Predictive model for differentiating malignant and benign small pulmonary nodules

Predictive model for differentiating malignant and benign small pulmonary nodules | 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 Predictive model for differentiating malignant and benign small pulmonary nodules Ming-Ze Li, Su-Qin Li, Yi-Bing Shi, Yao-Yao Wang, Tao Meng This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7347617/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 5 You are reading this latest preprint version Abstract Background Current evaluation strategies for pulmonary nodules (PNs) emphasize size-based risk stratification. Predictive models are increasingly utilized in the preoperative assessment of PNs. While several models have been proposed to predict malignancy in PNs, few are specifically tailored to small pulmonary nodules (SPNs, ≤ 20 mm). This study entailed the development and validation of a predictive model specifically for differentiating malignant from benign SPNs. Methods A retrospective study was conducted using patient data from two medical centers between January 2019 and December 2024. Patients were separated into training and testing cohorts based on the hospital of origin. All patients underwent computed tomography (CT) and positron emission tomography (PET)/CT examinations. The predictive model was constructed using the training cohort and subsequently validated with the testing cohort. Model performance was also compared against that of an existing SPN predictive model. Results The training set comprised 155 patients (111 malignant and 44 benign), while the testing set included 82 patients (64 malignant and 18 benign). Logistic regression analysis identified lobulation (P = 0.015) and elevated standardized maximum uptake value (SUV max , P = 0.015) as independent predictors of malignancy. The final prediction model was defined as: X = − 4.446 + 1.305 × lobulation (present = 1, absent = 0) + 0.327 × SUV max . In the training cohort, the area under the curve (AUC) for the new model was 0.821, compared to 0.715 for the existing model. In the testing cohort, AUCs were 0.853 and 0.701, respectively. Calibration curves demonstrated strong concordance between predicted and actual malignancy probabilities. Conclusions The proposed predictive model demonstrated high diagnostic accuracy differentiating malignant from benign SPNs and may help to reduce the unnecessary biopsies. Predictive Model PET/CT Small Pulmonary nodule Solid Figures Figure 1 Figure 2 Figure 3 Figure 4 Introduction Pulmonary nodules (PNs) are defined as focal, rounded opacities within the lung parenchyma, measuring ≤ 3 cm in diameter [ 1 – 3 ]. Clinical guidelines recommend computed tomography (CT) follow-up for nodules larger than 6 mm [ 4 ], and tissue sampling for nodules ≥ 8 mm that exhibit high-risk features [ 4 ]. Common diagnostic procedures include biopsy and surgery [ 5 – 7 ]. Despite their minimally invasive nature, these interventions require careful preoperative evaluation to minimize unnecessary procedures. Predictive models are increasingly utilized in the preoperative assessment of PNs [ 8 ]. These models integrate clinical, radiological, and laboratory data to provide key advantages that include comprehensive risk profiling, risk coefficients for individual predictive variables, and malignancy probability scores to facilitate clinical decision-making [ 8 ]. Nodule size plays a pivotal role in malignancy risk stratification [ 9 – 11 ], underscoring the need for diameter-based assessment approaches. Most of the existing models were established based on the PNs ≤ 30 mm [ 8 ]. Zhao et al. [ 12 ] considered that the predictive models for PNs ≤ 30 mm may not suitable for the small PNs (SPNs, ≤ 20 mm) because the malignant/benign ratios among all PNs and SPNs may be different. Therefore, the specific predictive model for SPNs may provide more accurate diagnoses for SPNs than the existing models. While several models have been proposed to predict malignancy in PNs [ 8 ], few are specifically tailored to SPNs [ 9 , 12 , 13 ]. This study aimed to construct and externally validate a logistic regression-based model designed to differentiate malignant from benign SPNs, with the goal of enhancing early and accurate diagnosis. Methods Study design This retrospective study was conducted at two institutions: Xuzhou Central Hospital and Xuzhou First People's Hospital. Ethical approval was obtained from the institutional review boards of both centers, with a waiver for written informed consent. From January 2019 through December 2024, a total of 200 patients with SPNs were enrolled. Of these, 155 patients from Xuzhou Central Hospital were assigned to the training cohort, whereas 82 patients from Xuzhou First People's Hospital affiliated to China University of Mining and Technology constituted the testing cohort. Inclusion criteria were as follows: (a) presence of SPNs with a diameter ≤ 20 mm; (b) nodules with solid density on CT; (c) completion of preoperative chest CT, positron emission tomography (PET)/CT, and serum tumor marker tests; (d) availability of a definitive pathological diagnosis; and (e) interval ≤ 14 days between imaging and pathological confirmation. Exclusion criteria included: (a) history of malignancy within the previous five years; (b) nodules < 6 mm in diameter; and (c) incomplete clinical or imaging data. CT imaging protocol Chest CT was performed using lung (level: −500 HU; width: 1200 HU) and mediastinal (level: 50 HU; width: 450 HU) window settings. Imaging parameters included a tube voltage of 120 kVp, current of 100–200 mAs, pitch of 0.75–1.5, and collimation of 0.625–1.25 mm. Image slices were reconstructed to 1.0–1.25 mm thickness using a medium-sharp (B50) algorithm. CT scans were independently reviewed by 2 board-certified chest radiologists blinded to clinical and pathological information. PET/CT imaging protocol PET/CT scans were conducted with the Gemini TF PET/CT system (Philips) at Xuzhou Central Hospital and the Biograph mCT-S64 system (Siemens) at Xuzhou First People's Hospital affiliated to China University of Mining and Technology. Patients fasted for more than 6 hours before intravenous injection of 3.7 MBq/kg 18 F-fluorodeoxyglucose (FDG). Scanning commenced 1 hour post-injection. CT acquisition parameters were as follows: for the Philips system, parameters included 120 kV, 300 mA, pitch 0.829, 64 × 0.625 mm collimation, 0.5 s rotation time, 5.0 mm slice thickness; for the Siemens system, parameters included 120 kV, 255 mA, pitch 0.8, 64 × 0.6 mm collimation, 0.5 s rotation time, 3.0 mm slice thickness. PET scans were conducted using a 3D acquisition mode from the skull base to mid-thigh, with a scan time of 1.5 minutes per bed position. Data were reconstructed into PET, CT, and fused PET/CT images using Philips EBW 3.0 or PowerNMDMS 5.5 software. The standardized maximum uptake value (SUV max ) of each SPN was automatically calculated. The SUV = Activity in tumor/Injected activity × weight. All types of PET/CT scanners used the same formula of SUV. Pathological confirmation All nodules underwent pathological confirmation via surgical resection or percutaneous biopsy. Malignant and specific benign diagnoses (e.g., benign tumors, fungal infections, or tuberculosis) were established through histopathological analysis [ 14 ]. Non-specific benign findings (e.g., fibrosis or inflammation) on biopsy were not considered conclusive. Predictive model development and validation The predictive model was developed using data from the training cohort. Potential risk factors for malignancy in SPNs were detected using logistic regression analysis. Based on the significant predictors identified, both a logistic regression model and a corresponding nomogram were constructed. Model performance was assessed with a receiver operating characteristic (ROC) curve analysis. External validation was conducted using independent testing cohort data. Additionally, model clinical utility was evaluated through decision curve analysis (DCA) in both the training and testing datasets. Statistical analyses Normally distributed continuous data were presented as mean with standard deviation, otherwise, medians with interquartile range (Q1; Q3) was used. Mean and median values were compared with the Student’s t and Mann–Whitney U tests. Categorical data were compared with chi-square or Fisher’s exact test. Variables yielding a P-value less than 0.1 in the univariate logistic regression were included into multivariate analysis to identify independent predictors of malignancy. ROC curves were generated, and the corresponding area under the curve (AUC) values were compared using the DeLong test. The inter-observer agreement was analyzed to assess the degree of agreement by referring the statistical values (poor: 0-0.2; fair: 0.21–0.4; moderate: 0.41–0.6; good: 0.61–0.8,; and excellent: 0.81–1). Significance was defined as a two-tailed P < 0.05. Results Baseline data The training dataset included 155 patients with pathologically confirmed SPNs, comprising 111 malignant and 44 benign nodules. The testing cohort consisted of 82 patients, with 64 malignant and 18 benign SPNs, all confirmed by histopathology. Detailed baseline characteristics, imaging findings, and tumor marker results for both cohorts are presented in Table 1 . The inter-observer agreement analyses were shown in supplementary Table 1. Table 1 Baseline data of the training group. Training group (n = 155) Testing group (n = 82) P inter-groups Malignant (n = 111) Benign (n = 44) P Malignant (n = 64) Benign (n = 18) P Gender 0.838 0.236 0.749 Male 56 23 29 11 Female 55 21 35 7 Age (y) 64.8 ± 10.4 60.1 ± 12.6 < 0.001 65.3 ± 11.3 61.9 ± 13.4 0.292 0.502 Smoking history 35 9 0.168 14 5 0.600 0.387 Malignant history 5 1 0.676 1 0 1.000 0.069 Image features Diameter (mm) 16.1 ± 4.0 12.7 ± 4.5 < 0.001 14.8 ± 3.8 13.1 ± 3.8 0.093 0.253 Lobulation 86 18 < 0.001 55 7 < 0.001 0.526 Spiculation 68 12 < 0.001 47 6 0.002 0.055 Pleural retraction sign 39 6 0.008 27 1 0.004 0.417 Vascular convergence 14 4 0.537 6 2 0.583 0.663 CT bronchus sign 28 2 0.003 18 1 0.045 0.490 Calcification 1 4 0.023 1 3 0.009 0.527 Enlarged lymph node 14 1 0.097 13 0 0.037 0.161 SUV max 3.4 (2.3; 4.9) 1.6 (0; 2.5) < 0.001 4.3 (2.0; 6.7) 1.2 (0; 2.7) < 0.001 0.274 Lobes 0.539 0.142 0.033 Upper 57 25 20 9 Non-upper 54 19 44 9 Sides 0.083 0.152 0.172 Left 46 25 26 4 Right 65 19 38 14 Tumor markers CEA (ug/L) 2.8 (1.6; 3.8) 1.6 (1.2; 2.5) 0.001 3.0 (1.6; 5.9) 2.1 (1.3; 3.1) 0.055 0.147 Cyfra21-1 (ng/ml) 2.1 (1.5; 2.9) 1.9 (1.4; 2.4) 0.367 2.3 (1.7; 3.0) 1.8 (1.2; 2.2) 0.035 0.450 SCC (ug/L) 1.2 (0.8; 1.8) 1.2 (1.1; 1.8) 0.860 1.2 (0.6; 1.9) 1.2 (0.8; 1.8) 0.996 0.944 NSE (ng/ml) 12.5 (11.3; 14.7) 11.2 (10.1; 12.1) < 0.001 10.6 (3.7; 13.5) 10.0 (3.5; 12.0) 0.836 0.003 CEA: Carcinoembryonic antigen; CT: Computed tomography; Cyfra21-1: cytokeratin 19 fragment antigen 21 − 1; NSE: Neuron-specific enolase; SCC: Squamous cell carcinoma antigen; SUV max : Standard maximum uptake value. Predictive model construction Risk factors for malignant SPNs were initially identified via univariate logistic regression analysis in the training cohort. Significant associations were found with older age (P = 0.02), larger nodule diameter (P < 0.001), lobulation (P < 0.001), spiculation (P < 0.001), pleural retraction (P = 0.011), higher maximum standardized uptake value (SUV max ) (P 10 mm, P = 0.082), location in the left lung (P = 0.085), elevated neuron-specific enolase (NSE, P = 0.007), and elevated carcinoembryonic antigen (CEA, P = 0.006). Conversely, calcification was negatively associated with malignancy (P = 0.034). In the subsequent multivariate analysis, lobulation (P = 0.015) and increased SUV max (P = 0.015) emerged as independent predictors of malignancy. A summary of the univariate and multivariate logistic regression analyses is provided in Table 2 . Table 2 Predictors of malignancy. Variables Univariate analysis Multivariate analysis Hazard ratio 95% CI P value Hazard ratio 95% CI P value Age 1.038 1.006–1.071 0.02 1.007 0.964–1.052 0.753 Gender 1.076 0.535–2.164 0.838 Smoking history 1.791 0.777–4.127 0.171 Malignant history 2.208 0.230-17.873 0.524 Diameter 1.202 1.102–1.310 < 0.001 1.115 0.990–1.256 0.072 Lobulation 4.969 2.352–10.498 < 0.001 3.689 1.282–10.615 0.015 Spiculation 4.217 1.961–9.067 < 0.001 