Prostate Cancer Classification using Cat Swarm Optimization with Deep Learning on MRI Images

Prostate Cancer (PC) is a common malignancy amongst men globally, and accurate classification plays a significant part in its diagnoses and treatment planning. Due to its superior soft tissue contrast, Magnetic Resonance Imaging (MRI) has developed as a valuable imaging modality for the assessment of PC. PC classification on MRI is a major task in medical imaging analysis, and deep learning (DL) approaches are employed to accomplish automated and accurate classification. In recent times, Convolutional Neural Network (CNN) structure was selected for the classification of PC. This manuscript introduces an automated Prostate Cancer Classification using Cat Swarm Optimization with Deep Learning (PCC-CSODL) technique on MRI Images. The main purpose of the PCC-CSODL algorithm lies in the effectual and accurate identification of PC. For achieving this, the PCC-CSODL technique involves NASNet feature extractor at the initial stages. In addition, the CSO algorithm is used for the optimum hyperparameter selection of the NASNet model. For classification purposes, attention based long short term memory (ALSTM) model can be used. The results demonstrated the effectiveness of the PCC-CSODL technique in accurately classifying PC on MRI. It has the potential to assist clinicians in making informed decisions and enhancing patient outcomes in PC management.