Augmented reality aid in diagnostic assistance for breast cancer detection

Breast cancer is a significant public health concern with over 2.2 million cases reported in 2022 by the World Health Organization. Computer-aided detection (CAD) systems that use machine learning (ML) have become a crucial tool in assisting clinicians with breast cancer diagnostics. Our proposal offers an augmented reality (AR) system that allows surgeons to visualize cancer tumors in a cost-effective manner. The approach involves noise suppression using a Median filter, Top-Hat transform for blocking clear spots, and segmentation of active geometric contour models based on speckle (plane set) image. The pro-posed approach was validated experimentally, with a sensitivity of 90% accuracy of over 98% for tumor cells. Using 3D Slicer, the 3D breast mass reconstruction can be virtually augmented on the actual scene, which significantly improves breast mass extraction, additional 3D reconstruction, 3D interaction, and AR visualization. Overall, the proposal presents a promising approach for assisting clinicians with breast cancer diagnosis and treatment.