2.491 0.848–7.318 0.097 Pleural retraction 3.431 1.333–8.827 0.011 1.343 0.381–4.729 0.646 Vascular convergence 1.443 0.448–4.653 0.539 SUV max 1.745 1.351–2.254 < 0.001 1.372 1.067–1.804 0.015 CT bronchus sign 7.084 1.610–31.177 0.01 1.789 0.336–9.524 0.495 Calcification 0.091 0.01–0.838 0.034 0.085 0.007–1.041 0.054 Enlarged lymph node 6.206 0.791–48.708 0.082 5.543 0.402–76.403 0.201 Left lung 0.538 0.265–1.090 0.085 0.655 0.247–1.734 0.395 Upper lobe 0.802 0.397–1.621 0.539 NSE 1.146 1.038–1.265 0.007 1.074 0.942–1.225 0.285 CEA 1.488 1.119–1.979 0.006 1.172 0.861–1.595 0.313 SCC 1.065 0.824–1.378 0.629 Cyfra21-1 1.201 0.894–1.614 0.223 CEA: Carcinoembryonic antigen; CT: Computed tomography; Cyfra21-1: cytokeratin 19 fragment antigen 21 − 1; NSE: Neuron-specific enolase; SCC: Squamous cell carcinoma antigen; SUV max : Standard maximum uptake value. Using these two significant predictors, a logistic regression model was developed with the following equation: X = − 4.446 + 1.305 × lobulation (1 if present, 0 if absent) + 0.327 × SUV max . A nomogram corresponding to this model was also created to facilitate clinical application (Fig. 1 ). Based on ROC analysis, a threshold score of − 3.0904 was selected to optimize diagnostic sensitivity (90.1%) and specificity (68.2%) with the maximum Youden index (0.583). The positive and negative predictive values were 87.7% and 73.2%, respectively. SPNs with scores ≥ − 3.0904 were inclined to malignant, while those with lower scores were inclined to benign. To benchmark the newly developed model, it was compared with a previously published SPN prediction model developed by Zhao et al. [ 12 ], which is defined as: X = − 10.111 + 0.129 × age + 1.214 × pleural retraction sign (1 if present, 0 if absent) + 0.985 × CT bronchus sign (1 if present, 0 if absent) + 0.21 × CEA level. In the training dataset, the AUC for our model was 0.821, significantly outperforming the Zhao model, which had an AUC of 0.715 (P = 0.025; Table 3 , Fig. 2 a). Table 3 Diagnostic performance of the predictive model. Training group (n = 155) Testing group (n = 82) AUC Sensitivity Specificity AUC Sensitivity Specificity Predictive model 0.821 0.901 0.682 0.853 0.844 0.778 Zhao’ s model 0.715 0.694 0.614 0.701 0.641 0.611 AUC: area under the curve; SUV max : Standard maximum uptake value. Model validation When applied to the independent testing cohort, our model maintained strong discriminatory performance, achieving an AUC of 0.853. The sensitivity, specificity, positive predictive value, and negative predictive value were 84.4%, 77.8%, 93.1%, and 58.3%, respectively. In comparison, the Zhao model yielded an AUC of 0.701 (Table 3 , Fig. 2 b). The difference had the statistical significance (P = 0.023). Evaluation of clinical utility Calibration curve analysis demonstrated good agreement between predicted probabilities and observed outcomes for both the training and testing cohorts (Fig. 3 ). Decision curve analysis further supported the clinical value of the developed model and its nomogram, revealing a net benefit across a range of threshold probabilities (Fig. 4 ). Discussion In this study, a predictive model capable of distinguishing between malignant and benign SPNs was developed and validated. The model incorporated one morphological feature observable on CT imaging—lobulation—and one quantitative imaging parameter—SUV max derived from PET/CT. A corresponding nomogram was also constructed to provide an individualized, pretest estimate of malignancy risk for each patient, demonstrating strong predictive performance. During the clinical workflow, lobulation and SUV max also are the common features of the SPNs, and our predictive model can be easily understood and it can play an important role during the diagnosis of SPNs. Notably, the model in this study utilized only two predictive variables, fewer than the four or five factors commonly used in previously reported models for pulmonary nodules [ 9 , 12 , 13 ]. Despite this, the model achieved AUC values exceeding 0.8 in both the training and testing cohorts, suggesting robust diagnostic capability. These findings indicate that even with a simplified structure, the model retains strong discriminative power for SPNs. A key contributor to the model's high accuracy is likely the inclusion of SUV max , a quantitative parameter obtained from PET/CT imaging. Malignant SPNs typically exhibit elevated glucose metabolism, resulting in increased uptake of 18 F-FDG relative to benign lesions [ 15 ]. A prior meta-analysis reported that PET/CT alone could distinguish benign from malignant pulmonary nodules with an AUC of 0.82 [ 16 ]. Similarly, Wang et al. [ 15 ] demonstrated that PET/CT achieved high diagnostic performance, with AUC values of 0.915 in the training group and 0.843 in the testing group, when applied to pulmonary nodules. However, the SUV max values may be affected by multiple factors, such as scanner type or glucose levels. In this study, the the PET/CT scanners were different types in the 2 centers. However, the pre-examination preparation (fasted for ≥ 6 hours) and concentration of 18 F-FDG (3.7 MBq/kg) were same for each patient. These factors may decrease the bias risk of SUV max values. Lobulation, a common CT morphological feature associated with malignancy, also contributed significantly to our model. Chen et al. [ 8 ] reported that lobulation alone yielded an AUC of 0.74, with a sensitivity of 57% and specificity of 80%. However, as a binary categorical variable, lobulation lacks the continuous granularity of SUV max , which may limit its standalone diagnostic utility. This underscores the importance of combining multiple imaging features—such as spiculation, pleural retraction, and calcification—for a more comprehensive assessment of malignancy risk. The spiculation and pleural retraction signs were not presented as significance in this study. These findings may be attributed to that both lung cancers and inflammatory lesions can exhibit spiculation and pleural retraction signs [ 12 ]. Although some researchers considered that short and long spiculation represented the malignant and inflammatory lesions [ 12 ], there are no strict criteria to define the length of spiculation. Therefore, the spiculation was usually presented as a binary parameter (presence or absence), which limited its diagnostic performance. Consistent with findings from previous SPN studies [ 12 , 13 ], our results showed that nodule diameter was not a significant predictor of malignancy in the context of SPNs. While She et al. [ 10 ] reported a 1.1-fold increase in malignancy risk for each 1 mm increase in diameter, many predictive models have focused on nodules ≤ 30 mm in size [ 8 ]. The influence of diameter may be more pronounced in larger nodules (20–30 mm), whereas our study specifically targeted SPNs—typically defined as nodules < 20 mm—where diameter may exert a lesser effect on malignancy prediction [ 12 , 13 ]. Although CEA is widely recognized as a tumor marker in lung cancer and has been included in several predictive models [ 8 ], it was not retained as a significant variable in our model. This may be due to two primary factors: (a) the median CEA levels in both malignant and benign groups fell within the normal reference range (0–5 ng/mL) because CEA is not a specific tumor marker for lung cancers, especially for the non-adenocarcinoma, thereby diminishing its discriminatory value in our cohort; and (b) CEA concentrations can be influenced by non-malignant factors such as age and smoking status [ 17 ], which may have contributed to elevated CEA levels even in patients with benign SPNs. To further contextualize this newly developed model’s performance, it was compared with a previously published SPN model developed by Zhao et al. [ 12 ]. Our model achieved a significantly higher AUC, which may be attributable to the integration of PET/CT imaging. Unlike Zhao’s model, which did not incorporate PET/CT data, our model leveraged SUV max as a continuous variable rather than a categorical classification of FDG uptake, as seen in other studies [ 18 , 19 ]. This continuous representation likely offers a more precise reflection of metabolic activity, enhancing diagnostic performance. Machine learning-based models (e.g., radiomics or deep learning) also have been used for predict the malignant or benign SPNs and the AUC of such model can reach to more than 0.9 [ 9 ]. However, for radiomics signatures, there usually lacks the reproducibility and standardizability across different centers [ 20 ]. Therefore, the radiomics has not been widely used in the clinic. Nevertheless, radiomics and artificial intelligence (AI) are the future direction and the studies regarding radiomics and AI are still needed. There are several limitations to this study. First, the retrospective design introduces the possibility of selection bias. Second, the pathological diagnoses were confirmed by surgical resection or percutaneous biopsy, the different pathological diagnostic methods may also cause the risk of bias. Third, the PET/CT systems used in the two participating centers were from different manufacturers, and the SUV max values may be influenced. Furthermore, the feature “lobulation” also may be associated with interobserver variability. Fourth, both of the training and testing cohorts were from the same city, and it may limit geographic diversity and applicability to broader populations. Further international validation should be performed in the future study. Fifty, this predictive model also had the risk of overfitting. Finally, imbalances in baseline characteristics between the training and testing cohorts could have influenced model performance. Nonetheless, both cohorts demonstrated similarly high AUC values (> 0.8), suggesting that the predictive model is stable and generalizable across datasets. Conclusion In summary, a predictive model that effectively differentiates between malignant and benign SPNs was developed in this study, achieving strong diagnostic accuracy in both internal and external validation cohorts. The model offers a simplified yet reliable approach for clinical risk stratification of SPNs and may help to reduce the unnecessary biopsies. However, further prospective validation with lager sample size is still needed. Abbreviations AI: artificial intelligence; AUC: area under the curve; CEA: carcinoembryonic antigen; CT: computed tomography; DCA: decision curve analysis; FDG: 18 F-fluorodeoxyglucose; PET: positron emission tomography; PN: pulmonary nodule; ROC: receiver operating characteristic; SPN: small PN; SUV max : standardized maximum uptake value. Declarations Ethics approval and consent to participate : This study was approved by Ethics Committee of Xuzhou Central Hospital and Xuzhou First People's Hospital. The need for written informed consent was waived by the Ethics Committee of Xuzhou Central Hospital and Xuzhou First People's Hospital. All methods were carried out in accordance with Declaration of Helsinki. Consent for publication : Not applicable. Clinical trial number: Not applicable. Availability of data and materials : The data that support the findings of this study are available from the corresponding author upon reasonable request. Competing interests : None. Funding: None. Authors’ contributions: YYW and TM designed this work. YBS, MZL, and TM collected the clinical data. SQL and MZL performed the statistical analyses. MZL wrote this article. Final manuscript was approved by all authors. Acknowledgments: None References de Margerie-Mellon C, Chassagnon G. Artificial intelligence: A critical review of applications for lung nodule and lung cancer. Diagn Interv Imaging. 2023;104(1):11-17. doi: 10.1016/j.diii.2022.11.007. Godoy MCB, Odisio EGLC, Truong MT, de Groot PM, Shroff GS, Erasmus JJ. Pulmonary Nodule Management in Lung Cancer Screening: A Pictorial Review of Lung-RADS Version 1.0. Radiol Clin North Am. 2018;56(3):353-363. doi: 10.1016/j.rcl.2018.01.003. Mankidy BJ, Mohammad G, Trinh K, Ayyappan AP, Huang Q, Bujarski S, et al. High risk lung nodule: A multidisciplinary approach to diagnosis and management. Respir Med. 2023;214:107277. doi: 10.1016/j.rmed.2023.107277. MacMahon H, Naidich DP, Goo JM, Lee KS, Leung ANC, Mayo JR, et al. Guidelines for Management of Incidental Pulmonary Nodules Detected on CT Images: From the Fleischner Society 2017. Radiology. 