Methodology for Automatic diagnosing Colon and Lung Carcinoma using TSA-PSODL Algorithm

Artificial Intelligence (AI) has significantly advanced in biomedical diagnostics and research, particularly in disease evaluation, medication development, and therapeutic interventions. This technology is revolutionizing the field, and one of its most critical applications is in the identification of cancer, which kills one in six people worldwide and is the leading cause of mortality. Among the various types of cancer, colon and lung cancers are particularly prevalent and lethal. Early detection of these cancers can greatly improve survival rates. ML and DL methods are being utilized with growing frequency to expedite the process of cancer detection. These methods enable researchers to analyze vast amounts of patient data more efficiently and cost-effectively. A promising new approach involves using the Tuna Swarm Algorithm (TSA) and Particle Swarm Optimization (PSO) using a deep learning model referred to as the TSA-PSODL approach, advanced analysis of biomedical images is being developed to detect colon and lung cancer. The TSA-PSODL technique preprocesses input images with Laplacian of Gaussian filtering (LoG) to enhance image quality and highlight key features necessary for accurate cancer detection. In addition, the TSA-PSODL approach employs InceptionNetV5 feature extractor to generate a comprehensive set of feature vectors. The Adaptive Fuzzy Adam Optimizer (AFAO) is then used to fine-tune the hyperparameters of these feature extractors, improving their performance. Finally, the TSA-PSODL approach is utilized to use an Echo State Network (ESN) classifier to identify lung and colon cancer, providing a robust and effective solution for early diagnosis. This innovative methodology showcases the immense potential of AI in revolutionizing cancer detection and treatment, paving the way for more efficient and accurate diagnostic tools. An comprehensive experimental result is conducted in order to illustrate the more amazing consequence of the TSA-PSODL system. With a maximum accuracy of 99.44%, when compared to other approaches, the comprehensive comparison study showed that the TSA-PSODL strategy was more efficient.