Detection of Lung Cancer in Histopathological Images Utilizing Self-Guided Quantum Generative Adversarial Network with Cat Hunting Optimization
Law Kumar Singh, G Ranjitham., Prasanta Kumar Parida, Loganathan Guganathan, Sridhar. S · 2025
One of the most deadly illnesses of the last few decades is lung cancer. According to research, histopathological images acquired during biopsy are the most accurate way to diagnose cancer. In bioinformatics, deep learning methods have demonstrated notable success, especially in medical imaging. Many current techniques are inaccurate, which raises the possibility of a misdiagnosis and the risks of incorrect treatments. In order to overcome these constraints, this research proposed a Harr Wavelet Quantum Generative Adversarial Network with Cat Hunting Optimization (HWQGAN-CHO) method. Using a modified SR-SHAKF for pre-processing improves estimation and reduces noise, increasing data accuracy. The Scale-Aware Modulation Meet Transformer and Quantized Discrete Haar Wavelet Transform are utilized for segmentation and feature extraction in image analysis at different frequency levels.CHO is used to optimize SGQGAN hyperparameters by simulating a cat's hunting behavior, enhancing the model's performance and resulting in more accurate results using the LC25000 database. On the LC25000 dataset, it detected lung cancer in histopathology images with an accuracy of 99.9%. These results demonstrate the strategy's ability to outperform existing methods and open the path for serious advancements in lung cancer diagnosis.