Answers are grounded in this paper and cite their sources.
Lung cancer remains the leading cause of cancer-related mortality worldwide, with 2.21 million new cases and 1.80 million deaths annually. Existing diagnostic methods often lack accuracy, accessibility, and cost-effectiveness, especially in low-resource settings. This study proposes a robust ensemble machine learning framework for lung cancer risk prediction using a dataset of 309 patients with 15 clinical and lifestyle features. After advanced preprocessing and feature engineering, 25 variables were selected. Eleven machine learning algorithms, including Random Forest, Gradient Boosting, Neural Networks, and Support Vector Machines, were evaluated. The Hard Voting Ensemble classifier achieved the highest performance with 95.16% accuracy, 98.11% precision, 96.30% recall, 97.20% F1-score, and a Matthews Correlation Coefficient of 0.798. Stratified 5-fold cross-validation yielded a mean accuracy of 91.52% ± 5.82%. Key predictors included total symptom burden, general symptom score, and allergy status. The proposed ensemble model outperformed individual algorithms, demonstrating strong clinical relevance and potential integration into clinical decision support systems.
No takes yet. Share an insight, caveat, or question.
Sahare et al. (2025) studied this question.