Optimized Reformed Histogram Equalization and SE-CNN with Chan Vese Segmentation for an Automatic Prediction of Lung Cancer

Lung cancer is one of the deadliest illnesses and the leading cause of illness and death, when detected in its early stages. While many classical machines and deep learning automatic systems can identify lung cancer with effectiveness, they still have issues with recognition accuracy and take longer to analyze. The Squeeze and Excitation Convolutional Neural Network (SENET), a deep learning technique, is proposed in this research. To increase the clarity of an image, lung CT input pictures are pre-processed employing the Adaptive Bilateral Filter (ABF) and Reformed Histogram Equalization (RHE). Tuna-swam optimization is utilized to determine the tunable parameters of RHE approach. Chan Vese segmentation model is used for segment the object and background of pre-processed image. This segmented image is given to SENET classifier for lung cancer prediction. The suggested strategy achieves 98. 7% accuracy, 2.3% error, and 97.4% precision, based on experimental outcomes. As a result, the proposed framework successfully predicts lung cancer and raises the standard of living for patients.