell lung cancer. J Thorac Dis. 2014;6(12):1697\\u0026ndash;703.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eTravis WD, Brambilla E, Nicholson AG, Yatabe Y, Austin JHM, Beasley MB, et al. The 2015 World Health Organization classification of lung tumors: Impact of genetic, clinical and radiologic advances since the 2004 classification. J Thorac Oncol. 2015;10(9):1243\\u0026ndash;60.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eRami-Porta R, Asamura H, Travis WD, Rusch VW. Lung cancer - major changes in the american joint committee on cancer eighth edition cancer staging manual. CA Cancer J Clin. 2017;67(2):138\\u0026ndash;55.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eWang T, She Y, Yang Y, Liu X, Chen S, Zhong Y, et al. Radiomics for survival risk stratification of clinical and pathologic stage IA pure-solid non-small cell lung cancer. Radiology. 2022;302(2):425\\u0026ndash;34.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eZhuo Y, Feng M, Yang S, Zhou L, Ge D, Lu S, et al. Radiomics nomograms of tumors and peritumoral regions for the preoperative prediction of spread through air spaces in lung adenocarcinoma. Transl Oncol. 2020;13(10):100820.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eShafiq-Ul-Hassan M, Zhang GG, Latifi K, Ullah G, Hunt DC, Balagurunathan Y, et al. Intrinsic dependencies of CT radiomic features on voxel size and number of gray levels. Med Phys. 2017;44(3):1050\\u0026ndash;62.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003evan Griethuysen JJM, Fedorov A, Parmar C, Hosny A, Aucoin N, Narayan V, et al. Computational Radiomics System to Decode the Radiographic Phenotype. Cancer Res. 2017;77(21):e104\\u0026ndash;e7.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eChoe J, Lee SM, Do KH, Lee G, Lee JG, Lee SM, et al. Deep Learning-based Image Conversion of CT Reconstruction Kernels Improves Radiomics Reproducibility for Pulmonary Nodules or Masses. Radiology. 2019;292(2):365\\u0026ndash;73.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eVasquez MM, Hu C, Roe DJ, Halonen M, Guerra S. Measurement error correction in the least absolute shrinkage and selection operator model when validation data are available. Stat Methods Med Res. 2019;28(3):670\\u0026ndash;80.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eVickers AJ, Elkin EB. Decision curve analysis: a novel method for evaluating prediction models. Med Decis Making. 2006;26(6):565\\u0026ndash;74.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eDeLong ER, DeLong DM, Clarke-Pearson DL. Comparing the areas under two or more correlated receiver operating characteristic curves: A nonparametric approach. Biometrics. 1988;44(3):837\\u0026ndash;45.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eCai JS, Yang F, Wang X. Occult lymph node metastasis is not a favorable factor for resected NSCLC patients. BMC Cancer. 2023;23(1):822.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eBeyaz F, Verhoeven RLJ, Schuurbiers OCJ, Verhagen A, van der Heijden E. Occult lymph node metastases in clinical N0/N1 NSCLC; A single center in-depth analysis. Lung Cancer. 2020;150:186\\u0026ndash;94.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eGw\\u0026oacute;źdź P, Pasieka-Lis M, Kołodziej K, Pankowski J, Banaś R, Wiłkojć M, et al. Prognosis of Patients With Stages I and II Non-Small Cell Lung Cancer With Nodal Micrometastases. Ann Thorac Surg. 2018;105(5):1551\\u0026ndash;7.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eHaque W, Singh A, Park HS, Teh BS, Butler EB, Zeng M, et al. Quantifying the rate and predictors of occult lymph node involvement in patients with clinically node-negative non-small cell lung cancer. Acta Oncol. 2022;61(4):403\\u0026ndash;8.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eDeng J, Zhong Y, Wang T, Yang M, Ma M, Song Y, et al. Lung cancer with PET/CT-defined occult nodal metastasis yields favourable prognosis and benefits from adjuvant therapy: a multicentre study. Eur J Nucl Med Mol Imaging. 2022;49(7):2414\\u0026ndash;24.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eMoon Y, Choi SY, Park JK, Lee KY. Risk Factors for Occult Lymph Node Metastasis in Peripheral Non-Small Cell Lung Cancer with Invasive Component Size 3 cm or Less. World J Surg. 2020;44(5):1658\\u0026ndash;65.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eDingemans AC, Fr\\u0026uuml;h M, Ardizzoni A, Besse B, Faivre-Finn C, Hendriks LE, et al. Small-cell lung cancer: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up(☆). Ann Oncol. 2021;32(7):839\\u0026ndash;53.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eRudin CM, Brambilla E, Faivre-Finn C, Sage J. Small-cell lung cancer. Nat Rev Dis Primers. 2021;7(1):3.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eHe XQ, Luo TY, Li X, Huo JW, Gong JW, Li Q. Clinicopathological and computed tomographic features associated with occult lymph node metastasis in patients with peripheral solid non-small cell lung cancer. Eur J Radiol. 2021;144:109981.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eG\\u0026oacute;mez-Caro A, Boada M, Caba\\u0026ntilde;as M, Sanchez M, Arguis P, Lome\\u0026ntilde;a F, et al. False-negative rate after positron emission tomography/computer tomography scan for mediastinal staging in cI stage non-small-cell lung cancer. Eur J Cardiothorac Surg. 2012;42(1):93\\u0026ndash;100. discussion.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eKim JY, Park JE, Jo Y, Shim WH, Nam SJ, Kim JH, et al. Incorporating diffusion- and perfusion-weighted MRI into a radiomics model improves diagnostic performance for pseudoprogression in glioblastoma patients. Neuro Oncol. 2019;21(3):404\\u0026ndash;14.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eLim K, Hsin MKY, Commentary. Resection for small cell lung cancer should be offered more often, and preferably anatomical. J Thorac Cardiovasc Surg. 2021;161(4):1495\\u0026ndash;6.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eNavani N, Spiro SG. PET scanning is important in lung cancer; but it has its limitations. Respirology. 2010;15(8):1149\\u0026ndash;51.\\u003c/span\\u003e\\u003c/li\\u003e\\u003c/ol\\u003e\"}],\"fulltextSource\":\"\",\"fullText\":\"\",\"funders\":[],\"hasAdminPriorityOnWorkflow\":false,\"hasManuscriptDocX\":true,\"hasOptedInToPreprint\":true,\"hasPassedJournalQc\":\"\",\"hasAnyPriority\":false,\"hideJournal\":false,\"highlight\":\"\",\"institution\":\"\",\"isAcceptedByJournal\":true,\"isAuthorSuppliedPdf\":false,\"isDeskRejected\":\"\",\"isHiddenFromSearch\":false,\"isInQc\":false,\"isInWorkflow\":false,\"isPdf\":false,\"isPdfUpToDate\":true,\"isWithdrawnOrRetracted\":false,\"journal\":{\"display\":true,\"email\":\"info@researchsquare.com\",\"identity\":\"respiratory-research\",\"isNatureJournal\":false,\"hasQc\":true,\"allowDirectSubmit\":false,\"externalIdentity\":\"rere\",\"sideBox\":\"Learn more about [Respiratory