JoVE - Journal of Visualized Experiments
JoVE Visualize
Contact Us

Machine learning driven prediction of drug efficacy in lung cancer: based on protein biomarkers and clinical features

  • 0Beijing University of Chinese Medicine, Beijing, China.
Life Sciences +

|

May 12, 2025

|

May 12, 2025

View abstract on PubMed

Summary

This summary is machine-generated.

Machine learning models, particularly CatBoost, show promise in predicting lung cancer patient survival and stratifying risk. These AI tools can aid clinicians in developing personalized treatment plans for better patient outcomes.

Area Of Science

  • Oncology
  • Medical Informatics
  • Artificial Intelligence

Background

  • Chemotherapy is the primary lung cancer treatment, but efficacy assessment is challenging.
  • Machine learning (AI) offers advanced prediction capabilities by integrating diverse patient data.
  • Accurate prediction of survival and risk stratification is crucial for optimizing lung cancer treatment.

Purpose Of The Study

  • To evaluate the efficacy of machine learning models in predicting overall survival (OS) and progression-free survival (PFS) in lung cancer patients.
  • To compare the performance of ten different machine learning models for survival prediction.
  • To assess the utility of these models in stratifying lung cancer patients into high-risk and low-risk groups.

Main Methods

  • Collected clinical and hematological data from 2115 lung cancer patients.
  • Trained and evaluated ten machine learning models (Decision Tree, Random Forest, Logistic Regression, k-NN, AdaBoost, XGBoost, CatBoost) for OS and PFS prediction.
  • Assessed model performance using Area Under the Curve (AUC) and analyzed risk stratification capabilities.

Main Results

  • The CatBoost model demonstrated superior performance in predicting 3-year OS and PFS, achieving AUCs of 0.97 and 0.95, respectively.
  • CatBoost excelled in distinguishing between high-risk and low-risk patients, showing strong predictive power across various time points (1, 3, and 5 years).
  • All evaluated models showed effectiveness in risk stratification, with CatBoost being the most proficient.

Conclusions

  • Machine learning models, especially CatBoost, are effective tools for predicting lung cancer survival and stratifying patient risk.
  • These AI-driven models can assist clinicians in making more informed decisions for personalized lung cancer treatment strategies.
  • The study highlights the potential of AI in improving clinical prediction and treatment planning for lung cancer.

Keywords: