Radiomics Nomogram for Preoperative Differentiation Between Clinical Stage IA Solitary Pulmonary Nodule-Type Invasive Mucinous Adenocarcinoma and Invasive Non-mucinous Adenocarcinoma

Radiomics Nomogram for Preoperative Differentiation Between Clinical Stage IA Solitary Pulmonary Nodule-Type Invasive Mucinous Adenocarcinoma and Invasive Non-mucinous Adenocarcinoma | 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 Radiomics Nomogram for Preoperative Differentiation Between Clinical Stage IA Solitary Pulmonary Nodule-Type Invasive Mucinous Adenocarcinoma and Invasive Non-mucinous Adenocarcinoma Sen Hong, Wu Ge, Yanping Wu, Yinjun Zhou, Haibo Liu, Shanyue Lin This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-3831470/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Background Radiomic applications for differentiating clinical stage IA solitary pulmonary nodule (SPN)-type invasive mucinous adenocarcinoma (IMA) from SPN-presenting lung adenocarcinoma (LADC) are lacking. Therefore, this study aimed to develop and validate predictive models for the preoperative differentiation between SPN-IMA and invasive non-mucinous LADC using computed tomography (CT) radiological and radiomic features. Methods In this bicentric study, we collected 507 SPNs, of which 42 were diagnosed as IMA and 465 as invasive non-mucinous LADC. The patients were randomly divided into training and test sets at a ratio of 7:3. The minimal redundancy maximal relevance filter was used to extract radiomic features, and the least absolute shrinkage and selection operator regression was used to screen these features and calculate the individualized radiomic score (rad score). We constructed a prediction nomogram that integrated radiomics and CT radiological features by applying multivariate logistic regression. Diagnostic capabilities were assessed by comparing the receiver operating characteristic and area under the curve (AUC) values. Results The combined model achieved AUC values of 0.789 and 0.798 for the training and test sets, respectively, surpassing those of the radiomics model in both the training (p = 0.038) and test (p = 0.021) sets. Moreover, the combined model performed better than the clinical model in the training (p = 0.017) and test (p = 0.025) sets. We transformed this combined model into a nomogram that accurately quantifies the risk of IMA and demonstrates exceptional discrimination and calibration. Conclusions The combined nomogram, incorporating radiomics and CT radiological features, is potentially valuable for the preoperative differentiation between clinical stage IA SPN-type IMA and invasive non-mucinous LADC. adenocarcinoma nomogram radiomics mucinous Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Background Lung cancer, particularly lung adenocarcinoma (LADC), is a significant global public health concern, accounting for > 40% of all cases [ 1 ]. Invasive mucinous adenocarcinoma (IMA) is a rare histological adenocarcinoma subtype, accounting for only 2–5% of all LADCs. Despite its low prevalence, various studies have investigated the unique characteristics of IMA, including its distinct clinical, pathological, and genomic profiles, in comparison with those of invasive non-mucinous LADC [ 2 – 5 ]. Notably, patients with IMA often exhibit specific gene mutations, such as anaplastic lymphoma kinase or Kirsten rat sarcoma viral oncogene [ 2 – 5 ]. Histologically, IMAs are characterized by tumor cells exhibiting a goblet or columnar cell morphology, accompanied by abundant intracytoplasmic mucin. Moreover, patients diagnosed with IMA often exhibit the frequent occurrence of aerogenous dissemination and the presence of satellite tumors surrounding the primary lesion [ 6 ]. In 2015, recognizing these distinctive features, the World Health Organization officially designated IMA as a variant of invasive LADC [ 7 ]. Furthermore, scientific investigations have revealed that IMA generally exhibits an inferior overall prognosis than non-mucinous LADC, even in the early stages [ 8 – 10 ]. IMA has a distinctive feature within the spectrum of LADC; therefore, achieving a preoperative diagnosis is crucial for optimizing patient outcomes, particularly in early-stage cases. Previous studies investigated the imaging characteristics of IMA and categorized them into two main types based on computed tomography (CT) findings: solitary pulmonary nodule (SPN)-type and pneumonia-type IMA [ 11 ]. Nie et al. [ 12 ] discovered that pneumonic-type IMA exhibited significantly poorer disease-free survival than SPN-type IMA. In addition, Wang et al. [ 13 ] demonstrated that the prognosis of pneumonic-type IMA was significantly poorer than that of SPN-type IMA. Traditionally, mucinous bronchioloalveolar carcinomas have been linked with either a multifocal disease or a pattern resembling pneumonia on CT scans [ 14 ]. However, recent studies have shown that SPN-type IMA is more common than pneumonia-type IMA [ 8 , 15 ]. CT images currently provide limited information for differentiating between SPN-type IMA and non-mucinous LADC. Previous studies have suggested that certain CT features, including vacuolar signs, air bronchograms, and abnormal vascular changes, may help predict SPN-type IMA [ 16 – 18 ]. However, interpreting these features can be subjective and lack consistency or typicality. Therefore, the accurate differentiation between IMA and invasive non-mucinous LADC based solely on preoperative CT images remains challenging. Radiomics provides a modality to extract quantitative and high-throughput information from pulmonary images, capturing the intrinsic pathophysiology and offering valuable insights into tumor phenotypes [ 19 ]. Yu et al. [ 20 ] developed a nomogram that combined clinical variables and CT-based radiomic features, showing good diagnostic performance to effectively differentiate pneumonia-type IMA from pneumonia. Similarly, Zhang et al. [ 21 ] established the diagnostic value of a radiomics-based nomogram to distinguish SPN-type IMA from pulmonary tuberculoma, demonstrating favorable diagnostic performance. However, radiomic applications that differentiate early-stage SPN-IMA from SPN-presenting LADC are currently lacking. Therefore, we aimed to develop a radiomics nomogram specifically for this purpose, which can help in clinical decision-making. Methods Patient selection Following the Declaration of Helsinki, ethical approval was obtained from the Ethics Committee (reference number: 2021-07-009) of the participating hospital, which ensured that the rights and interests of the research participants were not compromised. Notably, all methodologies were implemented in strict adherence to the applicable guidelines and regulations. The study participants were granted an exemption from the requirement of consent. Between March 2020 and October 2023, 935 surgically resected solitary LADCs were identified at Center 1 (Guilin Medical College Affiliated Hospital) and Center 2 (Xiangtan Central Hospital). The resected tumor glass slides were assessed by experienced pathologists [ 18 ]. This study included 77 patients with SPN-type IMA and 569 with solitary invasive non-mucinous LADC. The patients' inclusion criteria were as follows: (1) histopathological confirmation of IMA, (2) solid nodules with a maximum diameter ranging from 0.5 cm to 3.0 cm, showing no cavities, calcification, vacuoles, and lacking ground glass density, (3) availability of complete thin-slice CT image data (0.625–1.25 mm/slice) within a span of 2 weeks prior to the pathological diagnosis. Patients were excluded if they met any of the following criteria: (1) presence of multiple LADCs, (2) receiving antitumor therapy before undergoing CT examination and receiving the pathological diagnosis, (3) pathological diagnosis of noninvasive non-mucinous LADCs, or (4) demonstrating lymph node or distant metastases. It is worth noting that lymph node involvement and distant metastasis are less prevalent in IMA compared to other invasive non-mucinous LADCs [ 6 ]. Ultimately, a total of 507 solitary pulmonary nodules were obtained, comprising 42 cases diagnosed as IMA and 465 diagnosed as invasive non-mucinous LADC. The division of training and test sets was randomly performed in a 7:3 ratio, without any deliberate partitioning, resulting in 354 cases allocated for the training set and 153 cases for the test set. Figure 1 provides a flowchart illustrating the patient inclusion process. CT examinations Two centers conducted unenhanced CT using a 64- or 128-detector row CT system. Center 1 employed the Revolution CT (GE Healthcare, Chicago, IL, USA) or MX16 CT (Philips Healthcare, Best, The Netherlands). Center 2 used the uCT550 or uCT760 systems (Shanghai United Imaging Healthcare, Shanghai, China). The scans implemented specific parameters, including a 120 kV tube voltage, tube current-time product ranging from 180 to 280 mA, beam pitch of 0.515 and 0.758, matrix size of 512 × 512, and standard resolution algorithms. CT radiological features evaluation Following CT scanning, the unprocessed data underwent transfer to a post-processing workstation to enable multiplanar reconstruction. Subsequently, analysis was performed using the lung window, characterized by a window width of 1200 HU and a level of -600 HU, as well as the mediastinal window with a window width of 400 HU and a level of 40 HU. The recorded and analyzed CT image features included size, density, shape, boundary, spiculation, lobulation, vascular convergence, and vacuole signs. Two board-certified thoracic radiologists (5 and 10 years of experience in chest CT imaging) independently analyzed the CT radiological features. The patients were kept unaware of the clinical and histological findings. Disagreements between the radiologists were resolved through consensus achieved during discussions to reconcile any discrepancies in the qualitative indicators. Preprocessing of CT images and tumor segmentation Initially, the CT images underwent standard resampling and grayscale discretization. Subsequently, utilizing ITK-SNAP software (version 4.0, www.itksnap.org ), an adept thoracic radiologist with 5 years of experience meticulously delineated the tumor boundary slice by slice, thereby generating a precise volume of interest. Following this, a second radiologist, with a decade of experience in the same field, thoroughly reviewed the lesion delineations and made requisite adjustments as needed. Radiomics feature extraction and data preprocessing The Pyradiomics function package ( https://pyradiomics.readthedocs.io ) was used to extract radiomic features from the CT images. This powerful package facilitates the extraction of a comprehensive set of 1239 radiomic features. CT images sourced from diverse hospitals and various protocols were included; therefore, all the radiomic features' intensities were effectively normalized using the z-score transformation (z = [x-µ]/σ). Radiomics feature selection and model construction We employed the minimum redundancy maximum correlation (mRMR) approach to prevent overfitting by selecting the most pertinent features for tumor classification while minimizing redundancy. We used the least absolute shrinkage and selection operator (LASSO) regression model with 10-fold cross-validation in the feature selection phase to identify features with nonzero coefficients under the optimal λ. The selected features were then used to construct a radiomics model. The rad score was computed by summing the selected features, each weighted by its respective coefficient. Wilcoxon