Computer Vision News - May 2022

48 AI for Urology prostate cancer, the Prostate Imaging– Reporting and Data System (PI-RADS) scoring system was developed . It utilizes several MRI sequences and specific lesion characteristics to provide a numerical score for risk assessment. However, calculating the PI-RADS score is time-consuming and subjective . Image quality, which depends on technician experience and hardware, highly affects the lesion characteristics. Also, each radiologist views the MRI sequences differently, and determines the lesion based on personal experience, leading to inter-user variability. This process takes time, and as this condition is prevalent, there are many prostate MRI scans to view, taking up a large portion of the radiologists’ workload. RSIP Vision has recently developed a new tool to assist in PI- RADS scoring . This tool performs segmentation of the prostate, its sub- sections, and lesions. It analyzes the lesions’ intensity, restriction, size, and shape, and provides a baseline for Prostate Imaging–Reporting and Data System score. This information can be used In a recent article published on our website, we described the challenges in Prostate Imaging– Reporting and Data System (PI- RADS) scoring, and that Artificial Intelligence (AI) plays a critical role in facing these challenges. We are now excited to share RSIP Vision’s solution: the PI-RADS Assistant. by Oren Wintner Prostate cancer is the most common male-cancer, affecting nearly %10 of the population. Similar to all cancers, early diagnosis and treatment significantly improve survival rates and reduce complications. Diagnosis of prostatic cancer requires several clinical and imaging tests . Firstly, blood PSA level is tested. Imaging tests usually involve ultrasound (US) and MRI scans. To construct an objective measure for risk of PI-RADS ASSISTANT: NEW AI TOOL FOR PROSTATE MRI ANALYSIS