Objective: Progressive ground-glass nodules (GGNs) are associated with a risk of malignancy. Early identification and intervention are crucial for lung cancer prevention. Traditional Chinese medicine (TCM) has advantages of syndrome differentiation, holistic adjustment, and whole-process management. In this study, we built a risk assessment model based on the integration of TCM symptoms and imaging features to identify risks of GGN progression and to guide personalized TCM management and treatment.
Methods: This study enrolled 864 patients with GGNs from January 2015 to December 2023. Assessment factors included general demographic information, TCM symptoms, and imaging features. Machine learning algorithms were used to screen risk factors, regression models were used to establish GGN progression risk assessment models, and a nomogram was used to assess the risk of progression.
Results: Age, second-hand smoke exposure, acid reflux, breast or abdominal pain, chest tightness, depression, heartburn, tongue with little coating, and the diameter of the largest pulmonary nodule were significant risk factors for GGN progression. In the nomogram model, the TCM symptom of acid reflux made the highest contribution. The constructed model using these key factors was used to assess the risk of GGN progression. The receiver operating characteristic curve showed that the area under the curve (AUC) in the training set was 0.70 (95% CI [0.65, 0.75]), and the AUC in the test set was 0.74 (95% CI [0.65, 0.82]).
Conclusion: This study was the first to establish a GGN progression risk assessment model using TCM clinical symptoms and imaging features. The model can help individuals and doctors better understand the health status and take timely preventive measures. Please cite this article as: Maolan A, Yu LH, Li Y, Yan ZS, Xiang XH, Wang LF, Guo QJ, Hu JQ, Zhang GH, Ren XL, Li J, Sun TH, Yang W, Liu R, Hua BJ. A progression risk assessment model for pulmonary ground-glass nodules based on traditional Chinese medicine clinical symptoms combined with imaging. J Integr Med. 2026; Epub ahead of print.
扫码关注我们
求助内容:
应助结果提醒方式:
