Chameleon-Inspired Evolutionary Optimization Enhanced with Deep Learning for Improved Lung Cancer Classification

Abstract

The development of computer methods in the field of medical image analysis in recent years has shown promising potential for enhancing lung cancer classification. This research paper presents a novel approach that combines evolutionary optimization inspired by the adaptability of chameleons with deep learning methodologies to raise the precision and effectiveness of lung cancer identification and categorization. In the first step of our methodology, we employ pre-processing techniques including Canny edge detection and Gaussian blur to enhance the quality of the input lung cancer images. Subsequently, leveraging the principles of evolutionary optimization, particularly inspired by the Chameleon abilities, the proposed approach aims to dynamically adapt its parameters to optimize the classification performance on lung cancer images. Additionally, deep learning techniques, including convolutional neural networks (CNNs) and dilated convolution, are integrated into the optimization framework to further enhance feature extraction and classification accuracy. The outcomes of the experiment show how effective the suggested strategy in achieving superior performance compared to existing methods, with significant improvements in accuracy in lung cancer detection and classification tasks.