tests were conducted to discern disparities between the invasive non-mucinous LADC and IMA groups. These procedures were initially applied to the training set and were subsequently extended to the test set. Clinical model and nomogram construction A multivariate logistic regression analysis was conducted in the training set using clinical and CT radiological features that exhibited a p-value less than 0.1 in the univariate logistic analysis. To establish the clinical prediction model, the best combinations of variables were selected through a backward stepwise selection process. The clinical model incorporated the predictive variables, and the radiomics model contributed the rad score, resulting in the creation of a combined prediction model and a corresponding nomogram. The predictive performance of each model was evaluated in both the training and test sets. The radiomics process, encompassing region of interest delineation, feature extraction, dimensionality reduction, feature selection, and model construction, is vividly illustrated in Fig. 2. Statistical analysis The R software (version 4.3.1; https://www.r-project.org ) was employed for the statistical analysis. The Student's t-test was applied to assess normally distributed continuous variables, while the Mann–Whitney U test was utilized for non-normally distributed data. For categorical variables, the chi-square test was employed. Receiver operating characteristic analysis was conducted to compute the area under the curve (AUC), along with the corresponding sensitivity, specificity, and accuracy measures. DeLong's test was employed to assess the statistical significance of differences between the AUC values. Statistical significance was established at a p-value < 0.05. Results Baseline characteristics In total, 507 patients with SPNs were included in the study; 465 (91.7%) were diagnosed pathologically with invasive non-mucinous LADC and 42 (8.3%) with IMA. Among them, 376 and 162 patients were included in the training and test sets, respectively. The two cohorts showed no significant differences (Table 1 ). Table 1 Comparative Analysis of Radiological Features among Training, and Test set Variables Total (N = 507) Training Set (N = 354) Test Set (N = 153) p-value Group, N (%) 0.951 Invasive Non-Mucinous LADC 465 (91.7) 324 (91.5) 141 (92.2) IMA 42 (8.3) 30 (8.5) 12 (7.8) Location, N (%) 0.253 RUL 145 (28.6) 105 (29.7) 40 (26.1) RLL 119 (23.5) 81 (22.9) 38 (24.8) RML 40 (7.9) 22 (6.2) 18 (11.8) LUL 118 (23.3) 86 (24.3) 32 (20.9) LLL 85 (16.7) 60 (16.9) 25 (16.4) Boundary, N (%) 0.318 Ill-Defined 107 (21.1) 70 (19.8) 37 (24.2) Well-Defined 400 (78.9) 284 (80.2) 116 (75.8) Shape, N (%) 0.235 Irregular 273 (53.8) 184 (52.0) 89 (58.2) Others 234 (46.2) 170 (48.0) 64 (41.8) Lobulation, N (%) 1.000 Absence 120 (23.7) 84 (23.7) 36 (23.5) Presence 387 (76.3) 270 (76.3) 117 (76.5) Spiculation, N (%) 0.312 Absence 150 (29.6) 110 (31.1) 40 (26.1) Presence 357 (70.4) 244 (68.9) 113 (73.9) Vascular Convergence Sign, N (%) 0.821 Absence 264 (52.1) 186 (52.5) 78 (51.0) Presence 243 (47.9) 168 (47.5) 75 (49.0) Vacuole Sign, N (%) 0.652 Absence 403 (79.5) 279 (78.8) 124 (81) Presence 104 (20.5) 75 (21.2) 29 (19) Pleural Indentation, N (%) 1 Absence 117 (23.1) 82 (23.2) 35 (22.9) Presence 390 (76.9) 272 (76.8) 118 (77.1) Sex, N (%) 0.05 Male 244 (48.1) 181 (51.1) 63 (41.2) Female 263 (51.9) 173 (48.9) 90 (58.8) Age, Median (Q1, Q3) 61 (53, 67) 61.5 (54, 67.8) 60 (53, 67) 0.539 Clinical Stage, N (%) 0.579 cT1a 23 (4.5) 18 (5.1) 5 (3.3) cT1b 217 (42.8) 148 (41.8) 69 (45.1) cT1c 267 (52.7) 188 (53.1) 79 (51.6) Abbreviation : LUL Left Upper Lobe, LLL Left Lower Lobe, RUL Right Upper Lobe, RML Right Middle Lobe, RLL Right Lower Lobe, IMA Invasive Mucinous Adenocarcinoma, LADC Lung Adenocarcinoma Table S1 provides information on SPN-type IMA and invasive non-mucinous LADC. The training set showed no significant differences in demographic data, such as age and sex, but notable differences were found in CT radiological features. Specifically, SPN-type IMA exhibited a relatively higher proportion of vascular convergence signs, pleural indentation, and a higher clinical stage than invasive non-mucinous LADC (p = 0.003, p = 0.044, p = 0.002). Clinical prediction model Univariate and multivariate logistic analyses were performed to identify several CT features, such as the vascular convergence sign (odds ratio [OR], 1.053 [0.995–1.115], p = 0.047) and clinical stage (OR, 1.135 [0.994–1.296], p = 0.041), as independent risk factors (Table 2 ). These significant risk factors were used to select the most suitable combination of predictive variables for developing the clinical model. The AUC values for the training and test sets were 0.714 and 0.715, respectively. Table 2 Univariate and Multivariate Analysis for Discriminating Clinical Stage IA IMA from Invasive Non-Mucinous LADC Variables Univariate analysis p-value Multivariate analysis p-value Odd Ratio (95%CI) Odd Ratio (95%CI) Location 0.76 (0.26–2.2) 0.61 Boundary 0.54 (0.24–1.2) 0.15 Shape 0.6 (0.28–1.3) 0.2 Lobulation 3 (0.89-10) 0.077 Spiculation 2.4 (0.89–6.4) 0.083 Vascular convergence sign 2 (0.94–4.4) 0.043 1.053(0.995–1.115) 0.047 Vacuole sign 1.1 (0.47–2.8) 0.76 Pleural indentation 4.6 (1.1–20) 0.04 1.054(0.983–1.129) 0.139 Sex 0.91 (0.43–1.9) 0.8 Age 0.97 (0.94-1) 0.14 Clinical Stage 1.1 (1.1–1.2) < 0.001 1.135(0.994–1.296) 0.041 Abbreviation : IMA invasive mucinous adenocarcinoma, LADC lung adenocarcinoma Radiomics model Following the elimination of redundant and irrelevant features through the mRMR method, LASSO regression was employed to meticulously curate an optimized subset of features necessary for the construction of the ultimate model. A 10-fold cross-validation was subsequently executed to ascertain the optimal hyperparameter λ, as illustrated in Fig. 3. With an optimal λ value of 0.031, six features were discerningly chosen for the development of the radiomics models, as depicted in Fig. 4. Notably, the rad score for IMA exhibited a significant elevation compared to that of invasive non-mucinous LADC in both the training and test sets (all p < 0.05), as elucidated in Fig. 5. For the radiomic model, AUC values of 0.770 and 0.753 were obtained for the training and test sets, respectively. Figure 6 shows the detailed performance of the radiomics model using 5-fold cross-validation for a more comprehensive evaluation. The radiomics model using the cross-validated analysis demonstrated favorable predictive performance, with a mean AUC value of 0.771, in differentiating IMA from invasive non-mucinous LADC. Efficacy evaluation of combined models The predictive variables utilized for constructing the combined model and the corresponding nomogram (Fig. 7) consisted of rad scores derived from the radiomics model and the radiological CT features integrated in the clinical model. The AUC values of the combined model were calculated to be 0.789 and 0.798 for the training and test sets, respectively (Table 3 ). Importantly, based on the outcomes of the DeLong test, it was demonstrated that the combined model outperformed the radiomics model in both the training (p = 0.038) and test sets (p = 0.021). In addition, when compared with the clinical model, the combined model exhibited superior performance in both the training (p = 0.017) and test (p = 0.025) sets. These comparisons are depicted in Fig. 8. Table 3 Diagnostic Efficacy of the Combined Nomogram, Clinical model, and Radiomic Approaches in the Training and Validation Cohorts Prediction models AUC ACC Sensitivity Specificity Training set Combined 0.789 0.653 0.833 0.636 Clinical 0.714 0.525 0.833 0.497 Radiomics 0.770 0.647 0.833 0.630 Test set Combined 0.798 0.935 0.583 0.965 Clinical 0.715 0.569 0.833 0.546 Radiomics 0.753 0.686 0.750 0.681 Abbreviation : AUC, Area Under the Curve; ACC, Accuracy. In addition, the calibration plots of the nomogram in the training and test sets revealed that the uncalibrated and calibration prediction curves were closely aligned with the diagonal and did not significantly deviate from the ideal curve, indicating that the nomogram had a good calibration (Fig. 9 ). Discussion Our findings strongly indicate that integrating radiological CT features with radiomics models can significantly enhance the ability to differentiate between patients with IMA and those with invasive non-mucinous LADC, surpassing the diagnostic efficacy achieved using only radiological CT features or radiomics. Therefore, we successfully developed a combined nomogram with excellent discrimination and calibration to accurately distinguish between patients with SPN-type IMA and those with invasive non-mucinous LADC. SPN-type IMA is the primary manifestation of IMA, and ground-glass nodules are rare [ 17 ]. This manifestation distinction may be attributed to IMA’s histological characteristics, such as mucin-rich tumor cells, fibrosis, and central fibrosis with alveolar mucin-filled spaces [ 6 ]. Recent studies have indicated a higher prevalence of SPN-type IMA compared with pneumonia-type IMA [ 8 , 15 ]. Therefore, differentiating between SPN-type IMA and non-mucinous LADC is crucial, particularly in the early stages. This study found that SPN-type IMA exhibited more vascular convergence signs, pleural indentation, and advanced clinical stage than invasive non-mucinous LADC. These observations may be attributed to mucus leakage from the nodule's margin and the migration of macrophages along the alveolar walls and pores, leading to signs of vascular convergence and pleural indentation. These findings are consistent with those of Wu et al., Zhang et al., and Cha et al. [ 16 – 18 ]. However, the clinical prediction model using CT radiological features demonstrated a modest diagnostic efficiency, as evidenced by the AUC values of 0.714 and 0.715 for the training and test sets, respectively. This could be due to the subjective interpretation of the radiological features, which may lack consistency, typicality, and generalizability, particularly during the early stages. Radiomics is an advanced, noninvasive approach that uses intelligent algorithms to construct models based on original medical images. This method yields additional insights and potentially reveals pertinent phenotypic information by capturing tumor heterogeneity [ 19 ]. To our knowledge, no study has investigated the use of radiomic models in distinguishing between clinical stage IA SPN-type IMA and non-mucinous LADC. Notably, our study revealed an independent association between higher rad scores and IMA in the radiomics model. This observation can be explained by the heightened IMA heterogeneity, attributed to the abundant presence of mucin within the cytoplasm of IMA tumor cells and the aggressive nature of tumor neovascularization, closely associated with its malignant behavior [ 17 ]. The radiomics model achieved AUC values of 0.770 and 0.753 for the training and test sets, respectively, indicating higher diagnostic efficiency than that of the clinical model. This is because radiomics provides a notable advantage in evaluating tumor imaging phenotypes. It enables the extraction of quantitative features from CT images, allowing the identification of additional factors that cannot be easily comprehended visually using CT radiology. In addition, through five cross-validated analyses, the radiomics model demonstrated favorable predictive performance, with a mean AUC value of 0.771 in differentiating IMA from invasive non-mucinous LADC, indicating the model’s good robustness and avoidance of overfitting. By amalgamating crucial data from clinical and radiomics models, the combined model accurately distinguished between IMA and invasive non-mucinous LADC. It performed better than the clinical or radiomics models when used independently. We hypothesized that integrating diverse and abundant information within the combined model would increase diagnostic efficiency when distinguishing IMA from invasive non-mucinous LADC. This study has some limitations. First, it is crucial to acknowledge our study’s retrospective nature and the relatively low incidence rate of IMA stemming from the rarity of this condition. In addition, potential selection bias in our research arose from including only patients with postoperative pathologic results. Second, the nature of our bicentric study introduced variations in acquisition parameters, image quality, and potential co-registration errors, which could confound the analysis by contributing to the sources of variability. Furthermore, artifacts and technical limitations may affect the reliability and reproducibility of radiomic features. Third, we could not develop a predictive model for disease outcomes due to the relatively short postoperative follow-up period. Finally, our study focused solely on SPN-type IMA and invasive non-mucinous LADC with a diameter of ≤ 3 cm, which may limit the generalizability of our findings to other stages. In conclusion, our study successfully developed an innovative model using preoperative radiological and radiomic features to effectively differentiate between IMA and invasive non-mucinous LADC. We transformed this combined model into a nomogram that accurately quantified the risk of IMA. Notably, the nomogram exhibited exceptional discrimination and calibration, highlighting its potential value in clinical practice. Abbreviations AUC, area under the curve CT, computed tomography IMA, invasive mucinous adenocarcinoma LADC, lung adenocarcinoma LASSO, least absolute shrinkage and selection operator mRMR, minimum redundancy maximum correlation SPN, solitary pulmonary nodule Declarations Ethics approval and consent to participate Ethical approval was obtained from Medical Ethics Committee of Xiangtan Central Hospital (reference number: No. 2021-07-009), which ensured that the rights and interests of the research participants were not compromised. All methods were performed following the relevant guidelines and regulations. Medical Ethics Committee of Xiangtan Central Hospital approved the retrospective study and waived the requirement for informed consent. Consent for publication NA Availability of data and materials The datasets generated during and analyzed during the current study are publicly available. Competing interests The authors declare that they have No competing financial interests exist. Funding NA Authors' contributions S.H., W.G. and Y.W. performed the analyses and wrote the manuscript. Y.Z. and H.L. contributed to the conception of the paper. S.H. and W.G. contributed to analysis and prepared figures. Y.W prepared Tables. S.L. helped perform the analysis with constructive discussions. All authors contributed to the article and approved the submitted version. Acknowledgments We would like to express our sincere gratitude to S.H., W.G., Y.W., Y.Z., H.L. and S.L. for their invaluable contributions to this study. We confirm that all individuals have given permission to be named in the Acknowledgments section. References Lee HY, Lee SW, Lee KS, Jeong JY, Choi JY, Kwon OJ, Song SH, Kim EY, Kim J, Shim YM: Role of CT and PET Imaging in Predicting Tumor Recurrence and Survival in Patients with Lung Adenocarcinoma: A Comparison with the International Association for the Study of Lung Cancer/American Thoracic Society/European Respiratory Society Classification of Lung Adenocarcinoma. J Thorac Oncol 2015, 10 (12):1785-1794. Shim HS, Kenudson M, Zheng Z, Liebers M, Cha YJ, Hoang Ho Q, Onozato M, Phi Le L, Heist RS, Iafrate AJ: Unique Genetic and Survival Characteristics of Invasive Mucinous Adenocarcinoma of the Lung. J Thorac Oncol 2015, 10 (8):1156-1162. Kadota K, Yeh YC, D'Angelo SP, Moreira AL, Kuk D, Sima CS, Riely GJ, Arcila ME, Kris MG, Rusch VW et al: Associations between mutations and histologic patterns of mucin in lung adenocarcinoma: invasive mucinous pattern and extracellular mucin are associated with KRAS mutation. Am J Surg Pathol 2014, 38 (8):1118-1127. Guo M, Tomoshige K, Meister M, Muley T, Fukazawa T, Tsuchiya T, Karns R, Warth A, Fink-Baldauf IM, Nagayasu T et al: Gene signature driving invasive mucinous adenocarcinoma of the lung. EMBO Mol Med 2017, 9 (4):462-481. Shang G, Jin Y, Zheng Q, Shen X, Yang M, Li Y, Zhang L: Histology and oncogenic driver alterations of lung adenocarcinoma in Chinese. Am J Cancer Res 2019, 9 (6):1212-1223. Ichinokawa H, Ishii G, Nagai K, Yoshida J, Nishimura M, Hishida T, Suzuki K, Ochiai A: Clinicopathological characteristics of primary lung adenocarcinoma predominantly composed of goblet cells in surgically resected cases. Pathol Int 2011, 61 (7):423-429. Travis WD, Brambilla E, Nicholson AG, Yatabe Y, Austin JHM, Beasley MB, Chirieac LR, Dacic S, Duhig E, Flieder DB 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):1243-1260. Lee HY, Cha MJ, Lee KS, Lee HY, Kwon OJ, Choi JY, Kim HK, Choi YS, Kim J, Shim YM: Prognosis in Resected Invasive Mucinous Adenocarcinomas of the Lung: Related Factors and Comparison with Resected Nonmucinous Adenocarcinomas. J Thorac Oncol 2016, 11 (7):1064-1073. Matsui T, Sakakura N, Koyama S, Nakanishi K, Sasaki E, Kato S, Hosoda W, Murakami Y, Kuroda H, Yatabe Y: Comparison of Surgical Outcomes Between Invasive Mucinous and Non-Mucinous Lung Adenocarcinoma. Ann Thorac Surg 2021, 112 (4):1118-1126. Xu X, Shen W, Wang D, Li N, Huang Z, Sheng J, Rucker AJ, Mao W, Xu H, Cheng G: Clinical features and prognosis of resectable pulmonary primary invasive mucinous adenocarcinoma. Transl Lung Cancer Res 2022, 11 (3):420-431. Beck KS, Sung YE, Lee KY, Han DH: Invasive mucinous adenocarcinoma of the lung: Serial CT findings, clinical features, and treatment and survival outcomes. Thorac Cancer. 2020 , 11(12):3463-3472. Nie K, Nie W, Zhang YX, Yu H: Comparing clinicopathological features and prognosis of primary pulmonary invasive mucinous adenocarcinoma based on computed tomography findings. Cancer Imaging 2019, 19 (1):47. Wang T, Yang Y, Liu X, Deng J, Wu J, Hou L, Wu C, She Y, Sun X, Xie D et al: Primary Invasive Mucinous Adenocarcinoma of the Lung: Prognostic Value of CT Imaging Features Combined with Clinical Factors. Korean J Radiol 2021, 22 (4):652-662. Casali C, Rossi G, Marchioni A, Sartori G, Maselli F, Longo L, Tallarico E, Morandi U: A single institution-based retrospective study of surgically treated bronchioloalveolar adenocarcinoma of the lung: clinicopathologic analysis, molecular features, and possible pitfalls in routine practice. J Thorac Oncol 2010, 5 (6):830-836. Watanabe H, Saito H, Yokose T, Sakuma Y, Murakami S, Kondo T, Oshita F, Ito H, Nakayama H, Yamada K et al: Relation between thin-section computed tomography and clinical findings of mucinous adenocarcinoma. Ann Thorac Surg 2015, 99 (3):975-981. Wu J, Zhang T, Pan J, Zhang Q, Lin X, Chang L, Chen YC, Xue X: Characteristics of the Computed Tomography Imaging Findings in 72 Patients with Airway-Invasive Pulmonary Aspergillosis. Med Sci Monit 2021, 27 :e931162. Zhang X, Qiao W, Kang Z, Pan C, Chen Y, Li K, Shen W, Zhang L: CT Features of Stage IA Invasive Mucinous Adenocarcinoma of the Lung and Establishment of a Prediction Model. Int J Gen Med 2022, 15 :5455-5463. Cha MJ, Lee KS, Kim TJ, Kim HS, Kim TS, Chung MJ, Kim BT, Kim YS: Solitary Nodular Invasive Mucinous Adenocarcinoma of the Lung: Imaging Diagnosis Using the Morphologic-Metabolic Dissociation Sign. Korean J Radiol 2019, 20 (3):513-521. Tunali I, Gillies RJ, Schabath MB: Application of Radiomics and Artificial Intelligence for Lung Cancer Precision Medicine. Cold Spring Harb Perspect Med 2021, 11 (8). Yu X, Zhang S, Xu J, Huang Y, Luo H, Huang C, Nie P, Deng Y, Mao N, Zhang R et al: Nomogram Using CT Radiomics Features for Differentiation of Pneumonia-Type Invasive Mucinous Adenocarcinoma and Pneumonia: Multicenter Development and External Validation Study. AJR Am J Roentgenol 2023, 220 (2):224-234. Zhang J, Hao L, Qi M, Xu Q, Zhang N, Feng H, Shi G: Radiomics nomogram for preoperative differentiation of pulmonary mucinous adenocarcinoma from tuberculoma in solitary pulmonary solid nodules. BMC Cancer 2023, 23 (1):261. Additional Declarations No competing interests reported. Supplementary Files TableS1.doc Cite Share Download PDF Status: Posted Version 1 posted 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-3831470","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":266118036,"identity":"1b65ea21-6a35-4bf2-bd12-019b5d0ad6fd","order_by":0,"name":"Sen Hong","email":"","orcid":"","institution":"Guangzhou Women and Children's Medical Center, Guangzhou Medical University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Sen","middleName":"","lastName":"Hong","suffix":""},{"id":266118037,"identity":"1be17dd5-336c-46dc-973c-59c24e23837b","order_by":1,"name":"Wu Ge","email":"","orcid":"","institution":"Xiangtan Central Hospital","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Wu","middleName":"","lastName":"Ge","suffix":""},{"id":266118038,"identity":"4213b591-5b85-4f07-9613-d27ba488bfb6","order_by":2,"name":"Yanping Wu","email":"","orcid":"","institution":"Xiangtan Central Hospital","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Yanping","middleName":"","lastName":"Wu","suffix":""},{"id":266118039,"identity":"dda18e9c-0531-4335-9cef-2257ac377eac","order_by":3,"name":"Yinjun Zhou","email":"","orcid":"","institution":"Xiangtan Central Hospital","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Yinjun","middleName":"","lastName":"Zhou","suffix":""},{"id":266118040,"identity":"c822c612-06d9-4256-877f-7ea6b0245c5c","order_by":4,"name":"Haibo Liu","email":"","orcid":"","institution":"Xiangtan Central Hospital","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Haibo","middleName":"","lastName":"Liu","suffix":""},{"id":266118041,"identity":"6cd5bd6e-b1fe-404c-bbdc-e5a54ba15949","order_by":5,"name":"Shanyue Lin","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA/ElEQVRIie3RsUoDMRjA8e8I5Bw+zZqjPkTgBikUfZVIodMNfQNzBJz6ALm3EArOX8jgIroKXSq+wGV0qcaODjlHoflDIIT8SEIASqV/2EUaFA9fjLPKhM/DAoUwecIBKu84q0Vt+w/gq8vG0SRhATkTzebJtsDDQhk9QcSz8g75TL3d3s/W+IoKqBpjlyGyUzRKbI/EyR1eMcOa4TFHMJ2i5PKHpPkO54Y4O8+RdLGAWt09HIl+QUV6gkCXCGnWbLxtkegPRK7WfjDERN3374NZYuO8zb5FiLAdYyLpa/YUzfWNENaPMUMAztTvlcrk9qfq/cSGUqlUOvm+AcPfVZuTpmNWAAAAAElFTkSuQmCC","orcid":"","institution":"Affiliated Hospital of Guilin Medical University","correspondingAuthor":true,"submittingAuthor":false,"prefix":"","firstName":"Shanyue","middleName":"","lastName":"Lin","suffix":""}],"badges":[],"createdAt":"2024-01-03 09:30:44","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-3831470/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-3831470/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":49437439,"identity":"a6206c85-e3f3-4355-9921-9a72a16c6056","added_by":"auto","created_at":"2024-01-10 20:33:51","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":219209,"visible":true,"origin":"","legend":"\u003cp\u003eThis flowchart depicts the patient enrollment process for the study.\u003c/p\u003e","description":"","filename":"Figure1.png","url":"https://assets-eu.researchsquare.com/files/rs-3831470/v1/1ad5866d55a9080ff39fff25.png"},{"id":49437028,"identity":"e888d327-93d6-49c4-b4e6-c367aef6a589","added_by":"auto","created_at":"2024-01-10 20:17:51","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":1148623,"visible":true,"origin":"","legend":"\u003cp\u003eThis visual representation outlines the radiomics process, encompassing image acquisition, image segmentation, feature extraction, feature selection, model establishment, and model evaluation.\u003c/p\u003e","description":"","filename":"Figure2.png","url":"https://assets-eu.researchsquare.com/files/rs-3831470/v1/6f73050b9200284bdfe47db2.png"},{"id":49436688,"identity":"aefb5514-0053-44d8-8027-dcc957f208e9","added_by":"auto","created_at":"2024-01-10 20:09:51","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":121853,"visible":true,"origin":"","legend":"\u003cp\u003eEmploying LASSO regression for feature selection and assessing the predictive accuracy of the radiomics signature (3a). The tuning parameter (λ) selection was carried out through 10-fold cross-validation, adhering to stringent criteria (3b).\u003c/p\u003e","description":"","filename":"Figure3.png","url":"https://assets-eu.researchsquare.com/files/rs-3831470/v1/9fbe5be6d7957a04805af3ee.png"},{"id":49436691,"identity":"120824ec-6b38-481c-bdd0-c532e56d03f8","added_by":"auto","created_at":"2024-01-10 20:09:51","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":71307,"visible":true,"origin":"","legend":"\u003cp\u003eUnveiling the pinnacle radiomics signature, meticulously identified by harnessing the discerning capabilities of the LASSO classifier and the rigorous 10-fold cross-validation methodology.\u003c/p\u003e","description":"","filename":"Figure4.png","url":"https://assets-eu.researchsquare.com/files/rs-3831470/v1/87c73419261f7f6668fbc2f0.png"},{"id":49437238,"identity":"a16b309e-2650-41fa-a943-27266597ea9d","added_by":"auto","created_at":"2024-01-10 20:25:51","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":66437,"visible":true,"origin":"","legend":"\u003cp\u003ePresenting visually captivating violin plots that artfully portray the disparities in Rad scores between the IMA and invasive non-mucinous LADCs groups, meticulously observed across the training and validation cohorts.\u003c/p\u003e","description":"","filename":"Figure5.png","url":"https://assets-eu.researchsquare.com/files/rs-3831470/v1/7d178516d968609027754f34.png"},{"id":49436686,"identity":"f7dc9c45-8950-433e-8324-70de5e02cf78","added_by":"auto","created_at":"2024-01-10 20:09:51","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":57202,"visible":true,"origin":"","legend":"\u003cp\u003eDisplaying comprehensive performance metrics of the radiomics model utilizing 5-fold cross-validation.\u003c/p\u003e","description":"","filename":"Figure6.png","url":"https://assets-eu.researchsquare.com/files/rs-3831470/v1/f99a6a14c94c30c569f94e62.png"},{"id":49437025,"identity":"def3df40-21c8-4521-9822-3abb3fed42f6","added_by":"auto","created_at":"2024-01-10 20:17:51","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":57117,"visible":true,"origin":"","legend":"\u003cp\u003eIntroducing a groundbreaking hybrid nomogram, meticulously designed to provide an insightful scoring system for the evaluation of SPN-type IMA risk.\u003c/p\u003e","description":"","filename":"Figure7.png","url":"https://assets-eu.researchsquare.com/files/rs-3831470/v1/289000521a535a4b03322c4a.png"},{"id":49436693,"identity":"381542eb-e3c8-466f-83c7-d2fdf60d369d","added_by":"auto","created_at":"2024-01-10 20:09:51","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":412795,"visible":true,"origin":"","legend":"\u003cp\u003ePropelling the epitome of excellence, ROC curves meticulously crafted for the combined nomogram, clinical model, and radiomics approach in both the training (8a) and test sets (8b).\u003c/p\u003e","description":"","filename":"Figure8.png","url":"https://assets-eu.researchsquare.com/files/rs-3831470/v1/76679c7f4a350309ee8a565e.png"},{"id":49436687,"identity":"cf34f436-2ae0-43c1-957f-8f9639f73bae","added_by":"auto","created_at":"2024-01-10 20:09:51","extension":"png","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":322315,"visible":true,"origin":"","legend":"\u003cp\u003eShedding light on the extraordinary calibration of the combined nomogram, highlighting the flawless alignment between predicted risk and observed frequency, thereby reinforcing confidence in its precision, in both the training (8a) and test sets (8b).\u003c/p\u003e","description":"","filename":"Figure9.png","url":"https://assets-eu.researchsquare.com/files/rs-3831470/v1/62db876262f16001883848c6.png"},{"id":54496440,"identity":"1be8b280-781d-463c-8ce3-a4b6b790232c","added_by":"auto","created_at":"2024-04-11 11:45:01","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2246147,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-3831470/v1/57941c28-230a-4a6d-8ff7-599374a1e182.pdf"},{"id":49437237,"identity":"a833075e-805d-4c88-b2c2-7863a123b1d8","added_by":"auto","created_at":"2024-01-10 20:25:51","extension":"doc","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":56320,"visible":true,"origin":"","legend":"","description":"","filename":"TableS1.doc","url":"https://assets-eu.researchsquare.com/files/rs-3831470/v1/fcfed4a1643b8e4d84f6d72e.doc"}],"financialInterests":"No competing interests reported.","formattedTitle":"Radiomics Nomogram for Preoperative Differentiation Between Clinical Stage IA Solitary Pulmonary Nodule-Type Invasive Mucinous Adenocarcinoma and Invasive Non-mucinous Adenocarcinoma","fulltext":[{"header":"Background","content":"\u003cp\u003eLung cancer, particularly lung adenocarcinoma (LADC), is a significant global public health concern, accounting for \u0026gt;\u0026thinsp;40% of all cases [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]. Invasive mucinous adenocarcinoma (IMA) is a rare histological adenocarcinoma subtype, accounting for only 2\u0026ndash;5% of all LADCs. Despite its low prevalence, various studies have investigated the unique characteristics of IMA, including its distinct clinical, pathological, and genomic profiles, in comparison with those of invasive non-mucinous LADC [\u003cspan additionalcitationids=\"CR3 CR4\" citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. Notably, patients with IMA often exhibit specific gene mutations, such as anaplastic lymphoma kinase or Kirsten rat sarcoma viral oncogene [\u003cspan additionalcitationids=\"CR3 CR4\" citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. Histologically, IMAs are characterized by tumor cells exhibiting a goblet or columnar cell morphology, accompanied by abundant intracytoplasmic mucin. Moreover, patients diagnosed with IMA often exhibit the frequent occurrence of aerogenous dissemination and the presence of satellite tumors surrounding the primary lesion [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]. In 2015, recognizing these distinctive features, the World Health Organization officially designated IMA as a variant of invasive LADC [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. Furthermore, scientific investigations have revealed that IMA generally exhibits an inferior overall prognosis than non-mucinous LADC, even in the early stages [\u003cspan additionalcitationids=\"CR9\" citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. IMA has a distinctive feature within the spectrum of LADC; therefore, achieving a preoperative diagnosis is crucial for optimizing patient outcomes, particularly in early-stage cases.\u003c/p\u003e \u003cp\u003ePrevious studies investigated the imaging characteristics of IMA and categorized them into two main types based on computed tomography (CT) findings: solitary pulmonary nodule (SPN)-type and pneumonia-type IMA [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]. Nie et al. [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e] discovered that pneumonic-type IMA exhibited significantly poorer disease-free survival than SPN-type IMA. In addition, Wang et al. [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e] demonstrated that the prognosis of pneumonic-type IMA was significantly poorer than that of SPN-type IMA. Traditionally, mucinous bronchioloalveolar carcinomas have been linked with either a multifocal disease or a pattern resembling pneumonia on CT scans [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]. However, recent studies have shown that SPN-type IMA is more common than pneumonia-type IMA [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]. CT images currently provide limited information for differentiating between SPN-type IMA and non-mucinous LADC. Previous studies have suggested that certain CT features, including vacuolar signs, air bronchograms, and abnormal vascular changes, may help predict SPN-type IMA [\u003cspan additionalcitationids=\"CR17\" citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. However, interpreting these features can be subjective and lack consistency or typicality. Therefore, the accurate differentiation between IMA and invasive non-mucinous LADC based solely on preoperative CT images remains challenging.\u003c/p\u003e \u003cp\u003eRadiomics provides a modality to extract quantitative and high-throughput information from pulmonary images, capturing the intrinsic pathophysiology and offering valuable insights into tumor phenotypes [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e]. Yu et al. [\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e] developed a nomogram that combined clinical variables and CT-based radiomic features, showing good diagnostic performance to effectively differentiate pneumonia-type IMA from pneumonia. Similarly, Zhang et al. [\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e] established the diagnostic value of a radiomics-based nomogram to distinguish SPN-type IMA from pulmonary tuberculoma, demonstrating favorable diagnostic performance. However, radiomic applications that differentiate early-stage SPN-IMA from SPN-presenting LADC are currently lacking. Therefore, we aimed to develop a radiomics nomogram specifically for this purpose, which can help in clinical decision-making.\u003c/p\u003e"},{"header":"Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\n \u003ch2\u003ePatient selection\u003c/h2\u003e\n \u003cp\u003eFollowing the Declaration of Helsinki, ethical approval was obtained from the Ethics Committee (reference number: 2021-07-009) of the participating hospital, which ensured that the rights and interests of the research participants were not compromised. Notably, all methodologies were implemented in strict adherence to the applicable guidelines and regulations. The study participants were granted an exemption from the requirement of consent.\u003c/p\u003e\n \u003cp\u003eBetween March 2020 and October 2023, 935 surgically resected solitary LADCs were identified at Center 1 (Guilin Medical College Affiliated Hospital) and Center 2 (Xiangtan Central Hospital). The resected tumor glass slides were assessed by experienced pathologists [\u003cspan class=\"CitationRef\"\u003e18\u003c/span\u003e]. This study included 77 patients with SPN-type IMA and 569 with solitary invasive non-mucinous LADC.\u003c/p\u003e\n \u003cp\u003eThe patients\u0026apos; inclusion criteria were as follows: (1) histopathological confirmation of IMA, (2) solid nodules with a maximum diameter ranging from 0.5 cm to 3.0 cm, showing no cavities, calcification, vacuoles, and lacking ground glass density, (3) availability of complete thin-slice CT image data (0.625\u0026ndash;1.25 mm/slice) within a span of 2 weeks prior to the pathological diagnosis. Patients were excluded if they met any of the following criteria: (1) presence of multiple LADCs, (2) receiving antitumor therapy before undergoing CT examination and receiving the pathological diagnosis, (3) pathological diagnosis of noninvasive non-mucinous LADCs, or (4) demonstrating lymph node or distant metastases. It is worth noting that lymph node involvement and distant metastasis are less prevalent in IMA compared to other invasive non-mucinous LADCs [\u003cspan class=\"CitationRef\"\u003e6\u003c/span\u003e]. Ultimately, a total of 507 solitary pulmonary nodules were obtained, comprising 42 cases diagnosed as IMA and 465 diagnosed as invasive non-mucinous LADC. The division of training and test sets was randomly performed in a 7:3 ratio, without any deliberate partitioning, resulting in 354 cases allocated for the training set and 153 cases for the test set. Figure\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e provides a flowchart illustrating the patient inclusion process.\u003c/p\u003e\n \u003cp\u003e\u003cbr\u003e\u003c/p\u003e\n \u003cdiv id=\"Sec4\" class=\"Section3\"\u003e\n \u003ch2\u003eCT examinations\u003c/h2\u003e\n \u003cp\u003eTwo centers conducted unenhanced CT using a 64- or 128-detector row CT system. Center 1 employed the Revolution CT (GE Healthcare, Chicago, IL, USA) or MX16 CT (Philips Healthcare, Best, The Netherlands). Center 2 used the uCT550 or uCT760 systems (Shanghai United Imaging Healthcare, Shanghai, China). The scans implemented specific parameters, including a 120 kV tube voltage, tube current-time product ranging from 180 to 280 mA, beam pitch of 0.515 and 0.758, matrix size of 512 \u0026times; 512, and standard resolution algorithms.\u003c/p\u003e\n \u003c/div\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec5\" class=\"Section2\"\u003e\n \u003ch2\u003eCT radiological features evaluation\u003c/h2\u003e\n \u003cp\u003eFollowing CT scanning, the unprocessed data underwent transfer to a post-processing workstation to enable multiplanar reconstruction. Subsequently, analysis was performed using the lung window, characterized by a window width of 1200 HU and a level of -600 HU, as well as the mediastinal window with a window width of 400 HU and a level of 40 HU. The recorded and analyzed CT image features included size, density, shape, boundary, spiculation, lobulation, vascular convergence, and vacuole signs. Two board-certified thoracic radiologists (5 and 10 years of experience in chest CT imaging) independently analyzed the CT radiological features. The patients were kept unaware of the clinical and histological findings. Disagreements between the radiologists were resolved through consensus achieved during discussions to reconcile any discrepancies in the qualitative indicators.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec6\" class=\"Section2\"\u003e\n \u003ch2\u003ePreprocessing of CT images and tumor segmentation\u003c/h2\u003e\n \u003cp\u003eInitially, the CT images underwent standard resampling and grayscale discretization. Subsequently, utilizing ITK-SNAP software (version 4.0, \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ewww.itksnap.org\u003c/span\u003e\u003c/span\u003e), an adept thoracic radiologist with 5 years of experience meticulously delineated the tumor boundary slice by slice, thereby generating a precise volume of interest. Following this, a second radiologist, with a decade of experience in the same field, thoroughly reviewed the lesion delineations and made requisite adjustments as needed.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec7\" class=\"Section2\"\u003e\n \u003ch2\u003eRadiomics feature extraction and data preprocessing\u003c/h2\u003e\n \u003cp\u003eThe Pyradiomics function package (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://pyradiomics.readthedocs.io\u003c/span\u003e\u003c/span\u003e) was used to extract radiomic features from the CT images. This powerful package facilitates the extraction of a comprehensive set of 1239 radiomic features. CT images sourced from diverse hospitals and various protocols were included; therefore, all the radiomic features\u0026apos; intensities were effectively normalized using the z-score transformation (z = [x-\u0026micro;]/\u0026sigma;).\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e\n \u003ch2\u003eRadiomics feature selection and model construction\u003c/h2\u003e\n \u003cp\u003eWe employed the minimum redundancy maximum correlation (mRMR) approach to prevent overfitting by selecting the most pertinent features for tumor classification while minimizing redundancy. We used the least absolute shrinkage and selection operator (LASSO) regression model with 10-fold cross-validation in the feature selection phase to identify features with nonzero coefficients under the optimal \u0026lambda;. The selected features were then used to construct a radiomics model. The rad score was computed by summing the selected features, each weighted by its respective coefficient. Wilcoxon tests were conducted to discern disparities between the invasive non-mucinous LADC and IMA groups. These procedures were initially applied to the training set and were subsequently extended to the test set.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec9\" class=\"Section2\"\u003e\n \u003ch2\u003eClinical model and nomogram construction\u003c/h2\u003e\n \u003cp\u003eA multivariate logistic regression analysis was conducted in the training set using clinical and CT radiological features that exhibited a p-value less than 0.1 in the univariate logistic analysis. To establish the clinical prediction model, the best combinations of variables were selected through a backward stepwise selection process. The clinical model incorporated the predictive variables, and the radiomics model contributed the rad score, resulting in the creation of a combined prediction model and a corresponding nomogram. The predictive performance of each model was evaluated in both the training and test sets. The radiomics process, encompassing region of interest delineation, feature extraction, dimensionality reduction, feature selection, and model construction, is vividly illustrated in Fig.\u0026nbsp;2.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e\n \u003ch2\u003eStatistical analysis\u003c/h2\u003e\n \u003cp\u003eThe R software (version 4.3.1; \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.r-project.org\u003c/span\u003e\u003c/span\u003e) was employed for the statistical analysis. The Student\u0026apos;s t-test was applied to assess normally distributed continuous variables, while the Mann\u0026ndash;Whitney U test was utilized for non-normally distributed data. For categorical variables, the chi-square test was employed. Receiver operating characteristic analysis was conducted to compute the area under the curve (AUC), along with the corresponding sensitivity, specificity, and accuracy measures. DeLong\u0026apos;s test was employed to assess the statistical significance of differences between the AUC values. Statistical significance was established at a p-value\u0026thinsp;\u0026lt;\u0026thinsp;0.05.\u003c/p\u003e\n\u003c/div\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eBaseline characteristics\u003c/h2\u003e \u003cp\u003eIn total, 507 patients with SPNs were included in the study; 465 (91.7%) were diagnosed pathologically with invasive non-mucinous LADC and 42 (8.3%) with IMA. Among them, 376 and 162 patients were included in the training and test sets, respectively. The two cohorts showed no significant differences (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\u003eComparative Analysis of Radiological Features among Training, and Test set\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\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 \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVariables\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTotal (N\u0026thinsp;=\u0026thinsp;507)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eTraining Set (N\u0026thinsp;=\u0026thinsp;354)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eTest Set (N\u0026thinsp;=\u0026thinsp;153)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003ep-value\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGroup, N (%)\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\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.951\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eInvasive Non-Mucinous LADC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e465 (91.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e324 (91.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e141 (92.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIMA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e42 (8.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e30 (8.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e12 (7.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLocation, N (%)\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\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.253\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\u003e145 (28.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e105 (29.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e40 (26.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\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\u003e119 (23.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e81 (22.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e38 (24.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\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\u003e40 (7.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e22 (6.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e18 (11.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\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\u003e118 (23.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e86 (24.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e32 (20.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\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\u003e85 (16.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e60 (16.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e25 (16.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBoundary, N (%)\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\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.318\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIll-Defined\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e107 (21.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e70 (19.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e37 (24.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWell-Defined\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e400 (78.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e284 (80.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e116 (75.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eShape, N (%)\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\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.235\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\u003e273 (53.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e184 (52.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e89 (58.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOthers\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e234 (46.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e170 (48.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e64 (41.