Breast cancer remains the most diagnosed cancer in women universally, and an estimated 1.54 million new cases are registered every year. Non-invasive medical imaging apart, traditional diagnostics, which are mostly based on mammography, have long been variable capacity for detection of small or occult tumours, most notoriously in dense breast tissue. This failure contributes to late staging and high mortality, most in under-resourced regions such as India, where one in every nine women is at risk. The proposed system combines the strengths of two-state-of-the-art deep learning networks, YOLOv8 and Convolutional Neural Networks, YOLOv8, as an object detector in real-time. This approach is also applied to suspicious mass detection as well as malignant and benign tumour separation. However, in an attempt to overcome its limitation in creating elaborated tumour characterizations, CNN are used for fine-grained analysis, such as border detection and malignancy size estimation. Such a two-model architecture is likely to provide both detection accuracy as well as diagnostic depth. Additionally, various pre-trained CNN models, including DenseNet-121, VGG-16, MobileNetV2, ResNet152V2, and InceptionV3, are compared to determine the best model to classify between cases and controls for breast cancer. Among them, the most remarkable accuracy was recorded by DenseNet-121 at 99%, followed by VGG-16 (98%) and MobileNetV2 (97%), which states the performance as well as feasibility in medical image analysis. The diverse multimodal dataset was used to validate the system by considering generalizability for a variety of tumour types as well as imaging conditions. The findings demonstrated a significant reduction in both false positives and false negatives, shows the new method’s potential for both improved patient outcomes and early identification. Deep learning methods has significant potential for improving diagnostics processes, particularly in underserved areas where qualified radiologists and pathologists are not easily available. It highlights how AI can revolutionize the diagnosis of breast cancer and opens the door for future studies into real-time clinical application, electronic medical record interface, and further development of AI models for potential use in medical imaging. It will be essential to keep improving these intelligent systems in order to help oncologists, advance individualized care, and eventually lower the number of breast cancer fatalities worldwide.