2017;284(1):228-243. doi: 10.1148/radiol.2017161659. Bourgouin PP, Rodriguez KJ, Fintelmann FJ. Image-Guided Percutaneous Lung Needle Biopsy: How we do it. Tech Vasc Interv Radiol. 2021;24(3):100770. doi: 10.1016/j.tvir.2021.100770. Azour L, Liu S, Washer SL, Moore WH. Percutaneous Transthoracic Lung Biopsy: Optimizing Yield and Mitigating Risk. J Comput Assist Tomogr. 2021;45(5):765-775. doi: 10.1097/RCT.0000000000001192. Salfity H, Tong BC. VATS and Minimally Invasive Resection in Early-Stage NSCLC. Semin Respir Crit Care Med. 2020;41(3):335-345. doi: 10.1055/s-0039-3401991. Chen G, Bai T, Wen LJ, Li Y. Predictive model for the probability of malignancy in solitary pulmonary nodules: a meta-analysis. J Cardiothorac Surg. 2022;17(1):102. doi: 10.1186/s13019-022-01859-x. Sun JX, Zhou XX, Yu YJ, Wei YM, Shi YB, Xu QS, et al. CT radiomics based model for differentiating malignant and benign small (≤20mm) solid pulmonary nodules. Front Oncol. 2025;15:1502932. doi: 10.3389/fonc.2025.1502932. She Y, Zhao L, Dai C, Ren Y, Jiang G, Xie H, et al. Development and validation of a nomogram to estimate the pretest probability of cancer in Chinese patients with solid solitary pulmonary nodules: A multi-institutional study. J Surg Oncol. 2017;116(6):756-762. doi: 10.1002/jso.24704. Li Y, Chen KZ, Wang J. Development and validation of a clinical prediction model to estimate the probability of malignancy in solitary pulmonary nodules in Chinese people. Clin Lung Cancer. 2011;12(5):313-9. doi: 10.1016/j.cllc.2011.06.005. Zhao HC, Xu QS, Shi YB, Ma XJ. Clinical-radiological predictive model in differential diagnosis of small (≤ 20 mm) solitary pulmonary nodules. BMC Pulm Med. 2021;21(1):281. doi: 10.1186/s12890-021-01651-y. Chen XB, Yan RY, Zhao K, Zhang DF, Li YJ, Wu L, et al. Nomogram For The Prediction Of Malignancy In Small (8-20 mm) Indeterminate Solid Solitary Pulmonary Nodules In Chinese Populations. Cancer Manag Res. 2019;11:9439-9448. doi: 10.2147/CMAR.S225739. Li GC, Fu YF, Cao W, Shi YB, Wang T. Computed tomography-guided percutaneous cutting needle biopsy for small (≤ 20 mm) lung nodules. Medicine (Baltimore). 2017;96(46):e8703. doi: 10.1097/MD.0000000000008703. Wang XZ, Wang JY, Meng T, Shi YB, Sun JJ. Non-malignant pathological results from CT-guided biopsy for pulmonary nodules: a predictive model for identifying false-negative results. J Cardiothorac Surg. 2024;19(1):386. doi: 10.1186/s13019-024-02898-2. Li Y, Shi YB, Hu CF. 18F-FDG PET/CT based model for predicting malignancy in pulmonary nodules: a meta-analysis. J Cardiothorac Surg. 2024;19(1):148. doi: 10.1186/s13019-024-02614-0. Li L, Guo C, Wan JL, Fan QS, Xu XL, Fu YF. The use of carcinoembryonic antigen levels to predict lung nodule malignancy: a meta-analysis. Acta Clin Belg. 2022;77(1):227-232. doi: 10.1080/17843286.2020.1797330. Xiang Y, Sun Y, Gao W, Han B, Chen Q, Ye X, et al. Establishment of a predicting model to evaluate the probability of malignancy or benign in patients with solid solitary pulmonary nodules. Zhonghua Yi Xue Za Zhi. 2016;96(17):1354-1358. doi: 10.3760/cma.j.issn.0376-2491.2016.17.011. Herder GJ, van Tinteren H, Golding RP, Kostense PJ, Comans EF, Smit EF, et al. Clinical prediction model to characterize pulmonary nodules: validation and added value of 18F-fluorodeoxyglucose positron emission tomography. Chest. 2005;128(4):2490-2496. doi: 10.1378/chest.128.4.2490. Teng F, Fu YF, Wu AL, Xian YT, Lin J, Han R, et al. Computed Tomography-Based Predictive Model for the Probability of Lymph Node Metastasis in Gastric Cancer: A Meta-analysis. J Comput Assist Tomogr. 2024;48(1):19-25. doi: 10.1097/RCT.0000000000001530. Additional Declarations No competing interests reported. Supplementary Files supplement.doc Cite Share Download PDF Status: Under Review Version 1 posted Reviewers invited by journal 25 Sep, 2025 Editor invited by journal 31 Aug, 2025 Editor assigned by journal 29 Aug, 2025 Submission checks completed at journal 29 Aug, 2025 First submitted to journal 11 Aug, 2025 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-7347617","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":524895058,"identity":"7296804d-c69e-46a6-ad08-c99c5d432676","order_by":0,"name":"Ming-Ze Li","email":"","orcid":"","institution":"Xuzhou First People's Hospital","correspondingAuthor":false,"prefix":"","firstName":"Ming-Ze","middleName":"","lastName":"Li","suffix":""},{"id":524895059,"identity":"1821c982-26f1-4e9c-8032-05d39c2e516b","order_by":1,"name":"Su-Qin Li","email":"","orcid":"","institution":"Xuzhou Central Hospital","correspondingAuthor":false,"prefix":"","firstName":"Su-Qin","middleName":"","lastName":"Li","suffix":""},{"id":524895060,"identity":"3aa945a0-3518-46c7-9c76-315d93407365","order_by":2,"name":"Yi-Bing Shi","email":"","orcid":"","institution":"Xuzhou Central Hospital","correspondingAuthor":false,"prefix":"","firstName":"Yi-Bing","middleName":"","lastName":"Shi","suffix":""},{"id":524895061,"identity":"c9af7937-41a4-4f53-8917-393f8674cb05","order_by":3,"name":"Yao-Yao Wang","email":"","orcid":"","institution":"Xuzhou Central Hospital","correspondingAuthor":false,"prefix":"","firstName":"Yao-Yao","middleName":"","lastName":"Wang","suffix":""},{"id":524895062,"identity":"dcc012db-c4e3-45bf-bd91-5f5ded745e62","order_by":4,"name":"Tao Meng","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA3UlEQVRIiWNgGAWjYBADAzYGBsYHCRU1RKo/ANHCbPDgzDEStAApNsmHLcyEVeu29x5+/aHGxphPuv1aRWIDGwN/e3cCXi1mZ86lWRw4lmbGJnOm7EbiDhkGiTNnN+DXciPHzOAA22EbNomctBuJZ9gYDCRyCWi5/wao5R9ES0FiGzMRWm7wGD842HbYjE0i/RgDcVrO5JgxnO1LMwbawiyRcOYYD2G/HD9j/KHim43h/BnpDz/+qKiR42/vxa8FCNgkIDSPAZgkpBwEmD9AaPYHxKgeBaNgFIyCEQgAU5BL3yJYapwAAAAASUVORK5CYII=","orcid":"","institution":"Xuzhou Central Hospital","correspondingAuthor":true,"prefix":"","firstName":"Tao","middleName":"","lastName":"Meng","suffix":""}],"badges":[],"createdAt":"2025-08-11 14:53:34","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7347617/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7347617/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":93062625,"identity":"000dafb7-47c9-4e23-aa03-50f65a001ae0","added_by":"auto","created_at":"2025-10-08 16:38:00","extension":"tif","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":99012,"visible":true,"origin":"","legend":"","description":"","filename":"Fig1.tif","url":"https://assets-eu.researchsquare.com/files/rs-7347617/v1/0bf10f416f4773cb032e98e1.tif"},{"id":93062570,"identity":"2e38dadf-8ef0-4027-92c6-049353bbeea9","added_by":"auto","created_at":"2025-10-08 16:37:59","extension":"tif","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":89696,"visible":true,"origin":"","legend":"","description":"","filename":"Fig2a.tif","url":"https://assets-eu.researchsquare.com/files/rs-7347617/v1/3172ee56ea56776d73654cfc.tif"},{"id":93065914,"identity":"2016c681-58ce-4e76-a1da-d10e4862e877","added_by":"auto","created_at":"2025-10-08 16:46:07","extension":"doc","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":108544,"visible":true,"origin":"","legend":"","description":"","filename":"article.doc","url":"https://assets-eu.researchsquare.com/files/rs-7347617/v1/b1f715646086df97fdfa160f.doc"},{"id":93062656,"identity":"01561300-9b5e-4a2e-a772-74abd009c392","added_by":"auto","created_at":"2025-10-08 16:38:01","extension":"doc","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":110592,"visible":true,"origin":"","legend":"","description":"","filename":"table.doc","url":"https://assets-eu.researchsquare.com/files/rs-7347617/v1/2f4841ff883df5733df873d3.doc"},{"id":93062906,"identity":"77a2ed22-0f4a-468e-a703-9ebd62048ca1","added_by":"auto","created_at":"2025-10-08 16:38:05","extension":"tif","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":2445236,"visible":true,"origin":"","legend":"","description":"","filename":"Fig2b.tif","url":"https://assets-eu.researchsquare.com/files/rs-7347617/v1/1f112ef7f061f41697999115.tif"},{"id":93063082,"identity":"7a24d258-3f48-46e7-ad2c-4da8428cf00c","added_by":"auto","created_at":"2025-10-08 16:38:07","extension":"tif","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":118100,"visible":true,"origin":"","legend":"","description":"","filename":"Fig3a.tif","url":"https://assets-eu.researchsquare.com/files/rs-7347617/v1/cc1370b0fb876ab9854e22cd.tif"},{"id":93063078,"identity":"bb2c9972-c64a-4dbe-9b6b-a7dc5719b4e0","added_by":"auto","created_at":"2025-10-08 16:38:07","extension":"tif","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1851740,"visible":true,"origin":"","legend":"","description":"","filename":"Fig3b.tif","url":"https://assets-eu.researchsquare.com/files/rs-7347617/v1/b28c06d67e217354e3567456.tif"},{"id":93065915,"identity":"7021ce27-5f90-4650-a9a2-d2d8bb6bfa07","added_by":"auto","created_at":"2025-10-08 16:46:08","extension":"tif","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":137540,"visible":true,"origin":"","legend":"","description":"","filename":"Fig4a.tif","url":"https://assets-eu.researchsquare.com/files/rs-7347617/v1/adc70c41a299155c13f89654.tif"},{"id":93063084,"identity":"d93b3196-07c9-460d-8db1-9950943ad082","added_by":"auto","created_at":"2025-10-08 16:38:07","extension":"tif","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1846432,"visible":true,"origin":"","legend":"","description":"","filename":"Fig4b.tif","url":"https://assets-eu.researchsquare.com/files/rs-7347617/v1/430fc3e9594bdce024c693a9.tif"},{"id":93062602,"identity":"e069e541-bcb9-4e3c-8808-c6a1c5194bdc","added_by":"auto","created_at":"2025-10-08 16:38:00","extension":"json","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":6832,"visible":true,"origin":"","legend":"","description":"","filename":"a490ac25eeab4100937750d813f43bcb.json","url":"https://assets-eu.researchsquare.com/files/rs-7347617/v1/b275e24b234a201d5fd3f164.json"},{"id":93062693,"identity":"3edac29d-125d-4ebb-b105-7d55a154d9cb","added_by":"auto","created_at":"2025-10-08 16:38:01","extension":"doc","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":29696,"visible":true,"origin":"","legend":"","description":"","filename":"supplement.doc","url":"https://assets-eu.researchsquare.com/files/rs-7347617/v1/a30b3c1d6f0fd40744373e01.doc"},{"id":93062630,"identity":"edbb6edf-84ef-4d84-9370-ef9524c53f10","added_by":"auto","created_at":"2025-10-08 16:38:01","extension":"xml","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":101068,"visible":true,"origin":"","legend":"","description":"","filename":"a490ac25eeab4100937750d813f43bcb1enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7347617/v1/f123834eb1b230488cb5e1a4.xml"},{"id":93062452,"identity":"0a7e5db2-0df9-4d0d-b13c-1e57278390c2","added_by":"auto","created_at":"2025-10-08 16:37:57","extension":"tif","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":99012,"visible":true,"origin":"","legend":"","description":"","filename":"Fig1.tif","url":"https://assets-eu.researchsquare.com/files/rs-7347617/v1/6e8844ff1ceefd18bbd7822c.tif"},{"id":93063354,"identity":"cfefc7ce-7796-4664-a75a-4fab8a0489a5","added_by":"auto","created_at":"2025-10-08 16:38:13","extension":"tif","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":89696,"visible":true,"origin":"","legend":"","description":"","filename":"Fig2a.tif","url":"https://assets-eu.researchsquare.com/files/rs-7347617/v1/6df8bd3505c7d0f191f2618e.tif"},{"id":93063232,"identity":"9c8c8b0e-b328-465c-83eb-cd69fb88702f","added_by":"auto","created_at":"2025-10-08 16:38:10","extension":"tif","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":2445236,"visible":true,"origin":"","legend":"","description":"","filename":"Fig2b.tif","url":"https://assets-eu.researchsquare.com/files/rs-7347617/v1/5517518bb0258a9ad28a0a2e.tif"},{"id":93062824,"identity":"13b8e554-b8e1-4480-b1f5-2381b711b228","added_by":"auto","created_at":"2025-10-08 16:38:04","extension":"tif","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":118100,"visible":true,"origin":"","legend":"","description":"","filename":"Fig3a.tif","url":"https://assets-eu.researchsquare.com/files/rs-7347617/v1/237269a2b037d8a5e1758448.tif"},{"id":93063061,"identity":"40dcfa81-c020-4f88-a6cc-3f81e856ab66","added_by":"auto","created_at":"2025-10-08 16:38:07","extension":"tif","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1851740,"visible":true,"origin":"","legend":"","description":"","filename":"Fig3b.tif","url":"https://assets-eu.researchsquare.com/files/rs-7347617/v1/5889fa805f416d40b5d45e0a.tif"},{"id":93062626,"identity":"c4afdd19-921c-484d-8d35-5c71de192161","added_by":"auto","created_at":"2025-10-08 