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLobulation, N (%)\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\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAbsence\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e120 (23.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e84 (23.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e36 (23.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePresence\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e387 (76.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e270 (76.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e117 (76.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSpiculation, N (%)\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\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.312\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAbsence\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e150 (29.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e110 (31.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e40 (26.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePresence\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e357 (70.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e244 (68.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e113 (73.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003eVascular Convergence Sign, N (%)\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 \u003cp\u003e0.821\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAbsence\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e264 (52.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e186 (52.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e78 (51.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePresence\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e243 (47.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e168 (47.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e75 (49.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVacuole Sign, N (%)\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\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.652\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAbsence\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e403 (79.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e279 (78.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e124 (81)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePresence\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e104 (20.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e75 (21.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e29 (19)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003ePleural Indentation, N (%)\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 \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAbsence\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e117 (23.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e82 (23.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e35 (22.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePresence\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e390 (76.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e272 (76.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e118 (77.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSex, N (%)\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\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.05\u003c/p\u003e \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\u003e244 (48.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e181 (51.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e63 (41.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\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\u003e263 (51.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e173 (48.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e90 (58.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge, Median (Q1, Q3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e61 (53, 67)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e61.5 (54, 67.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e60 (53, 67)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.539\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eClinical Stage, N (%)\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\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.579\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ecT1a\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e23 (4.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e18 (5.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e5 (3.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ecT1b\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e217 (42.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e148 (41.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e69 (45.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ecT1c\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e267 (52.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e188 (53.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e79 (51.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"5\" nameend=\"c5\" namest=\"c1\"\u003e \u003cp\u003e\u003cb\u003eAbbreviation\u003c/b\u003e: LUL Left Upper Lobe, LLL Left Lower Lobe, RUL Right Upper Lobe, RML Right Middle Lobe, RLL Right Lower Lobe, IMA Invasive Mucinous Adenocarcinoma, LADC Lung Adenocarcinoma\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\u003eTable \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e provides information on SPN-type IMA and invasive non-mucinous LADC. The training set showed no significant differences in demographic data, such as age and sex, but notable differences were found in CT radiological features. Specifically, SPN-type IMA exhibited a relatively higher proportion of vascular convergence signs, pleural indentation, and a higher clinical stage than invasive non-mucinous LADC (p\u0026thinsp;=\u0026thinsp;0.003, p\u0026thinsp;=\u0026thinsp;0.044, p\u0026thinsp;=\u0026thinsp;0.002).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eClinical prediction model\u003c/h2\u003e \u003cp\u003eUnivariate and multivariate logistic analyses were performed to identify several CT features, such as the vascular convergence sign (odds ratio [OR], 1.053 [0.995\u0026ndash;1.115], p\u0026thinsp;=\u0026thinsp;0.047) and clinical stage (OR, 1.135 [0.994\u0026ndash;1.296], p\u0026thinsp;=\u0026thinsp;0.041), as independent risk factors (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). These significant risk factors were used to select the most suitable combination of predictive variables for developing the clinical model. The AUC values for the training and test sets were 0.714 and 0.715, respectively.\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\u003eUnivariate and Multivariate Analysis for Discriminating Clinical Stage IA IMA from Invasive Non-Mucinous LADC\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"6\"\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 \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eVariables\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eUnivariate analysis\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003ep-value\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eMultivariate analysis\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003ep-value\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eOdd Ratio (95%CI)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eOdd Ratio (95%CI)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLocation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.76 (0.26\u0026ndash;2.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.61\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 \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBoundary\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.54 (0.24\u0026ndash;1.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.15\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 \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 \u003cp\u003e0.6 (0.28\u0026ndash;1.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.2\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 \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 \u003cp\u003e3 (0.89-10)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.077\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 \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSpiculation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2.4 (0.89\u0026ndash;6.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.083\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 \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVascular convergence sign\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2 (0.94\u0026ndash;4.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.043\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1.053(0.995\u0026ndash;1.115)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.047\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVacuole sign\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.1 (0.47\u0026ndash;2.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.76\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 \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePleural indentation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e4.6 (1.1\u0026ndash;20)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.04\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1.054(0.983\u0026ndash;1.129)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.139\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSex\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.91 (0.43\u0026ndash;1.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.8\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 \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\u003e0.97 (0.94-1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.14\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 \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eClinical Stage\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.1 (1.1\u0026ndash;1.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1.135(0.994\u0026ndash;1.296)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.041\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"6\" nameend=\"c6\" namest=\"c1\"\u003e \u003cp\u003e\u003cb\u003eAbbreviation\u003c/b\u003e: IMA invasive mucinous adenocarcinoma, LADC lung adenocarcinoma\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eRadiomics model\u003c/h2\u003e \u003cp\u003eFollowing the elimination of redundant and irrelevant features through the mRMR method, LASSO regression was employed to meticulously curate an optimized subset of features necessary for the construction of the ultimate model. A 10-fold cross-validation was subsequently executed to ascertain the optimal hyperparameter λ, as illustrated in Fig.\u0026nbsp;3. With an optimal λ value of 0.031, six features were discerningly chosen for the development of the radiomics models, as depicted in Fig.\u0026nbsp;4. Notably, the rad score for IMA exhibited a significant elevation compared to that of invasive non-mucinous LADC in both the training and test sets (all p\u0026thinsp;\u0026lt;\u0026thinsp;0.05), as elucidated in Fig.\u0026nbsp;5.