Deep Fuzzy SegNet-based lung nodule segmentation and optimized deep learning for lung cancer detection

Globally, lung cancer has a high fatality rate and is a lethal disease. Since lung cancer affects both men and women, it requires extra consideration when evaluating various diseases. Furthermore, early detection is even more important in order to increase the survival percentage of affected patients. There are many methods for detecting lung cancer, but it can be difficult to locate the affected area due to low visibility of the tumor section and imaging failure rates. Due to poor image quality, which distorts the segmentation process, the standard strategies failed to increase the accuracy rate. In order to diagnose lung cancer disease, this research created an approach known as Bat Deer Hunting Optimization Algorithm-based Deep Convolutional Neural Network (BDHOA-based DCNN). Here, Computed Tomography pictures are used to predict the presence of lung cancer. The Bat Algorithm (BA) and Deer Hunting Optimization Algorithm have been integrated into the newly developed BDHOA algorithm (DHOA). To execute the lung cancer detection and classification, the lung lobe and nodule region is segmented from the lung picture. With accuracy, sensitivity, and specificity scores of 0.9243, 0.9421, and 0.8915, the suggested approach performed better.