16:38:01","extension":"tif","order_by":17,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":137540,"visible":true,"origin":"","legend":"","description":"","filename":"Fig4a.tif","url":"https://assets-eu.researchsquare.com/files/rs-7347617/v1/cdc667f219377d9697a6d5d4.tif"},{"id":93062558,"identity":"395b5410-eb95-4574-ac29-924a65ea6e87","added_by":"auto","created_at":"2025-10-08 16:37:59","extension":"tif","order_by":18,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1846432,"visible":true,"origin":"","legend":"","description":"","filename":"Fig4b.tif","url":"https://assets-eu.researchsquare.com/files/rs-7347617/v1/0d7a416a5a1381664e7ce4d4.tif"},{"id":93062977,"identity":"93a1db77-7dc1-42fb-9c47-b1c734f9d9f4","added_by":"auto","created_at":"2025-10-08 16:38:05","extension":"png","order_by":19,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":18314,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFig1.png","url":"https://assets-eu.researchsquare.com/files/rs-7347617/v1/ec03ef1e4845b8aa925c3a51.png"},{"id":93062634,"identity":"8d3dfc6a-6b43-4627-bbf9-d5be1b382761","added_by":"auto","created_at":"2025-10-08 16:38:01","extension":"png","order_by":20,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":24831,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFig2a.png","url":"https://assets-eu.researchsquare.com/files/rs-7347617/v1/8c5a2c2f84a1f1a37528809c.png"},{"id":93063111,"identity":"93a38b25-e925-47d9-a7ff-a80f13edcffe","added_by":"auto","created_at":"2025-10-08 16:38:08","extension":"png","order_by":21,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":23269,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFig2b.png","url":"https://assets-eu.researchsquare.com/files/rs-7347617/v1/3eafa43c4e0c53b50b77396a.png"},{"id":93062562,"identity":"fffd0ac5-f8a5-49ed-be30-ce41994ba744","added_by":"auto","created_at":"2025-10-08 16:37:59","extension":"png","order_by":22,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":20861,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFig3a.png","url":"https://assets-eu.researchsquare.com/files/rs-7347617/v1/485a517e2133fc86eaf0dec9.png"},{"id":93062535,"identity":"21392e37-d0cd-4552-af03-d3147fe6570d","added_by":"auto","created_at":"2025-10-08 16:37:59","extension":"png","order_by":23,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":21206,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFig3b.png","url":"https://assets-eu.researchsquare.com/files/rs-7347617/v1/3294719743a99dfb49d204dc.png"},{"id":93062580,"identity":"ff9fcad3-41fb-45e6-887e-505cee52bfd5","added_by":"auto","created_at":"2025-10-08 16:38:00","extension":"png","order_by":24,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":22804,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFig4a.png","url":"https://assets-eu.researchsquare.com/files/rs-7347617/v1/5962ef0e08096f266c886601.png"},{"id":93062449,"identity":"c5cf3d39-3baf-43cf-920f-ba8a3e2ad86e","added_by":"auto","created_at":"2025-10-08 16:37:57","extension":"png","order_by":25,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":28626,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFig4b.png","url":"https://assets-eu.researchsquare.com/files/rs-7347617/v1/35ec2741a2a49a488e84542c.png"},{"id":93062754,"identity":"69af7ba5-e85a-43a0-933b-fc08db286b03","added_by":"auto","created_at":"2025-10-08 16:38:03","extension":"xml","order_by":26,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":101263,"visible":true,"origin":"","legend":"","description":"","filename":"a490ac25eeab4100937750d813f43bcb1structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7347617/v1/bef8a7a651922d12868ec682.xml"},{"id":93062438,"identity":"936e6b7f-2807-478f-81a3-aba928b88991","added_by":"auto","created_at":"2025-10-08 16:37:56","extension":"html","order_by":27,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":108231,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7347617/v1/a3abbae8e4a7e8e3af614332.html"},{"id":93065916,"identity":"f4ef7d46-5722-4d5f-acea-b52efd83fe8b","added_by":"auto","created_at":"2025-10-08 16:46:08","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":79890,"visible":true,"origin":"","legend":"\u003cp\u003eThe nomogram of the predictive model. When assessing a SPN, we should find each point for the lobulation and SUV\u003csub\u003emax\u003c/sub\u003e, then we can calculate the total point of the SPN, finally, we can estimate the diagnostic possibility according to the total point.\u003c/p\u003e","description":"","filename":"Fig1.png","url":"https://assets-eu.researchsquare.com/files/rs-7347617/v1/84085ed29c278867e9509c19.png"},{"id":93062555,"identity":"26ec4d47-e9af-4e8e-9cdc-fa562bd80fbe","added_by":"auto","created_at":"2025-10-08 16:37:59","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":222158,"visible":true,"origin":"","legend":"\u003cp\u003eROC curves corresponding to the predictive model, SUV\u003csub\u003emax\u003c/sub\u003e, and lobulation sign in the (a) training and (b) testing groups.\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-7347617/v1/462f3aa75e8d2fda1e2943cd.png"},{"id":93063124,"identity":"3fffb6e1-0c15-419f-a28f-97d7e7e113d8","added_by":"auto","created_at":"2025-10-08 16:38:08","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":180634,"visible":true,"origin":"","legend":"\u003cp\u003eCalibration curves of predictive model in the (a) training (R2 = 0.369) and (b) testing (R2 = 0.355) groups.\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-7347617/v1/ffa91b36269839110cc4265b.png"},{"id":93062454,"identity":"9cea5d8d-8aa5-4983-ae45-125ebf1b880d","added_by":"auto","created_at":"2025-10-08 16:37:57","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":255428,"visible":true,"origin":"","legend":"\u003cp\u003eDecision curve analysis results of predictive model for the (a) training and (b) testing groups.\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-7347617/v1/3e44ebcd7913aa30fec52d78.png"},{"id":93066011,"identity":"12674b16-39de-4ff6-97f1-440bb35338ef","added_by":"auto","created_at":"2025-10-08 16:46:16","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1531405,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7347617/v1/ddb6e1c9-4fb2-4322-ac61-8dc51cd9e092.pdf"},{"id":93062613,"identity":"5ac4f5f0-5506-45c2-a77b-47f0d4e10c5a","added_by":"auto","created_at":"2025-10-08 16:38:00","extension":"doc","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":29696,"visible":true,"origin":"","legend":"","description":"","filename":"supplement.doc","url":"https://assets-eu.researchsquare.com/files/rs-7347617/v1/e6c9fd4f6d25092b2d269968.doc"}],"financialInterests":"No competing interests reported.","formattedTitle":"Predictive model for differentiating malignant and benign small pulmonary nodules","fulltext":[{"header":"Introduction","content":"\u003cp\u003ePulmonary nodules (PNs) are defined as focal, rounded opacities within the lung parenchyma, measuring\u0026thinsp;\u0026le;\u0026thinsp;3 cm in diameter [\u003cspan additionalcitationids=\"CR2\" citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. Clinical guidelines recommend computed tomography (CT) follow-up for nodules larger than 6 mm [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e], and tissue sampling for nodules\u0026thinsp;\u0026ge;\u0026thinsp;8 mm that exhibit high-risk features [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]. Common diagnostic procedures include biopsy and surgery [\u003cspan additionalcitationids=\"CR6\" citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. Despite their minimally invasive nature, these interventions require careful preoperative evaluation to minimize unnecessary procedures.\u003c/p\u003e\u003cp\u003ePredictive models are increasingly utilized in the preoperative assessment of PNs [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. These models integrate clinical, radiological, and laboratory data to provide key advantages that include comprehensive risk profiling, risk coefficients for individual predictive variables, and malignancy probability scores to facilitate clinical decision-making [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eNodule size plays a pivotal role in malignancy risk stratification [\u003cspan additionalcitationids=\"CR10\" citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e], underscoring the need for diameter-based assessment approaches. Most of the existing models were established based on the PNs\u0026thinsp;\u0026le;\u0026thinsp;30 mm [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. Zhao et al. [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e] considered that the predictive models for PNs\u0026thinsp;\u0026le;\u0026thinsp;30 mm may not suitable for the small PNs (SPNs, \u0026le; 20 mm) because the malignant/benign ratios among all PNs and SPNs may be different. Therefore, the specific predictive model for SPNs may provide more accurate diagnoses for SPNs than the existing models. While several models have been proposed to predict malignancy in PNs [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e], few are specifically tailored to SPNs [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eThis study aimed to construct and externally validate a logistic regression-based model designed to differentiate malignant from benign SPNs, with the goal of enhancing early and accurate diagnosis.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003eStudy design\u003c/p\u003e\u003cp\u003eThis retrospective study was conducted at two institutions: Xuzhou Central Hospital and Xuzhou First People's Hospital. Ethical approval was obtained from the institutional review boards of both centers, with a waiver for written informed consent.\u003c/p\u003e\u003cp\u003eFrom January 2019 through December 2024, a total of 200 patients with SPNs were enrolled. Of these, 155 patients from Xuzhou Central Hospital were assigned to the training cohort, whereas 82 patients from Xuzhou First People's Hospital affiliated to China University of Mining and Technology constituted the testing cohort. Inclusion criteria were as follows: (a) presence of SPNs with a diameter\u0026thinsp;\u0026le;\u0026thinsp;20 mm; (b) nodules with solid density on CT; (c) completion of preoperative chest CT, positron emission tomography (PET)/CT, and serum tumor marker tests; (d) availability of a definitive pathological diagnosis; and (e) interval\u0026thinsp;\u0026le;\u0026thinsp;14 days between imaging and pathological confirmation. Exclusion criteria included: (a) history of malignancy within the previous five years; (b) nodules\u0026thinsp;\u0026lt;\u0026thinsp;6 mm in diameter; and (c) incomplete clinical or imaging data.\u003c/p\u003e\u003cp\u003eCT imaging protocol\u003c/p\u003e\u003cp\u003eChest CT was performed using lung (level: \u0026minus;500 HU; width: 1200 HU) and mediastinal (level: 50 HU; width: 450 HU) window settings. Imaging parameters included a tube voltage of 120 kVp, current of 100\u0026ndash;200 mAs, pitch of 0.75\u0026ndash;1.5, and collimation of 0.625\u0026ndash;1.25 mm. Image slices were reconstructed to 1.0\u0026ndash;1.25 mm thickness using a medium-sharp (B50) algorithm. CT scans were independently reviewed by 2 board-certified chest radiologists blinded to clinical and pathological information.\u003c/p\u003e\u003cp\u003ePET/CT imaging protocol\u003c/p\u003e\u003cp\u003ePET/CT scans were conducted with the Gemini TF PET/CT system (Philips) at Xuzhou Central Hospital and the Biograph mCT-S64 system (Siemens) at Xuzhou First People's Hospital affiliated to China University of Mining and Technology. Patients fasted for more than 6 hours before intravenous injection of 3.7 MBq/kg \u003csup\u003e18\u003c/sup\u003eF-fluorodeoxyglucose (FDG). Scanning commenced 1 hour post-injection. CT acquisition parameters were as follows: for the Philips system, parameters included 120 kV, 300 mA, pitch 0.829, 64 \u0026times; 0.625 mm collimation, 0.5 s rotation time, 5.0 mm slice thickness; for the Siemens system, parameters included 120 kV, 255 mA, pitch 0.8, 64 \u0026times; 0.6 mm collimation, 0.5 s rotation time, 3.0 mm slice thickness. PET scans were conducted using a 3D acquisition mode from the skull base to mid-thigh, with a scan time of 1.5 minutes per bed position. Data were reconstructed into PET, CT, and fused PET/CT images using Philips EBW 3.0 or PowerNMDMS 5.5 software. The standardized maximum uptake value (SUV\u003csub\u003emax\u003c/sub\u003e) of each SPN was automatically calculated. The SUV\u0026thinsp;=\u0026thinsp;Activity in tumor/Injected activity \u0026times; weight. All types of PET/CT scanners used the same formula of SUV.