\u003c/p\u003e \u003cp\u003eFor the radiomic model, AUC values of 0.770 and 0.753 were obtained for the training and test sets, respectively. Figure\u0026nbsp;6 shows the detailed performance of the radiomics model using 5-fold cross-validation for a more comprehensive evaluation. The radiomics model using the cross-validated analysis demonstrated favorable predictive performance, with a mean AUC value of 0.771, in differentiating IMA from invasive non-mucinous LADC.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003eEfficacy evaluation of combined models\u003c/h2\u003e \u003cp\u003eThe predictive variables utilized for constructing the combined model and the corresponding nomogram (Fig.\u0026nbsp;7) consisted of rad scores derived from the radiomics model and the radiological CT features integrated in the clinical model. The AUC values of the combined model were calculated to be 0.789 and 0.798 for the training and test sets, respectively (Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). Importantly, based on the outcomes of the DeLong test, it was demonstrated that the combined model outperformed the radiomics model in both the training (p\u0026thinsp;=\u0026thinsp;0.038) and test sets (p\u0026thinsp;=\u0026thinsp;0.021). In addition, when compared with the clinical model, the combined model exhibited superior performance in both the training (p\u0026thinsp;=\u0026thinsp;0.017) and test (p\u0026thinsp;=\u0026thinsp;0.025) sets. These comparisons are depicted in Fig.\u0026nbsp;8.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eDiagnostic Efficacy of the Combined Nomogram, Clinical model, and Radiomic Approaches in the Training and Validation Cohorts\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\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 \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePrediction models\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAUC\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\u003eSensitivity\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSpecificity\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTraining set\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\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCombined\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.789\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.653\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.833\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.636\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=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.714\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.525\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.833\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.497\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRadiomics\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.770\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.647\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.833\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.630\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTest set\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\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCombined\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.798\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.935\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.583\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.965\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=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.715\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.569\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.833\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.546\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRadiomics\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.753\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.686\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.750\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.681\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"5\" nameend=\"c5\" namest=\"c1\"\u003e \u003cp\u003e\u003cb\u003eAbbreviation\u003c/b\u003e: AUC, Area Under the Curve; ACC, Accuracy.\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\u003eIn addition, the calibration plots of the nomogram in the training and test sets revealed that the uncalibrated and calibration prediction curves were closely aligned with the diagonal and did not significantly deviate from the ideal curve, indicating that the nomogram had a good calibration (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e9\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eOur findings strongly indicate that integrating radiological CT features with radiomics models can significantly enhance the ability to differentiate between patients with IMA and those with invasive non-mucinous LADC, surpassing the diagnostic efficacy achieved using only radiological CT features or radiomics. Therefore, we successfully developed a combined nomogram with excellent discrimination and calibration to accurately distinguish between patients with SPN-type IMA and those with invasive non-mucinous LADC.\u003c/p\u003e \u003cp\u003eSPN-type IMA is the primary manifestation of IMA, and ground-glass nodules are rare [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e]. This manifestation distinction may be attributed to IMA\u0026rsquo;s histological characteristics, such as mucin-rich tumor cells, fibrosis, and central fibrosis with alveolar mucin-filled spaces [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]. Recent studies have indicated a higher prevalence of SPN-type IMA compared with pneumonia-type IMA [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]. Therefore, differentiating between SPN-type IMA and non-mucinous LADC is crucial, particularly in the early stages. This study found that SPN-type IMA exhibited more vascular convergence signs, pleural indentation, and advanced clinical stage than invasive non-mucinous LADC. These observations may be attributed to mucus leakage from the nodule's margin and the migration of macrophages along the alveolar walls and pores, leading to signs of vascular convergence and pleural indentation. These findings are consistent with those of Wu et al., Zhang et al., and Cha et al. [\u003cspan additionalcitationids=\"CR17\" citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. However, the clinical prediction model using CT radiological features demonstrated a modest diagnostic efficiency, as evidenced by the AUC values of 0.714 and 0.715 for the training and test sets, respectively. This could be due to the subjective interpretation of the radiological features, which may lack consistency, typicality, and generalizability, particularly during the early stages.\u003c/p\u003e \u003cp\u003eRadiomics is an advanced, noninvasive approach that uses intelligent algorithms to construct models based on original medical images. This method yields additional insights and potentially reveals pertinent phenotypic information by capturing tumor heterogeneity [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e]. To our knowledge, no study has investigated the use of radiomic models in distinguishing between clinical stage IA SPN-type IMA and non-mucinous LADC. Notably, our study revealed an independent association between higher rad scores and IMA in the radiomics model. This observation can be explained by the heightened IMA heterogeneity, attributed to the abundant presence of mucin within the cytoplasm of IMA tumor cells and the aggressive nature of tumor neovascularization, closely associated with its malignant behavior [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e]. The radiomics model achieved AUC values of 0.770 and 0.753 for the training and test sets, respectively, indicating higher diagnostic efficiency than that of the clinical model. This is because radiomics provides a notable advantage in evaluating tumor imaging phenotypes. It enables the extraction of quantitative features from CT images, allowing the identification of additional factors that cannot be easily comprehended visually using CT radiology. In addition, through five cross-validated analyses, the radiomics model demonstrated favorable predictive performance, with a mean AUC value of 0.771 in differentiating IMA from invasive non-mucinous LADC, indicating the model\u0026rsquo;s good robustness and avoidance of overfitting.\u003c/p\u003e \u003cp\u003eBy amalgamating crucial data from clinical and radiomics models, the combined model accurately distinguished between IMA and invasive non-mucinous LADC. It performed better than the clinical or radiomics models when used independently. We hypothesized that integrating diverse and abundant information within the combined model would increase diagnostic efficiency when distinguishing IMA from invasive non-mucinous LADC.\u003c/p\u003e \u003cp\u003eThis study has some limitations. First, it is crucial to acknowledge our study\u0026rsquo;s retrospective nature and the relatively low incidence rate of IMA stemming from the rarity of this condition. In addition, potential selection bias in our research arose from including only patients with postoperative pathologic results. Second, the nature of our bicentric study introduced variations in acquisition parameters, image quality, and potential co-registration errors, which could confound the analysis by contributing to the sources of variability. Furthermore, artifacts and technical limitations may affect the reliability and reproducibility of radiomic features. Third, we could not develop a predictive model for disease outcomes due to the relatively short postoperative follow-up period. Finally, our study focused solely on SPN-type IMA and invasive non-mucinous LADC with a diameter of \u0026le;\u0026thinsp;3 cm, which may limit the generalizability of our findings to other stages.\u003c/p\u003e \u003cp\u003eIn conclusion, our study successfully developed an innovative model using preoperative radiological and radiomic features to effectively differentiate between IMA and invasive non-mucinous LADC. We transformed this combined model into a nomogram that accurately quantified the risk of IMA. Notably, the nomogram exhibited exceptional discrimination and calibration, highlighting its potential value in clinical practice.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cp\u003eAUC, area under the curve\u003c/p\u003e\n\u003cp\u003eCT,\u0026nbsp;computed tomography\u003c/p\u003e\n\u003cp\u003eIMA,\u0026nbsp;invasive mucinous adenocarcinoma\u003c/p\u003e\n\u003cp\u003eLADC,\u0026nbsp;lung adenocarcinoma\u003c/p\u003e\n\u003cp\u003eLASSO,\u0026nbsp;least absolute shrinkage and selection operator\u003c/p\u003e\n\u003cp\u003emRMR,\u0026nbsp;minimum redundancy maximum correlation\u003c/p\u003e\n\u003cp\u003eSPN,\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003esolitary pulmonary nodule\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eEthical approval was obtained from Medical Ethics Committee of Xiangtan Central Hospital (reference number: No. 2021-07-009), which ensured that the rights and interests of the research participants were not compromised. All methods were performed following the relevant guidelines and regulations. Medical Ethics Committee of Xiangtan Central Hospital approved the retrospective study and waived the requirement for informed consent.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNA\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003eAvailability of data and materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe datasets generated during and analyzed during the current study are publicly available.