Accurate object counting in medical images is a critical task for detect and monitoring various health conditions, including tumor detection and cell counting. Traditional methods of counting objects in images, often reliant on manual annotation or basic image processing techniques, can be labor-intensive and prone to human error, resulting in inconsistent and imprecise counts. This study aims to address these challenges by leveraging advanced machine learning techniques to enhance object counting accuracy in medical imaging. Existing approaches frequently fall short in handling complex image variations and may lack the precision required for high-stakes medical applications. Our proposed method utilizes state-of-the-art machine learning algorithms to automate and improve the precision of object counting, filling the gap in current methodologies by providing robust solutions for diverse and challenging medical imaging scenarios. The outcomes of this study are crucial as they could significantly improve diagnostic accuracy and efficiency, ultimately leading to better patient outcomes and streamlined healthcare processes. Addressing this problem is essential due to the growing need for reliable and scalable tools in medical image analysis, which can help in making timely and accurate medical decisions.