\u003c/p\u003e\u003cp\u003ePathological confirmation\u003c/p\u003e\u003cp\u003eAll nodules underwent pathological confirmation via surgical resection or percutaneous biopsy. Malignant and specific benign diagnoses (e.g., benign tumors, fungal infections, or tuberculosis) were established through histopathological analysis [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]. Non-specific benign findings (e.g., fibrosis or inflammation) on biopsy were not considered conclusive.\u003c/p\u003e\u003cp\u003ePredictive model development and validation\u003c/p\u003e\u003cp\u003eThe predictive model was developed using data from the training cohort. Potential risk factors for malignancy in SPNs were detected using logistic regression analysis. Based on the significant predictors identified, both a logistic regression model and a corresponding nomogram were constructed. Model performance was assessed with a receiver operating characteristic (ROC) curve analysis. External validation was conducted using independent testing cohort data. Additionally, model clinical utility was evaluated through decision curve analysis (DCA) in both the training and testing datasets.\u003c/p\u003e\u003cp\u003eStatistical analyses\u003c/p\u003e\u003cp\u003eNormally distributed continuous data were presented as mean with standard deviation, otherwise, medians with interquartile range (Q1; Q3) was used. Mean and median values were compared with the Student\u0026rsquo;s t and Mann\u0026ndash;Whitney U tests. Categorical data were compared with chi-square or Fisher\u0026rsquo;s exact test. Variables yielding a P-value less than 0.1 in the univariate logistic regression were included into multivariate analysis to identify independent predictors of malignancy. ROC curves were generated, and the corresponding area under the curve (AUC) values were compared using the DeLong test. The inter-observer agreement was analyzed to assess the degree of agreement by referring the statistical values (poor: 0-0.2; fair: 0.21\u0026ndash;0.4; moderate: 0.41\u0026ndash;0.6; good: 0.61\u0026ndash;0.8,; and excellent: 0.81\u0026ndash;1). Significance was defined as a two-tailed P\u0026thinsp;\u0026lt;\u0026thinsp;0.05.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003eBaseline data\u003c/p\u003e\u003cp\u003eThe training dataset included 155 patients with pathologically confirmed SPNs, comprising 111 malignant and 44 benign nodules. The testing cohort consisted of 82 patients, with 64 malignant and 18 benign SPNs, all confirmed by histopathology. Detailed baseline characteristics, imaging findings, and tumor marker results for both cohorts are presented in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e. The inter-observer agreement analyses were shown in supplementary Table\u0026nbsp;1.\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\u003eBaseline data of the training group.\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"8\"\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=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colspan=\"3\" nameend=\"c4\" namest=\"c2\"\u003e\u003cp\u003eTraining group (n\u0026thinsp;=\u0026thinsp;155)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colspan=\"3\" nameend=\"c7\" namest=\"c5\"\u003e\u003cp\u003eTesting group (n\u0026thinsp;=\u0026thinsp;82)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c8\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003eP inter-groups\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eMalignant (n\u0026thinsp;=\u0026thinsp;111)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eBenign (n\u0026thinsp;=\u0026thinsp;44)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eP\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eMalignant (n\u0026thinsp;=\u0026thinsp;64)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003eBenign (n\u0026thinsp;=\u0026thinsp;18)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c7\"\u003e\u003cp\u003eP\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eGender\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.838\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.236\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e0.749\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\u003e56\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e23\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e29\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e11\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\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\u003e55\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e21\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e35\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAge (y)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e64.8\u0026thinsp;\u0026plusmn;\u0026thinsp;10.4\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e60.1\u0026thinsp;\u0026plusmn;\u0026thinsp;12.6\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e65.3\u0026thinsp;\u0026plusmn;\u0026thinsp;11.3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e61.9\u0026thinsp;\u0026plusmn;\u0026thinsp;13.4\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.292\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e0.502\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSmoking history\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e35\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e9\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.168\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e14\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e5\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.600\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e0.387\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMalignant history\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e5\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.676\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e1.000\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e0.069\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eImage features\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\u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eDiameter (mm)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e16.1\u0026thinsp;\u0026plusmn;\u0026thinsp;4.0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e12.7\u0026thinsp;\u0026plusmn;\u0026thinsp;4.5\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e14.8\u0026thinsp;\u0026plusmn;\u0026thinsp;3.8\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e13.1\u0026thinsp;\u0026plusmn;\u0026thinsp;3.8\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.093\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e0.253\u003c/p\u003e\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\u003e86\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e18\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e55\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e0.526\u003c/p\u003e\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\u003e68\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e12\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e47\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e6\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.002\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e0.055\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePleural retraction sign\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e39\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e6\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.008\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e27\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.004\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e0.417\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eVascular convergence\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e14\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e4\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.537\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e6\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.583\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e0.663\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCT bronchus sign\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e28\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.003\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e18\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.045\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e0.490\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCalcification\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e4\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.023\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.009\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e0.527\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eEnlarged lymph node\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e14\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.097\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e13\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.037\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e0.161\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSUV\u003csub\u003emax\u003c/sub\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e3.4 (2.3; 4.9)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1.6 (0; 2.5)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e4.3 (2.0; 6.7)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e1.2 (0; 2.7)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e0.274\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLobes\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=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.539\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.142\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e0.033\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eUpper\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e57\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e25\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e20\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e9\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNon-upper\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e54\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e19\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e44\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e9\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSides\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=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.083\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.152\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e0.172\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLeft\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e46\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e25\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e26\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e4\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eRight\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e65\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e19\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e38\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e14\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTumor markers\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\u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCEA (ug/L)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e2.8 (1.6; 3.8)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1.6 (1.2; 2.5)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e3.0 (1.6; 5.9)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e2.1 (1.3; 3.1)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.055\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e0.147\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCyfra21-1 (ng/ml)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e2.1 (1.5; 2.9)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1.9 (1.4; 2.4)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.367\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e2.3 (1.7; 3.0)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e1.8 (1.2; 2.2)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.035\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e0.450\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSCC (ug/L)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1.2 (0.8; 1.8)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1.2 (1.1; 1.8)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.860\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e1.2 (0.6; 1.9)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e1.2 (0.8; 1.8)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.996\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e0.944\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNSE (ng/ml)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e12.5 (11.3; 14.7)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e11.2 (10.1; 12.1)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e10.6 (3.7; 13.5)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e10.0 (3.5; 12.0)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.836\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e0.003\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003ctfoot\u003e\u003ctr\u003e\u003ctd colspan=\"8\"\u003eCEA: Carcinoembryonic antigen; CT: Computed tomography; Cyfra21-1: cytokeratin 19 fragment antigen 21\u0026thinsp;\u0026minus;\u0026thinsp;1; NSE: Neuron-specific enolase; SCC: Squamous cell carcinoma antigen; SUV\u003csub\u003emax\u003c/sub\u003e: Standard maximum uptake value.