Research](http://respiratory-research.biomedcentral.com/)\",\"snPcode\":\"12931\",\"submissionUrl\":\"https://submission.nature.com/new-submission/12931/3\",\"title\":\"Respiratory Research\",\"twitterHandle\":\"@RespiratoryBMC\",\"acdcEnabled\":true,\"dfaEnabled\":true,\"editorialSystem\":\"em\",\"reportingPortfolio\":\"BMC/SO AJ\",\"inReviewEnabled\":true,\"inReviewRevisionsEnabled\":true},\"keywords\":\"Small-cell lung cancer, Occult lymph node metastases, Contrast-enhanced computed tomography, Prediction model Radiomics\",\"lastPublishedDoi\":\"10.21203/rs.3.rs-3832084/v1\",\"lastPublishedDoiUrl\":\"https://doi.org/10.21203/rs.3.rs-3832084/v1\",\"license\":{\"name\":\"CC BY 4.0\",\"url\":\"https://creativecommons.org/licenses/by/4.0/\"},\"manuscriptAbstract\":\"\\u003ch2\\u003eBackground\\u003c/h2\\u003e \\u003cp\\u003eThis study aimed to explore the incidence of occult lymph node metastasis (OLM) in clinical T\\u003csub\\u003e1\\u0026thinsp;\\u0026minus;\\u0026thinsp;2\\u003c/sub\\u003eN\\u003csub\\u003e0\\u003c/sub\\u003eM\\u003csub\\u003e0\\u003c/sub\\u003e (cT\\u003csub\\u003e1\\u0026thinsp;\\u0026minus;\\u0026thinsp;2\\u003c/sub\\u003eN\\u003csub\\u003e0\\u003c/sub\\u003eM\\u003csub\\u003e0\\u003c/sub\\u003e) small cell lung cancer (SCLC) patients and develop machine learning prediction models using preoperative intratumoral and peritumoral contrast-enhanced CT-based radiomic data.\\u003c/p\\u003e\\u003ch2\\u003eMethods\\u003c/h2\\u003e \\u003cp\\u003eBy conducting a retrospective analysis involving 242 eligible patients from 4 centres, we determined the incidence of OLM in cT\\u003csub\\u003e1\\u0026thinsp;\\u0026minus;\\u0026thinsp;2\\u003c/sub\\u003eN\\u003csub\\u003e0\\u003c/sub\\u003eM\\u003csub\\u003e0\\u003c/sub\\u003e SCLC patients. For each lesion, two ROIs were defined using the gross tumour volume (GTV) and peritumoral volume 15 mm around the tumour (PTV). By extracting a comprehensive set of 1595 enhanced CT-based radiomic features individually from the GTV and PTV, we rigorously evaluated the model performance using various metrics, including the area under the curve (AUC), accuracy, sensitivity, specificity, calibration curve, and decision curve analysis (DCA). For enhanced clinical applicability, we formulated a nomogram that integrates clinical parameters and the rad_score (GTV and PTV).\\u003c/p\\u003e\\u003ch2\\u003eResults\\u003c/h2\\u003e \\u003cp\\u003eThe initial investigation revealed a 33.9% OLM positivity rate in cT\\u003csub\\u003e1\\u0026thinsp;\\u0026minus;\\u0026thinsp;2\\u003c/sub\\u003eN\\u003csub\\u003e0\\u003c/sub\\u003eM\\u003csub\\u003e0\\u003c/sub\\u003e SCLC patients. Our combined model, which incorporates three radiomic features from the GTV and PTV, along with two clinical parameters (smoking status and shape), exhibited robust predictive capabilities. With a peak AUC value of 0.772 in the external validation cohort, the model outperformed the alternative models. The nomogram significantly enhanced diagnostic precision for radiologists and added substantial value to the clinical decision-making process for T\\u003csub\\u003e1\\u0026thinsp;\\u0026minus;\\u0026thinsp;2\\u003c/sub\\u003eN\\u003csub\\u003e0\\u003c/sub\\u003eM\\u003csub\\u003e0\\u003c/sub\\u003e SCLC patients.\\u003c/p\\u003e\\u003ch2\\u003eConclusions\\u003c/h2\\u003e \\u003cp\\u003eThe incidence of OLM in SCLC patients surpassed that in non-small cell lung cancer patients. The combined model demonstrated a notable generalization effect, effectively distinguishing between positive and negative OLMs in a noninvasive manner, thereby guiding individualized clinical decisions for patients with cT\\u003csub\\u003e1\\u0026thinsp;\\u0026minus;\\u0026thinsp;2\\u003c/sub\\u003eN\\u003csub\\u003e0\\u003c/sub\\u003eM\\u003csub\\u003e0\\u003c/sub\\u003e SCLC.