Empowering Cancer Diagnosis: Unveiling Bladder Cancer with Advanced Deep Learning

Abstract

Artificial intelligence and deep learning in particular are now considered as promising methods in numerous scientific fields including medical science because the former can handle large amounts of data with non-linear relationships. Urinary bladder cancer is another frequent neoplasm that has variable histological variants and requires correct classification for the appropriate treatment tactics and prediction. This study proposed a Convolutional Neural Network (CNN) model that is accurate and simple to classifying bladder cancer, where, comparing its performance against three transfer learning architectures: These models include VGG16, InceptionV3, and MobileNetV2 for the two datasets that were used. It is seen from the experimental results that the proposed deep CNN model achieves a higher overall accuracy than the established transfer learning models which is 99. 5% in the final prediction and this makes the system suitable as a diagnostic tool for diagnosing bladder cancer.