Lung cancer is a leading cause of cancer-related deaths, often diagnosed late due to its aggressive nature. This study presents a novel Adaptive Dendritic Neural Model (ADNM) to enhance diagnostic accuracy in high-dimensional healthcare data. Utilizing hyperparameter optimization and activation mechanisms, ADNM improves scalability and feature selection for multi-class lung cancer prediction. Using a Kaggle dataset, Particle Swarm Optimization (PSO) selected features, while bootstrap assessed performance. ADNM achieved 98.39% accuracy, 99% AUC, and a Cohen's kappa of 96.95%, with rapid convergence via the Adam optimizer, demonstrating its potential for improving early diagnosis and personalized treatment in oncology.
Keywords: Dendritic neural model; ensemble learning; lung cancer prediction; machine learning; optimization.