IBM Research – Cognitive Assistance for Radiologists

Together with IBM’s Israeli research center, MK&M Big Data Science Institute is working on automatically identifying breast cancer using mammography images from Maccabi’s database.

Researchers from MK&M Big Data Science Institute along with their counterparts at IBM are using image analysis technologies and deep learning techniques to ‘teach’ computers how to identify breast cancer findings in mammography images.

The ability of machines to detect breast cancer from mammography images can be used at first to assist doctors in decision-making. Later on, the vision is that it may be combined with additional patient data to create a predictive analysis algorithm that could actually help to prevent breast cancer.

Today, breast cancer comprises nearly 30% of all diagnosed cancers in women across the globe, making it the second leading cause of death for women. While much work has been performed to facilitate breast cancer detection—including advanced computer-aided diagnosis tools for mammography images—until now, little or no work has been done in the creation of advanced decision-support tools that combine multi-modal image analytics and clinical data analysis.

This new diagnostic technology uses sophisticated medical imaging, deep learning and clinical inference technologies to analyze patient cases using a systematic, clinical thought process. It begins with the system examining the patient case description. It analyzes the text, highlighting the relevant clinical concepts and summarizing the findings, and then analyzes the imaging study, trying to locate potential lesions. Suspicious regions are highlighted and classified as potential masses, and then conclusions are drawn. The radiologists handling the case now have a second opinion they can refer to, enabling them to make a more informed and even more accurate diagnosis.