Computer Vision News - April 2019

16 Computer Vision News Pharma - Automated RECIST Score Every month, Computer Vision News reviews a successful project. Our main purpose is to show how diverse image processing techniques contribute to solving technical challenges and real world constraints. This month we review a Medical Image Analysis project by RSIP Vision for the Pharma industry: Automated RECIST Score . Project One of the most common practices in clinical trial is: tracking the effect of a new drug on cancer tumors . The Pharma industry constantly needs a fast and accurate system for measuring tumors and tracking tumor progression along the treatment. The golden standard is the RECIST score (Response Evaluation Criteria in Solid Tumors) , used by the medical community to assess solid tumors (generally on CT scans) for almost 20 years. The RECIST score is based on the measurement of the diameter of the tumor, calculated as the longest line that can run through the tumor under any angle. To this day, radiologists measure the RECIST score manually , which is problematic for mainly two reasons: the first reason is the long time it takes the expert to view the tumor from all possible angles; the second reason is the chance of mistake (like skipping the angle with the longest diameter), together with the large inter- and intra- variability of human observations. Tumors vary in shapes and sizes, and they represent a severe challenge to the human observer . RSIP Vision developed a software and procedure to solve this problem. This solution supports the radiologists in their decision (without replacing them). The algorithm assists them by offering a faster and more accurate detection and measurement of the longest diameter in each tumor. by Miki Haimovich The algorithm assists radiologists by offering them faster and more accurate detection and measurement of the longest diameter in each tumor.