This study presents a binary classification method that predicts the suitability of ovarian cancer patients for Bevacizumab therapy based on their histopathology images. Using a pre-trained ResNet-18 architecture, the study attempts to distinguish patients likely to benefit from the Bevacizumab therapy from those who won’t. Further assessing the effectiveness of the suggested model is a comparison with the VGG16 model. Our model shows remarkable prediction accuracy with rigorous training and validation stages, underscoring the significant contribution artificial intelligence could make to expediting the oncology treatment plan decision-making process. According to the findings, AI-driven models can improve personalized medicine by correctly identifying patients who would benefit from Bevacizumab therapy, which will improve patient outcomes and therapeutic resource allocation. The findings highlight how AI-driven models can improve personalized medicine by correctly identifying patients who might benefit from Bevacizumab therapy. This will improve patient outcomes and resource allocation in the clinical setting.