\u003c/td\u003e\u003c/tr\u003e\u003c/tfoot\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003ePredictive model construction\u003c/p\u003e\u003cp\u003eRisk factors for malignant SPNs were initially identified via univariate logistic regression analysis in the training cohort. Significant associations were found with older age (P\u0026thinsp;=\u0026thinsp;0.02), larger nodule diameter (P\u0026thinsp;\u0026lt;\u0026thinsp;0.001), lobulation (P\u0026thinsp;\u0026lt;\u0026thinsp;0.001), spiculation (P\u0026thinsp;\u0026lt;\u0026thinsp;0.001), pleural retraction (P\u0026thinsp;=\u0026thinsp;0.011), higher maximum standardized uptake value (SUV\u003csub\u003emax\u003c/sub\u003e) (P\u0026thinsp;\u0026lt;\u0026thinsp;0.001), presence of a CT bronchus sign (P\u0026thinsp;=\u0026thinsp;0.01), presence of enlarged lymph nodes (short-axis\u0026thinsp;\u0026gt;\u0026thinsp;10 mm, P\u0026thinsp;=\u0026thinsp;0.082), location in the left lung (P\u0026thinsp;=\u0026thinsp;0.085), elevated neuron-specific enolase (NSE, P\u0026thinsp;=\u0026thinsp;0.007), and elevated carcinoembryonic antigen (CEA, P\u0026thinsp;=\u0026thinsp;0.006). Conversely, calcification was negatively associated with malignancy (P\u0026thinsp;=\u0026thinsp;0.034). In the subsequent multivariate analysis, lobulation (P\u0026thinsp;=\u0026thinsp;0.015) and increased SUV\u003csub\u003emax\u003c/sub\u003e (P\u0026thinsp;=\u0026thinsp;0.015) emerged as independent predictors of malignancy. A summary of the univariate and multivariate logistic regression analyses is provided in Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003ePredictors of malignancy.\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"7\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\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\" colspan=\"3\" nameend=\"c4\" namest=\"c2\"\u003e\u003cp\u003eUnivariate analysis\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colspan=\"3\" nameend=\"c7\" namest=\"c5\"\u003e\u003cp\u003eMultivariate analysis\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eHazard ratio\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003e95% CI\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eP value\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eHazard ratio\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003e95% CI\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c7\"\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\u003eAge\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e1.038\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1.006\u0026ndash;1.071\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.02\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e1.007\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.964\u0026ndash;1.052\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.753\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eGender\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e1.076\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.535\u0026ndash;2.164\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.838\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSmoking history\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e1.791\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.777\u0026ndash;4.127\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.171\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMalignant history\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e2.208\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.230-17.873\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.524\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eDiameter\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e1.202\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1.102\u0026ndash;1.310\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e1.115\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.990\u0026ndash;1.256\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.072\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLobulation\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e4.969\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e2.352\u0026ndash;10.498\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e3.689\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e1.282\u0026ndash;10.615\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e\u003cb\u003e0.015\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSpiculation\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e4.217\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1.961\u0026ndash;9.067\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e2.491\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.848\u0026ndash;7.318\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.097\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePleural retraction\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e3.431\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1.333\u0026ndash;8.827\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.011\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e1.343\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.381\u0026ndash;4.729\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.646\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eVascular convergence\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e1.443\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.448\u0026ndash;4.653\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.539\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSUV\u003csub\u003emax\u003c/sub\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e1.745\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1.351\u0026ndash;2.254\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e1.372\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e1.067\u0026ndash;1.804\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e\u003cb\u003e0.015\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCT bronchus sign\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e7.084\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1.610\u0026ndash;31.177\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.01\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e1.789\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.336\u0026ndash;9.524\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.495\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCalcification\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.091\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.01\u0026ndash;0.838\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.034\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.085\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.007\u0026ndash;1.041\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.054\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eEnlarged lymph node\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e6.206\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.791\u0026ndash;48.708\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.082\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e5.543\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.402\u0026ndash;76.403\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.201\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLeft lung\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.538\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.265\u0026ndash;1.090\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.085\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.655\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.247\u0026ndash;1.734\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.395\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eUpper lobe\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.802\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.397\u0026ndash;1.621\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.539\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNSE\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e1.146\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1.038\u0026ndash;1.265\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.007\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e1.074\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.942\u0026ndash;1.225\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.285\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCEA\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e1.488\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1.119\u0026ndash;1.979\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.006\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e1.172\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.861\u0026ndash;1.595\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.313\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSCC\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e1.065\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.824\u0026ndash;1.378\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.629\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCyfra21-1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e1.201\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.894\u0026ndash;1.614\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.223\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003ctfoot\u003e\u003ctr\u003e\u003ctd colspan=\"7\"\u003eCEA: Carcinoembryonic antigen; CT: Computed tomography; Cyfra21-1: cytokeratin 19 fragment antigen 21\u0026thinsp;\u0026minus;\u0026thinsp;1; NSE: Neuron-specific enolase; SCC: Squamous cell carcinoma antigen; SUV\u003csub\u003emax\u003c/sub\u003e: Standard maximum uptake value.\u003c/td\u003e\u003c/tr\u003e\u003c/tfoot\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eUsing these two significant predictors, a logistic regression model was developed with the following equation: X\u0026thinsp;=\u0026thinsp;\u0026minus;\u0026thinsp;4.446\u0026thinsp;+\u0026thinsp;1.305 \u0026times; lobulation (1 if present, 0 if absent)\u0026thinsp;+\u0026thinsp;0.327 \u0026times; SUV\u003csub\u003emax\u003c/sub\u003e. A nomogram corresponding to this model was also created to facilitate clinical application (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Based on ROC analysis, a threshold score of \u0026minus;\u0026thinsp;3.0904 was selected to optimize diagnostic sensitivity (90.1%) and specificity (68.2%) with the maximum Youden index (0.583). The positive and negative predictive values were 87.7% and 73.2%, respectively. SPNs with scores\u0026thinsp;\u0026ge;\u0026thinsp;\u0026minus;\u0026thinsp;3.0904 were inclined to malignant, while those with lower scores were inclined to benign.