Sign in Subscribe
Medical research

Personalized Treatment Duration for Pulmonary Tuberculosis Using Radiomics and Clinical Indicators

Strategic Insights

— 1 min read

A groundbreaking study conducted by researchers at Capital Medical University in Beijing, People's Republic of China, has developed predictive models for individualized treatment duration in newly diagnosed pulmonary tuberculosis (PTB) patients. The innovative approach combines radiomics features with clinical indicators to support personalized therapeutic strategies. A retrospective cohort of 242 newly diagnosed PTB patients was analyzed, and three models were constructed using Cox proportional hazards regression, achieving high predictive performance and calibration. The study concludes that this approach provides a practical tool for personalizing treatment duration and supports more precise management of PTB patients.

Key Takeaways:

  • The treatment duration for pulmonary tuberculosis (PTB) varies significantly based on disease severity, pathogen characteristics, and host immune status.
  • A retrospective cohort of 242 newly diagnosed PTB patients was analyzed, and radiomic features were selected via Least Absolute Shrinkage and Selection Operator (LASSO)-Cox regression.
  • Three models-a radiomics model, clinical model, and radiomics-clinical combined model-were constructed using Cox proportional hazards regression, achieving C-indices of 0.81 (training cohort) and 0.79 (testing cohort).
  • The combined model outperformed the radiomics-only and clinical-only models, achieving time-dependent AUCs consistently above 0.75.
  • Calibration curves demonstrated good agreement between predicted and observed outcomes, and Kaplan-Meier (K-M) analysis confirmed that the Rad score effectively stratified patients by treatment duration.
  • The study incorporated two radiomic features and three clinical variables into the final models, providing a practical tool for personalizing treatment duration in PTB patients.

Statistics:

  • 242 newly diagnosed PTB patients were analyzed in the retrospective cohort.
  • The training cohort consisted of 128 patients, and the testing cohort consisted of 114 patients.
  • The C-index evaluated the predictive performance of the models, with values ranging from 0.81 to 0.79.
  • The time-dependent AUC consistently exceeded 0.75, indicating high predictive performance.
  • The study included two radiomic features and three clinical variables in the final models.

Sources:

  • A radiomics-clinical nomogram for predicting individualised treatment duration in newly diagnosed pulmonary tuberculosis. Clinical Radiology, 2025;90:107059. Clinical Radiology can be contacted at: W B Saunders Co Ltd, 32 Jamestown Rd, London NW1 7BY, England. (Elsevier - www.elsevier.com; Clinical Radiology - www.journals.elsevier.com/clinical-radiology/)
  • NewsRx. Findings from Capital Medical University Has Provided New Information about Pulmonary Tuberculosis (A radiomics-clinical nomogram for predicting individualised treatment duration in newly diagnosed pulmonary tuberculosis). TB & Outbreaks Week. October 21, 2025; p 835.

Read more