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare that they have No competing financial interests exist.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNA\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003eAuthors\u0026apos; contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eS.H., W.G. and Y.W. performed the analyses and wrote the manuscript. Y.Z. and H.L. contributed to the conception of the paper. S.H. and W.G. contributed to analysis and prepared figures. Y.W prepared Tables. S.L. helped perform the analysis with constructive discussions. All authors contributed to the article and approved the submitted version.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003eAcknowledgments\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe would like to express our sincere gratitude to S.H., W.G., Y.W., Y.Z., H.L. and S.L. for their invaluable contributions to this study. We confirm that all individuals have given permission to be named in the Acknowledgments section.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eLee HY, Lee SW, Lee KS, Jeong JY, Choi JY, Kwon OJ, Song SH, Kim EY, Kim J, Shim YM: \u003cstrong\u003eRole of CT and PET Imaging in Predicting Tumor Recurrence and Survival in Patients with Lung Adenocarcinoma: A Comparison with the International Association for the Study of Lung Cancer/American Thoracic Society/European Respiratory Society Classification of Lung Adenocarcinoma.\u003c/strong\u003e J Thorac Oncol 2015, \u003cstrong\u003e10\u003c/strong\u003e(12):1785-1794.\u003c/li\u003e\n\u003cli\u003eShim HS, Kenudson M, Zheng Z, Liebers M, Cha YJ, Hoang Ho Q, Onozato M, Phi Le L, Heist RS, Iafrate AJ: \u003cstrong\u003eUnique Genetic and Survival Characteristics of Invasive Mucinous Adenocarcinoma of the Lung.\u003c/strong\u003e J Thorac Oncol 2015, \u003cstrong\u003e10\u003c/strong\u003e(8):1156-1162.\u003c/li\u003e\n\u003cli\u003eKadota K, Yeh YC, D\u0026apos;Angelo SP, Moreira AL, Kuk D, Sima CS, Riely GJ, Arcila ME, Kris MG, Rusch VW et al: \u003cstrong\u003eAssociations between mutations and histologic patterns of mucin in lung adenocarcinoma: invasive mucinous pattern and extracellular mucin are associated with KRAS mutation.\u003c/strong\u003e Am J Surg Pathol 2014, \u003cstrong\u003e38\u003c/strong\u003e(8):1118-1127.\u003c/li\u003e\n\u003cli\u003eGuo M, Tomoshige K, Meister M, Muley T, Fukazawa T, Tsuchiya T, Karns R, Warth A, Fink-Baldauf IM, Nagayasu T et al: \u003cstrong\u003eGene signature driving invasive mucinous adenocarcinoma of the lung.\u003c/strong\u003e EMBO Mol Med 2017, \u003cstrong\u003e9\u003c/strong\u003e(4):462-481.\u003c/li\u003e\n\u003cli\u003eShang G, Jin Y, Zheng Q, Shen X, Yang M, Li Y, Zhang L: \u003cstrong\u003eHistology and oncogenic driver alterations of lung adenocarcinoma in Chinese. \u003c/strong\u003eAm J Cancer Res 2019, \u003cstrong\u003e9\u003c/strong\u003e(6):1212-1223.\u003c/li\u003e\n\u003cli\u003eIchinokawa H, Ishii G, Nagai K, Yoshida J, Nishimura M, Hishida T, Suzuki K, Ochiai A: \u003cstrong\u003eClinicopathological characteristics of primary lung adenocarcinoma predominantly composed of goblet cells in surgically resected cases. \u003c/strong\u003ePathol Int 2011, \u003cstrong\u003e61\u003c/strong\u003e(7):423-429.\u003c/li\u003e\n\u003cli\u003eTravis WD, Brambilla E, Nicholson AG, Yatabe Y, Austin JHM, Beasley MB, Chirieac LR, Dacic S, Duhig E, Flieder DB et al:\u003cstrong\u003e The 2015 World Health Organization Classification of Lung Tumors: Impact of Genetic, Clinical and Radiologic Advances Since the 2004 Classification. \u003c/strong\u003eJ Thorac Oncol 2015, \u003cstrong\u003e10\u003c/strong\u003e(9):1243-1260.\u003c/li\u003e\n\u003cli\u003eLee HY, Cha MJ, Lee KS, Lee HY, Kwon OJ, Choi JY, Kim HK, Choi YS, Kim J, Shim YM: \u003cstrong\u003ePrognosis in Resected Invasive Mucinous Adenocarcinomas of the Lung: Related Factors and Comparison with Resected Nonmucinous Adenocarcinomas.\u003c/strong\u003e J Thorac Oncol 2016, \u003cstrong\u003e11\u003c/strong\u003e(7):1064-1073.\u003c/li\u003e\n\u003cli\u003eMatsui T, Sakakura N, Koyama S, Nakanishi K, Sasaki E, Kato S, Hosoda W, Murakami Y, Kuroda H, Yatabe Y: \u003cstrong\u003eComparison of Surgical Outcomes Between Invasive Mucinous and Non-Mucinous Lung Adenocarcinoma. \u003c/strong\u003eAnn Thorac Surg 2021, \u003cstrong\u003e112\u003c/strong\u003e(4):1118-1126.\u003c/li\u003e\n\u003cli\u003eXu X, Shen W, Wang D, Li N, Huang Z, Sheng J, Rucker AJ, Mao W, Xu H, Cheng G: \u003cstrong\u003eClinical features and prognosis of resectable pulmonary primary invasive mucinous adenocarcinoma.\u003c/strong\u003e Transl Lung Cancer Res 2022,\u003cstrong\u003e 11\u003c/strong\u003e(3):420-431.\u003c/li\u003e\n\u003cli\u003eBeck KS, Sung YE, Lee KY, Han DH: \u003cstrong\u003eInvasive mucinous adenocarcinoma of the lung: Serial CT findings, clinical features, and treatment and survival outcomes.\u003c/strong\u003e Thorac Cancer. 2020 , 11(12):3463-3472. \u003c/li\u003e\n\u003cli\u003eNie K, Nie W, Zhang YX, Yu H: \u003cstrong\u003eComparing clinicopathological features and prognosis of primary pulmonary invasive mucinous adenocarcinoma based on computed tomography findings.\u003c/strong\u003e Cancer Imaging 2019, \u003cstrong\u003e19\u003c/strong\u003e(1):47.\u003c/li\u003e\n\u003cli\u003eWang T, Yang Y, Liu X, Deng J, Wu J, Hou L, Wu C, She Y, Sun X, Xie D et al: \u003cstrong\u003ePrimary Invasive Mucinous Adenocarcinoma of the Lung: Prognostic Value of CT Imaging Features Combined with Clinical Factors.\u003c/strong\u003e Korean J Radiol 2021,\u003cstrong\u003e 22\u003c/strong\u003e(4):652-662.\u003c/li\u003e\n\u003cli\u003eCasali C, Rossi G, Marchioni A, Sartori G, Maselli F, Longo L, Tallarico E, Morandi U: \u003cstrong\u003eA single institution-based retrospective study of surgically treated bronchioloalveolar adenocarcinoma of the lung: clinicopathologic analysis, molecular features, and possible pitfalls in routine practice. \u003c/strong\u003eJ Thorac Oncol 2010, \u003cstrong\u003e5\u003c/strong\u003e(6):830-836.\u003c/li\u003e\n\u003cli\u003eWatanabe H, Saito H, Yokose T, Sakuma Y, Murakami S, Kondo T, Oshita F, Ito H, Nakayama H, Yamada K et al: \u003cstrong\u003eRelation between thin-section computed tomography and clinical findings of mucinous adenocarcinoma.\u003c/strong\u003e Ann Thorac Surg 2015,\u003cstrong\u003e 99\u003c/strong\u003e(3):975-981.\u003c/li\u003e\n\u003cli\u003eWu J, Zhang T, Pan J, Zhang Q, Lin X, Chang L, Chen YC, Xue X: \u003cstrong\u003eCharacteristics of the Computed Tomography Imaging Findings in 72 Patients with Airway-Invasive Pulmonary Aspergillosis. \u003c/strong\u003eMed Sci Monit 2021, \u003cstrong\u003e27\u003c/strong\u003e:e931162.\u003c/li\u003e\n\u003cli\u003eZhang X, Qiao W, Kang Z, Pan C, Chen Y, Li K, Shen W, Zhang L: \u003cstrong\u003eCT Features of Stage IA Invasive Mucinous Adenocarcinoma of the Lung and Establishment of a Prediction Model. \u003c/strong\u003eInt J Gen Med 2022,\u003cstrong\u003e 15\u003c/strong\u003e:5455-5463.\u003c/li\u003e\n\u003cli\u003eCha MJ, Lee KS, Kim TJ, Kim HS, Kim TS, Chung MJ, Kim BT, Kim YS: \u003cstrong\u003eSolitary Nodular Invasive Mucinous Adenocarcinoma of the Lung: Imaging Diagnosis Using the Morphologic-Metabolic Dissociation Sign. \u003c/strong\u003eKorean J Radiol 2019, \u003cstrong\u003e20\u003c/strong\u003e(3):513-521.\u003c/li\u003e\n\u003cli\u003eTunali I, Gillies RJ, Schabath MB: \u003cstrong\u003eApplication of Radiomics and Artificial Intelligence for Lung Cancer Precision Medicine.\u003c/strong\u003e Cold Spring Harb Perspect Med 2021, \u003cstrong\u003e11\u003c/strong\u003e(8).\u003c/li\u003e\n\u003cli\u003eYu X, Zhang S, Xu J, Huang Y, Luo H, Huang C, Nie P, Deng Y, Mao N, Zhang R et al: Nomogram Using CT Radiomics Features for Differentiation of Pneumonia-Type \u003cstrong\u003eInvasive Mucinous Adenocarcinoma and Pneumonia: Multicenter Development and External Validation Study.\u003c/strong\u003e AJR Am J Roentgenol 2023, \u003cstrong\u003e220\u003c/strong\u003e(2):224-234.\u003c/li\u003e\n\u003cli\u003eZhang J, Hao L, Qi M, Xu Q, Zhang N, Feng H, Shi G: \u003cstrong\u003eRadiomics nomogram for preoperative differentiation of pulmonary mucinous adenocarcinoma from tuberculoma in solitary pulmonary solid nodules. \u003c/strong\u003eBMC Cancer 2023, \u003cstrong\u003e23\u003c/strong\u003e(1):261.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"adenocarcinoma, nomogram, radiomics, mucinous","lastPublishedDoi":"10.21203/rs.3.rs-3831470/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-3831470/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e \u003cp\u003eRadiomic applications for differentiating clinical stage IA solitary pulmonary nodule (SPN)-type invasive mucinous adenocarcinoma (IMA) from SPN-presenting lung adenocarcinoma (LADC) are lacking. Therefore, this study aimed to develop and validate predictive models for the preoperative differentiation between SPN-IMA and invasive non-mucinous LADC using computed tomography (CT) radiological and radiomic features.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e \u003cp\u003eIn this bicentric study, we collected 507 SPNs, of which 42 were diagnosed as IMA and 465 as invasive non-mucinous LADC. The patients were randomly divided into training and test sets at a ratio of 7:3. The minimal redundancy maximal relevance filter was used to extract radiomic features, and the least absolute shrinkage and selection operator regression was used to screen these features and calculate the individualized radiomic score (rad score). We constructed a prediction nomogram that integrated radiomics and CT radiological features by applying multivariate logistic regression. Diagnostic capabilities were assessed by comparing the receiver operating characteristic and area under the curve (AUC) values.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eThe combined model achieved AUC values of 0.789 and 0.798 for the training and test sets, respectively, surpassing those of the radiomics model in both the training (p\u0026thinsp;=\u0026thinsp;0.038) and test (p\u0026thinsp;=\u0026thinsp;0.021) sets. Moreover, the combined model performed better than the clinical model in the training (p\u0026thinsp;=\u0026thinsp;0.017) and test (p\u0026thinsp;=\u0026thinsp;0.025) sets. We transformed this combined model into a nomogram that accurately quantifies the risk of IMA and demonstrates exceptional discrimination and calibration.\u003c/p\u003e\u003ch2\u003eConclusions\u003c/h2\u003e \u003cp\u003eThe combined nomogram, incorporating radiomics and CT radiological features, is potentially valuable for the preoperative differentiation between clinical stage IA SPN-type IMA and invasive non-mucinous LADC.\u003c/p\u003e","manuscriptTitle":"Radiomics Nomogram for Preoperative Differentiation Between Clinical Stage IA Solitary Pulmonary Nodule-Type Invasive Mucinous Adenocarcinoma and Invasive Non-mucinous Adenocarcinoma","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-01-10 20:09:46","doi":"10.21203/rs.3.rs-3831470/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"00d11cc4-647d-47dd-82ae-a995c3cf2278","owner":[],"postedDate":"January 10th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2024-04-11T11:36:53+00:00","versionOfRecord":[],"versionCreatedAt":"2024-01-10 20:09:46","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-3831470","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-3831470","identity":"rs-3831470","version":["v1"]},"buildId":"7rjqhiLT3MXkJMwkYKINL","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}