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eTo benchmark the newly developed model, it was compared with a previously published SPN prediction model developed by Zhao et al. [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e], which is defined as: X\u0026thinsp;=\u0026thinsp;\u0026minus;\u0026thinsp;10.111\u0026thinsp;+\u0026thinsp;0.129 \u0026times; age\u0026thinsp;+\u0026thinsp;1.214 \u0026times; pleural retraction sign (1 if present, 0 if absent)\u0026thinsp;+\u0026thinsp;0.985 \u0026times; CT bronchus sign (1 if present, 0 if absent)\u0026thinsp;+\u0026thinsp;0.21 \u0026times; CEA level. In the training dataset, the AUC for our model was 0.821, significantly outperforming the Zhao model, which had an AUC of 0.715 (P\u0026thinsp;=\u0026thinsp;0.025; Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e, Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ea).\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 performance of the predictive model.\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"7\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colspan=\"3\" nameend=\"c4\" namest=\"c2\"\u003e\u003cp\u003eTraining group (n\u0026thinsp;=\u0026thinsp;155)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colspan=\"3\" nameend=\"c7\" namest=\"c5\"\u003e\u003cp\u003eTesting group (n\u0026thinsp;=\u0026thinsp;82)\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eAUC\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eSensitivity\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eSpecificity\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eAUC\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003eSensitivity\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c7\"\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\u003ePredictive model\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.821\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.901\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.682\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.853\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.844\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.778\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eZhao\u0026rsquo; s model\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.715\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.694\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.614\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.701\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.641\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.611\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003ctfoot\u003e\u003ctr\u003e\u003ctd colspan=\"7\"\u003eAUC: area under the curve; SUV\u003csub\u003emax\u003c/sub\u003e: Standard maximum uptake value.\u003c/td\u003e\u003c/tr\u003e\u003c/tfoot\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eModel validation\u003c/p\u003e\u003cp\u003eWhen applied to the independent testing cohort, our model maintained strong discriminatory performance, achieving an AUC of 0.853. The sensitivity, specificity, positive predictive value, and negative predictive value were 84.4%, 77.8%, 93.1%, and 58.3%, respectively. In comparison, the Zhao model yielded an AUC of 0.701 (Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e, Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eb). The difference had the statistical significance (P\u0026thinsp;=\u0026thinsp;0.023).\u003c/p\u003e\u003cp\u003eEvaluation of clinical utility\u003c/p\u003e\u003cp\u003eCalibration curve analysis demonstrated good agreement between predicted probabilities and observed outcomes for both the training and testing cohorts (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). Decision curve analysis further supported the clinical value of the developed model and its nomogram, revealing a net benefit across a range of threshold probabilities (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eIn this study, a predictive model capable of distinguishing between malignant and benign SPNs was developed and validated. The model incorporated one morphological feature observable on CT imaging\u0026mdash;lobulation\u0026mdash;and one quantitative imaging parameter\u0026mdash;SUV\u003csub\u003emax\u003c/sub\u003e derived from PET/CT. A corresponding nomogram was also constructed to provide an individualized, pretest estimate of malignancy risk for each patient, demonstrating strong predictive performance. During the clinical workflow, lobulation and SUV\u003csub\u003emax\u003c/sub\u003e also are the common features of the SPNs, and our predictive model can be easily understood and it can play an important role during the diagnosis of SPNs.\u003c/p\u003e\u003cp\u003eNotably, the model in this study utilized only two predictive variables, fewer than the four or five factors commonly used in previously reported models for pulmonary nodules [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]. Despite this, the model achieved AUC values exceeding 0.8 in both the training and testing cohorts, suggesting robust diagnostic capability. These findings indicate that even with a simplified structure, the model retains strong discriminative power for SPNs.\u003c/p\u003e\u003cp\u003eA key contributor to the model's high accuracy is likely the inclusion of SUV\u003csub\u003emax\u003c/sub\u003e, a quantitative parameter obtained from PET/CT imaging. Malignant SPNs typically exhibit elevated glucose metabolism, resulting in increased uptake of \u003csup\u003e18\u003c/sup\u003eF-FDG relative to benign lesions [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]. A prior meta-analysis reported that PET/CT alone could distinguish benign from malignant pulmonary nodules with an AUC of 0.82 [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]. Similarly, Wang et al. [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e] demonstrated that PET/CT achieved high diagnostic performance, with AUC values of 0.915 in the training group and 0.843 in the testing group, when applied to pulmonary nodules. However, the SUV\u003csub\u003emax\u003c/sub\u003e values may be affected by multiple factors, such as scanner type or glucose levels. In this study, the the PET/CT scanners were different types in the 2 centers. However, the pre-examination preparation (fasted for \u0026ge;\u0026thinsp;6 hours) and concentration of \u003csup\u003e18\u003c/sup\u003eF-FDG (3.7 MBq/kg) were same for each patient. These factors may decrease the bias risk of SUV\u003csub\u003emax\u003c/sub\u003e values.\u003c/p\u003e\u003cp\u003eLobulation, a common CT morphological feature associated with malignancy, also contributed significantly to our model. Chen et al. [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e] reported that lobulation alone yielded an AUC of 0.74, with a sensitivity of 57% and specificity of 80%. However, as a binary categorical variable, lobulation lacks the continuous granularity of SUV\u003csub\u003emax\u003c/sub\u003e, which may limit its standalone diagnostic utility. This underscores the importance of combining multiple imaging features\u0026mdash;such as spiculation, pleural retraction, and calcification\u0026mdash;for a more comprehensive assessment of malignancy risk.\u003c/p\u003e\u003cp\u003eThe spiculation and pleural retraction signs were not presented as significance in this study. These findings may be attributed to that both lung cancers and inflammatory lesions can exhibit spiculation and pleural retraction signs [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]. Although some researchers considered that short and long spiculation represented the malignant and inflammatory lesions [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e], there are no strict criteria to define the length of spiculation. Therefore, the spiculation was usually presented as a binary parameter (presence or absence), which limited its diagnostic performance.\u003c/p\u003e\u003cp\u003eConsistent with findings from previous SPN studies [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e], our results showed that nodule diameter was not a significant predictor of malignancy in the context of SPNs. While She et al. [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e] reported a 1.1-fold increase in malignancy risk for each 1 mm increase in diameter, many predictive models have focused on nodules\u0026thinsp;\u0026le;\u0026thinsp;30 mm in size [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. The influence of diameter may be more pronounced in larger nodules (20\u0026ndash;30 mm), whereas our study specifically targeted SPNs\u0026mdash;typically defined as nodules\u0026thinsp;\u0026lt;\u0026thinsp;20 mm\u0026mdash;where diameter may exert a lesser effect on malignancy prediction [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eAlthough CEA is widely recognized as a tumor marker in lung cancer and has been included in several predictive models [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e], it was not retained as a significant variable in our model. This may be due to two primary factors: (a) the median CEA levels in both malignant and benign groups fell within the normal reference range (0\u0026ndash;5 ng/mL) because CEA is not a specific tumor marker for lung cancers, especially for the non-adenocarcinoma, thereby diminishing its discriminatory value in our cohort; and (b) CEA concentrations can be influenced by non-malignant factors such as age and smoking status [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e], which may have contributed to elevated CEA levels even in patients with benign SPNs.\u003c/p\u003e\u003cp\u003eTo further contextualize this newly developed model\u0026rsquo;s performance, it was compared with a previously published SPN model developed by Zhao et al. [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]. Our model achieved a significantly higher AUC, which may be attributable to the integration of PET/CT imaging. Unlike Zhao\u0026rsquo;s model, which did not incorporate PET/CT data, our model leveraged SUV\u003csub\u003emax\u003c/sub\u003e as a continuous variable rather than a categorical classification of FDG uptake, as seen in other studies [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e, \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e]. This continuous representation likely offers a more precise reflection of metabolic activity, enhancing diagnostic performance.\u003c/p\u003e\u003cp\u003eMachine learning-based models (e.g., radiomics or deep learning) also have been used for predict the malignant or benign SPNs and the AUC of such model can reach to more than 0.9 [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]. However, for radiomics signatures, there usually lacks the reproducibility and standardizability across different centers [\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e]. Therefore, the radiomics has not been widely used in the clinic. Nevertheless, radiomics and artificial intelligence (AI) are the future direction and the studies regarding radiomics and AI are still needed.\u003c/p\u003e\u003cp\u003eThere are several limitations to this study. First, the retrospective design introduces the possibility of selection bias. Second, the pathological diagnoses were confirmed by surgical resection or percutaneous biopsy, the different pathological diagnostic methods may also cause the risk of bias. Third, the PET/CT systems used in the two participating centers were from different manufacturers, and the SUV\u003csub\u003emax\u003c/sub\u003e values may be influenced. Furthermore, the feature \u0026ldquo;lobulation\u0026rdquo; also may be associated with interobserver variability. Fourth, both of the training and testing cohorts were from the same city, and it may limit geographic diversity and applicability to broader populations. Further international validation should be performed in the future study. Fifty, this predictive model also had the risk of overfitting. Finally, imbalances in baseline characteristics between the training and testing cohorts could have influenced model performance. Nonetheless, both cohorts demonstrated similarly high AUC values (\u0026gt;\u0026thinsp;0.8), suggesting that the predictive model is stable and generalizable across datasets.