Medical research indicates that prostate cancer is among the deadliest types of cancer. Detecting prostate cancer (PCa) early greatly impacts how effectively the tumor can be treated. Artificial intelligence (AI) is advancing the field of digital pathology, enabling researchers to analyze large datasets and diagnose prostate cancer tumors more quickly and accurately. One method for early and rapid detection of prostate cancer is through image analysis software that uses machine learning or deep learning algorithms. This study evaluates the effectiveness of two convolutional neural network (CNN) algorithms for detecting prostate cancer. Python programming, along with Keras and TensorFlow frameworks, are utilized to implement two deep learning models: InceptionV3 and ResNet50. These models were evaluated using the MyDataset4 prostate magnetic resonance images (MRI) dataset from the Kaggle platform, which contains both cancerous and non-cancerous images. Both models have been applied to various medical imaging diagnostic tasks and have shown outstanding performance. The models are assessed based on accuracy, F1 score, recall, and precision to gauge their performance. Test results indicate that ResNet50 achieved the highest accuracy at 93%, while InceptionV3 reached an accuracy of 91%.