$GTCH "According to our R&D plans 2021, we are moving forward to develop more advanced AI systems for the medical domain. Today's medical imaging consists of ultra high resolution scans that include numerous details. We started the development of our Imaging Predicting Analytics system, internal code name IPA, which we believe will be especially useful if fully developed when quick response is necessary. For example, in case of cardiovascular or respiratory conditions, a rapid imaging analytics may save lives by quickly finding anomalies that may be urgently addressed. The system is planned to have 2D and 3D simulation including rotations, cross sections and mapping. We already developed graphic analytics techniques within our Avant! AI system and now plan to further develop advanced deep learning imaging algorithms to analyze modern images of CT, MRI, X-RAY, PET and Ultrasound. We believe we will be able to target the system to assess images for evidence of abnormalities and alert providers to the potential diagnoses, with the goal of pointing subjects-of-interest to physicians, enabling faster treatment. The system is planned to use AI algorithms to learn the image parameters and train a neural network for pixels recognition and analytics. We intend for the the system to provide risk scores for areas of concern. We target to improve diagnostic's accuracy and potentially alert for risky factors, eliminating unnecessary tests and benign biopsies. We believe that medical imaging data is one of the richest sources of information, and one of the most complex ones. Artificial intelligence science has already proven to be an efficient assistant for radiologists and health professionals for the past decade, pursuing faster, more accurate, and comprehensive patient care. But our challenges are growing exponentially and so are our needs. 2D data is rapidly moving to 3D form, enabling better information representation in many medical fields. Today's imaging analytics complexity requires advanced computerized disciplines and a quantum leap approach in order to handle vast data, especially when real-time factor has become crucial. Our goal is to develop native graphic-rich disciplines to augment image analytics algorithms and flows, enabling better, and more accurate medicine and diagnostics."
http://www.wsj.com/articles/PR-CO-20201222-903332
http://www.wsj.com/articles/PR-CO-20201222-903332