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eIn summary, a predictive model that effectively differentiates between malignant and benign SPNs was developed in this study, achieving strong diagnostic accuracy in both internal and external validation cohorts. The model offers a simplified yet reliable approach for clinical risk stratification of SPNs and may help to reduce the unnecessary biopsies. However, further prospective validation with lager sample size is still needed.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cp\u003eAI: artificial intelligence;\u003c/p\u003e\n\u003cp\u003eAUC: area under the curve;\u003c/p\u003e\n\u003cp\u003eCEA:\u0026nbsp;carcinoembryonic antigen;\u003c/p\u003e\n\u003cp\u003eCT: computed tomography;\u003c/p\u003e\n\u003cp\u003eDCA:\u0026nbsp;decision curve analysis;\u003c/p\u003e\n\u003cp\u003eFDG: \u003csup\u003e18\u003c/sup\u003eF-fluorodeoxyglucose;\u003c/p\u003e\n\u003cp\u003ePET: positron emission tomography;\u003c/p\u003e\n\u003cp\u003ePN: pulmonary nodule;\u003c/p\u003e\n\u003cp\u003eROC:\u0026nbsp;receiver operating characteristic;\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eSPN: small PN;\u003c/p\u003e\n\u003cp\u003eSUV\u003csub\u003emax\u003c/sub\u003e: standardized maximum uptake value.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003cstrong\u003e:\u0026nbsp;\u003c/strong\u003eThis study was approved by Ethics Committee of Xuzhou Central Hospital and Xuzhou First People\u0026apos;s Hospital. The need for written informed consent was waived by the Ethics Committee of Xuzhou Central Hospital and Xuzhou First People\u0026apos;s Hospital. All methods were carried out in accordance with \u003cstrong\u003eDeclaration of Helsinki.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003cstrong\u003e:\u003c/strong\u003e Not applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eClinical trial number:\u003c/strong\u003e Not applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and materials\u003c/strong\u003e\u003cstrong\u003e:\u0026nbsp;\u003c/strong\u003eThe data that support the findings of this study are available from the corresponding author upon reasonable request.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003cstrong\u003e:\u003c/strong\u003e None.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding:\u003c/strong\u003e None.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors\u0026rsquo; contributions:\u003c/strong\u003e YYW and TM designed this work. YBS, MZL, and TM collected the clinical data. SQL and MZL performed the statistical analyses. MZL wrote this article. Final manuscript was approved by all authors.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgments:\u0026nbsp;\u003c/strong\u003eNone\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003ede Margerie-Mellon C, Chassagnon G. Artificial intelligence: A critical review of applications for lung nodule and lung cancer. Diagn Interv Imaging. 2023;104(1):11-17. doi: 10.1016/j.diii.2022.11.007. \u003c/li\u003e\n\u003cli\u003eGodoy MCB, Odisio EGLC, Truong MT, de Groot PM, Shroff GS, Erasmus JJ. Pulmonary Nodule Management in Lung Cancer Screening: A Pictorial Review of Lung-RADS Version 1.0. Radiol Clin North Am. 2018;56(3):353-363. doi: 10.1016/j.rcl.2018.01.003.\u003c/li\u003e\n\u003cli\u003eMankidy BJ, Mohammad G, Trinh K, Ayyappan AP, Huang Q, Bujarski S, et al. High risk lung nodule: A multidisciplinary approach to diagnosis and management. Respir Med. 2023;214:107277. doi: 10.1016/j.rmed.2023.107277.\u003c/li\u003e\n\u003cli\u003eMacMahon H, Naidich DP, Goo JM, Lee KS, Leung ANC, Mayo JR, et al. Guidelines for Management of Incidental Pulmonary Nodules Detected on CT Images: From the Fleischner Society 2017. Radiology. 2017;284(1):228-243. doi: 10.1148/radiol.2017161659.\u003c/li\u003e\n\u003cli\u003eBourgouin PP, Rodriguez KJ, Fintelmann FJ. Image-Guided Percutaneous Lung Needle Biopsy: How we do it. Tech Vasc Interv Radiol. 2021;24(3):100770. doi: 10.1016/j.tvir.2021.100770.\u003c/li\u003e\n\u003cli\u003eAzour L, Liu S, Washer SL, Moore WH. Percutaneous Transthoracic Lung Biopsy: Optimizing Yield and Mitigating Risk. J Comput Assist Tomogr. 2021;45(5):765-775. doi: 10.1097/RCT.0000000000001192.\u003c/li\u003e\n\u003cli\u003eSalfity H, Tong BC. VATS and Minimally Invasive Resection in Early-Stage NSCLC. Semin Respir Crit Care Med. 2020;41(3):335-345. doi: 10.1055/s-0039-3401991.\u003c/li\u003e\n\u003cli\u003eChen G, Bai T, Wen LJ, Li Y. Predictive model for the probability of malignancy in solitary pulmonary nodules: a meta-analysis. J Cardiothorac Surg. 2022;17(1):102. doi: 10.1186/s13019-022-01859-x.\u003c/li\u003e\n\u003cli\u003eSun JX, Zhou XX, Yu YJ, Wei YM, Shi YB, Xu QS, et al. CT radiomics based model for differentiating malignant and benign small (\u0026le;20mm) solid pulmonary nodules. Front Oncol. 2025;15:1502932. doi: 10.3389/fonc.2025.1502932.\u003c/li\u003e\n\u003cli\u003eShe Y, Zhao L, Dai C, Ren Y, Jiang G, Xie H, et al. Development and validation of a nomogram to estimate the pretest probability of cancer in Chinese patients with solid solitary pulmonary nodules: A multi-institutional study. J Surg Oncol. 2017;116(6):756-762. doi: 10.1002/jso.24704.\u003c/li\u003e\n\u003cli\u003eLi Y, Chen KZ, Wang J. Development and validation of a clinical prediction model to estimate the probability of malignancy in solitary pulmonary nodules in Chinese people. Clin Lung Cancer. 2011;12(5):313-9. doi: 10.1016/j.cllc.2011.06.005.\u003c/li\u003e\n\u003cli\u003eZhao HC, Xu QS, Shi YB, Ma XJ. Clinical-radiological predictive model in differential diagnosis of small (\u0026le;\u0026thinsp;20 mm) solitary pulmonary nodules. BMC Pulm Med. 2021;21(1):281. doi: 10.1186/s12890-021-01651-y.\u003c/li\u003e\n\u003cli\u003eChen XB, Yan RY, Zhao K, Zhang DF, Li YJ, Wu L, et al. Nomogram For The Prediction Of Malignancy In Small (8-20 mm) Indeterminate Solid Solitary Pulmonary Nodules In Chinese Populations. Cancer Manag Res. 2019;11:9439-9448. doi: 10.2147/CMAR.S225739.\u003c/li\u003e\n\u003cli\u003eLi GC, Fu YF, Cao W, Shi YB, Wang T. Computed tomography-guided percutaneous cutting needle biopsy for small (\u0026le; 20 mm) lung nodules. Medicine (Baltimore). 2017;96(46):e8703. doi: 10.1097/MD.0000000000008703.\u003c/li\u003e\n\u003cli\u003eWang XZ, Wang JY, Meng T, Shi YB, Sun JJ. Non-malignant pathological results from CT-guided biopsy for pulmonary nodules: a predictive model for identifying false-negative results. J Cardiothorac Surg. 2024;19(1):386. doi: 10.1186/s13019-024-02898-2.\u003c/li\u003e\n\u003cli\u003eLi Y, Shi YB, Hu CF. 18F-FDG PET/CT based model for predicting malignancy in pulmonary nodules: a meta-analysis. J Cardiothorac Surg. 2024;19(1):148. doi: 10.1186/s13019-024-02614-0.\u003c/li\u003e\n\u003cli\u003eLi L, Guo C, Wan JL, Fan QS, Xu XL, Fu YF. The use of carcinoembryonic antigen levels to predict lung nodule malignancy: a meta-analysis. Acta Clin Belg. 2022;77(1):227-232. doi: 10.1080/17843286.2020.1797330.\u003c/li\u003e\n\u003cli\u003eXiang Y, Sun Y, Gao W, Han B, Chen Q, Ye X, et al. Establishment of a predicting model to evaluate the probability of malignancy or benign in patients with solid solitary pulmonary nodules. Zhonghua Yi Xue Za Zhi. 2016;96(17):1354-1358. doi: 10.3760/cma.j.issn.0376-2491.2016.17.011.\u003c/li\u003e\n\u003cli\u003eHerder GJ, van Tinteren H, Golding RP, Kostense PJ, Comans EF, Smit EF, et al. Clinical prediction model to characterize pulmonary nodules: validation and added value of 18F-fluorodeoxyglucose positron emission tomography. Chest. 2005;128(4):2490-2496. doi: 10.1378/chest.128.4.2490.\u003c/li\u003e\n\u003cli\u003eTeng F, Fu YF, Wu AL, Xian YT, Lin J, Han R, et al. Computed Tomography-Based Predictive Model for the Probability of Lymph Node Metastasis in Gastric Cancer: A Meta-analysis. J Comput Assist Tomogr. 2024;48(1):19-25. doi: 10.1097/RCT.0000000000001530.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"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":"bmc-pulmonary-medicine","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"pulm","sideBox":"Learn more about [BMC Pulmonary Medicine](http://bmcpulmmed.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/pulm/default.aspx","title":"BMC Pulmonary Medicine","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Predictive, Model, PET/CT, Small, Pulmonary nodule, Solid","lastPublishedDoi":"10.21203/rs.3.rs-7347617/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7347617/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e\u003cp\u003eCurrent evaluation strategies for pulmonary nodules (PNs) emphasize size-based risk stratification. Predictive models are increasingly utilized in the preoperative assessment of PNs. While several models have been proposed to predict malignancy in PNs, few are specifically tailored to small pulmonary nodules (SPNs, \u0026le; 20 mm). This study entailed the development and validation of a predictive model specifically for differentiating malignant from benign SPNs.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e\u003cp\u003eA retrospective study was conducted using patient data from two medical centers between January 2019 and December 2024. Patients were separated into training and testing cohorts based on the hospital of origin. All patients underwent computed tomography (CT) and positron emission tomography (PET)/CT examinations. The predictive model was constructed using the training cohort and subsequently validated with the testing cohort. Model performance was also compared against that of an existing SPN predictive model.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e\u003cp\u003eThe training set comprised 155 patients (111 malignant and 44 benign), while the testing set included 82 patients (64 malignant and 18 benign). Logistic regression analysis identified lobulation (P\u0026thinsp;=\u0026thinsp;0.015) and elevated standardized maximum uptake value (SUV\u003csub\u003emax\u003c/sub\u003e, P\u0026thinsp;=\u0026thinsp;0.015) as independent predictors of malignancy. The final prediction model was defined as: X\u0026thinsp;=\u0026thinsp;\u0026minus;\u0026thinsp;4.446\u0026thinsp;+\u0026thinsp;1.305 \u0026times; lobulation (present\u0026thinsp;=\u0026thinsp;1, absent\u0026thinsp;=\u0026thinsp;0)\u0026thinsp;+\u0026thinsp;0.327 \u0026times; SUV\u003csub\u003emax\u003c/sub\u003e. In the training cohort, the area under the curve (AUC) for the new model was 0.821, compared to 0.715 for the existing model. In the testing cohort, AUCs were 0.853 and 0.701, respectively. Calibration curves demonstrated strong concordance between predicted and actual malignancy probabilities.\u003c/p\u003e\u003ch2\u003eConclusions\u003c/h2\u003e\u003cp\u003eThe proposed predictive model demonstrated high diagnostic accuracy differentiating malignant from benign SPNs and may help to reduce the unnecessary biopsies.\u003c/p\u003e","manuscriptTitle":"Predictive model for differentiating malignant and benign small pulmonary nodules","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-10-08 15:58:54","doi":"10.21203/rs.3.rs-7347617/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"reviewersInvited","content":"","date":"2025-09-25T12:09:44+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2025-09-01T02:53:23+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-08-29T10:10:41+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-08-29T10:09:33+00:00","index":"","fulltext":""},{"type":"submitted","content":"BMC Pulmonary Medicine","date":"2025-08-11T14:41:48+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"bmc-pulmonary-medicine","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"pulm","sideBox":"Learn more about [BMC Pulmonary Medicine](http://bmcpulmmed.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/pulm/default.aspx","title":"BMC Pulmonary Medicine","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"842f32bb-d316-47c1-8b06-cd2fbe0a3f08","owner":[],"postedDate":"October 8th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2025-10-08T15:58:55+00:00","versionOfRecord":[],"versionCreatedAt":"2025-10-08 15:58:54","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7347617","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7347617","identity":"rs-7347617","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}