\\u003c/p\\u003e\",\"manuscriptTitle\":\"Incidence Rate of Occult Lymph Node Metastasis in Clinical T 1-2 N 0 M 0 Small Cell Lung Cancer Patients and Radiomic Prediction Based on Contrast-enhanced CT Imaging: A Multicentre Study\",\"msid\":\"\",\"msnumber\":\"\",\"nonDraftVersions\":[{\"code\":1,\"date\":\"2024-01-08 17:10:08\",\"doi\":\"10.21203/rs.3.rs-3832084/v1\",\"editorialEvents\":[{\"type\":\"communityComments\",\"content\":0},{\"type\":\"decision\",\"content\":\"Revision requested\",\"date\":\"2024-02-02T01:24:56+00:00\",\"index\":\"\",\"fulltext\":\"\"},{\"type\":\"editorInvitedReview\",\"content\":\"\",\"date\":\"2024-02-01T09:57:44+00:00\",\"index\":\"hide\",\"fulltext\":\"\"},{\"type\":\"reviewerAgreed\",\"content\":\"c216d6ec-0627-47dc-a77b-83a777702fb5\",\"date\":\"2024-01-24T08:30:40+00:00\",\"index\":\"hide\",\"fulltext\":\"\"},{\"type\":\"editorInvitedReview\",\"content\":\"\",\"date\":\"2024-01-20T22:28:47+00:00\",\"index\":\"hide\",\"fulltext\":\"\"},{\"type\":\"reviewerAgreed\",\"content\":\"e4bdb599-ed02-4bd1-8953-c7127be68a97\",\"date\":\"2024-01-10T21:52:08+00:00\",\"index\":\"hide\",\"fulltext\":\"\"},{\"type\":\"reviewersInvited\",\"content\":\"\",\"date\":\"2024-01-08T02:56:04+00:00\",\"index\":\"\",\"fulltext\":\"\"},{\"type\":\"editorAssigned\",\"content\":\"\",\"date\":\"2024-01-04T19:23:42+00:00\",\"index\":\"\",\"fulltext\":\"\"},{\"type\":\"checksComplete\",\"content\":\"\",\"date\":\"2024-01-04T04:59:53+00:00\",\"index\":\"\",\"fulltext\":\"\"},{\"type\":\"submitted\",\"content\":\"Respiratory Research\",\"date\":\"2024-01-03T14:28:49+00:00\",\"index\":\"\",\"fulltext\":\"\"}],\"status\":\"published\",\"journal\":{\"display\":true,\"email\":\"info@researchsquare.com\",\"identity\":\"respiratory-research\",\"isNatureJournal\":false,\"hasQc\":true,\"allowDirectSubmit\":false,\"externalIdentity\":\"rere\",\"sideBox\":\"Learn more about [Respiratory Research](http://respiratory-research.biomedcentral.com/)\",\"snPcode\":\"12931\",\"submissionUrl\":\"https://submission.nature.com/new-submission/12931/3\",\"title\":\"Respiratory Research\",\"twitterHandle\":\"@RespiratoryBMC\",\"acdcEnabled\":true,\"dfaEnabled\":true,\"editorialSystem\":\"em\",\"reportingPortfolio\":\"BMC/SO AJ\",\"inReviewEnabled\":true,\"inReviewRevisionsEnabled\":true}}],\"origin\":\"\",\"ownerIdentity\":\"cc7d00c1-f469-4406-9192-e0d5f3fbb8b9\",\"owner\":[],\"postedDate\":\"January 8th, 2024\",\"published\":true,\"recentEditorialEvents\":[],\"rejectedJournal\":[],\"revision\":\"\",\"amendment\":\"\",\"status\":\"under-review\",\"subjectAreas\":[],\"tags\":[],\"updatedAt\":\"2024-05-16T00:23:26+00:00\",\"versionOfRecord\":[],\"versionCreatedAt\":\"2024-01-08 17:10:08\",\"video\":\"\",\"vorDoi\":\"\",\"vorDoiUrl\":\"\",\"workflowStages\":[]},\"version\":\"v1\",\"identity\":\"rs-3832084\",\"journalConfig\":\"researchsquare\"},\"__N_SSP\":true},\"page\":\"/article/[identity]/[[...version]]\",\"query\":{\"redirect\":\"/article/rs-3832084\",\"identity\":\"rs-3832084\",\"version\":[\"v1\"]},\"buildId\":\"qtupq5eGEP_6zYnWcrvyt\",\"isFallback\":false,\"isExperimentalCompile\":false,\"dynamicIds\":[84888],\"gssp\":true,\"scriptLoader\":[]}","source_license":"CC-BY-4.0","license_restricted":false}