Enhanced CNN-Based Lung Cancer Detection Using Chi-Squared Feature Selection and Self-Adaptive Differential Evolution Optimizer

Lung cancer, a leading cause of cancer-related deaths worldwide, accounts for a significant proportion of global cancer mortality due to delayed diagnosis and limited access to effective screening methods. Early detection plays a crucial role in improving survival rates, yet traditional diagnostic techniques often lack the necessary precision and scalability. This study introduces an advanced AI-driven approach to lung cancer detection, combining the strengths of Convolutional Neural Networks (CNNs) with Chi-Squared feature selection and hyperparameter optimization using the Self-Adaptive Differential Evolution (SADE) algorithm. The proposed model achieves a remarkable accuracy of 94%, significantly outperforming traditional methods such as Gated Recurrent Unit (GRU), Long Short-Term Memory (LSTM), and Support Vector Machine (SVM) in both accuracy and loss metrics. By addressing the limitations of existing techniques, this innovative model has the potential to revolutionize early-stage lung cancer diagnostics, enhancing precision, scalability, and patient outcomes. This work underscores the transformative role of AI in advancing medical diagnostics.