Accurate report generation from medical images is a critical aspect of healthcare, enhancing the efficiency and precision of diagnoses. Medical caption generation aims to assist healthcare professionals in interpreting and analyzing radiology images, facilitating faster and more accurate diagnostic processes. Additionally, there exist a scarcity of medical reports in low-resource languages which are incredibly valuable in rural and lowly populated areas. Our study focuses on the task of using the ImageClef dataset from the ImageCLEF medicalCaption task. The captions provided in the training dataset were translated to the Tamil language to assist in generating medical captions in a low-resource language. We implemented the Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generation (BLIP) model, which utilizes a multimodal mixture of encoder-decoder components. This proposed approach leverages the strengths of vision as well as language models to generate detailed medical image captions that are accurate. By integrating these advanced techniques, we aim to significantly improve the utility of automated medical captioning in clinical settings for a low - resource language like Tamil.