Deep Learning Model for Pathological Grading and Prognostic Assessment of Lung Cancer Using CT Imaging: A Study on NLST and External Validation Cohorts.
- Department of Epidemiology and Health Statistics, School of Public Health, Capital Medical University, Beijing, China (R.Y., W.L., S.Y., H.Z., X.L., L.T., X.G.); Beijing Municipal Key Laboratory of Clinical Epidemiology, Capital Medical University, Beijing, China (R.Y., W.L., S.Y., H.Z., X.L., L.T., X.G.). Electronic address: yrh@mail.ccmu.edu.cn.
- Department of Epidemiology and Health Statistics, School of Public Health, Capital Medical University, Beijing, China (R.Y., W.L., S.Y., H.Z., X.L., L.T., X.G.); Beijing Municipal Key Laboratory of Clinical Epidemiology, Capital Medical University, Beijing, China (R.Y., W.L., S.Y., H.Z., X.L., L.T., X.G.). Electronic address: lwming@ccmu.edu.cn.
- Department of Epidemiology and Health Statistics, School of Public Health, Capital Medical University, Beijing, China (R.Y., W.L., S.Y., H.Z., X.L., L.T., X.G.); Beijing Municipal Key Laboratory of Clinical Epidemiology, Capital Medical University, Beijing, China (R.Y., W.L., S.Y., H.Z., X.L., L.T., X.G.). Electronic address: ysq@mail.ccmu.edu.cn.
- Harvard T. H. Chan School of Public Health, Boston, Massachusetts (Z.W.). Electronic address: zhiyuanwu@hsph.harvard.edu.
- Department of Epidemiology and Health Statistics, School of Public Health, Capital Medical University, Beijing, China (R.Y., W.L., S.Y., H.Z., X.L., L.T., X.G.); Beijing Municipal Key Laboratory of Clinical Epidemiology, Capital Medical University, Beijing, China (R.Y., W.L., S.Y., H.Z., X.L., L.T., X.G.). Electronic address: yeszhang@mail.ccmu.edu.cn.
- Department of Epidemiology and Health Statistics, School of Public Health, Capital Medical University, Beijing, China (R.Y., W.L., S.Y., H.Z., X.L., L.T., X.G.); Beijing Municipal Key Laboratory of Clinical Epidemiology, Capital Medical University, Beijing, China (R.Y., W.L., S.Y., H.Z., X.L., L.T., X.G.). Electronic address: lxiangtong@ccmu.edu.cn.
- Department of Epidemiology and Health Statistics, School of Public Health, Capital Medical University, Beijing, China (R.Y., W.L., S.Y., H.Z., X.L., L.T., X.G.); Beijing Municipal Key Laboratory of Clinical Epidemiology, Capital Medical University, Beijing, China (R.Y., W.L., S.Y., H.Z., X.L., L.T., X.G.). Electronic address: taolixin@ccmu.edu.cn.
- Department of Mathematics and Statistics, La Trobe University, Melbourne, Australia (X.L.). Electronic address: lixia_new@163.com.
- School of Mathematical Sciences, University College Cork, Cork, Ireland (J.H.). Electronic address: j.huang@ucc.ie.
- Department of Epidemiology and Health Statistics, School of Public Health, Capital Medical University, Beijing, China (R.Y., W.L., S.Y., H.Z., X.L., L.T., X.G.); Beijing Municipal Key Laboratory of Clinical Epidemiology, Capital Medical University, Beijing, China (R.Y., W.L., S.Y., H.Z., X.L., L.T., X.G.); Centre for Precision Health, School of Medical and Health Sciences, Edith Cowan University, Joondalup, Australia (X.G.). Electronic address: statguo@ccmu.edu.cn.
RATIONALE AND OBJECTIVES: To develop and validate a deep learning model for automated pathological grading and prognostic assessment of lung cancer using CT imaging, thereby providing surgeons with a non-invasive tool to guide surgical planning.
MATERIAL AND METHODS: This study utilized 572 cases from the National Lung Screening Trial cohort, dividing them randomly into training (461 cases) and internal validation (111 cases) sets in an 8:2 ratio. Additionally, 224 cases from four cohorts obtained from the Cancer Imaging Archive, all diagnosed with non-small cell lung cancer, were included for external validation. The deep learning model, built on the MobileNetV3 architecture, was assessed in both internal and external validation sets using metrics such as accuracy, sensitivity, specificity, and area under the receiver operating characteristic curve (AUC). The model's prognostic value was further analyzed using Cox proportional hazards models.
RESULTS: The model achieved high accuracy, sensitivity, specificity, and AUC in the internal validation set (accuracy: 0.888, macro AUC: 0.968, macro sensitivity: 0.798, macro specificity: 0.956). External validation demonstrated comparable performance (accuracy: 0.807, macro AUC: 0.920, macro sensitivity: 0.799, macro specificity: 0.896). The model's predicted signatures correlated significantly with patient mortality and provided valuable insights for prognostic assessment (adjusted HR 2.016 [95% CI: 1.010, 4.022]).
CONCLUSIONS: This study successfully developed and validated a deep learning model for the preoperative grading of lung cancer pathology. The model's accurate predictions could serve as a useful adjunct in treatment planning for lung cancer patients, enabling more effective and customized interventions to improve patient outcomes.