Computer Vision News - April 2022

38 Medical Imaging Applications anonymized data available to build better robotics . “ It’s a massive piece of work with a huge amount of surgical data to analyze, ” Karim tells us. “ We’re using deep learning and convolutional neural networks to develop an algorithm that can detect what went well and what didn’t go so well. The goal is to have an algorithm that provides a high degree of accuracy. We’ll compare it to what happens in surgery so that the correlation is 99%. Then we’ll have reached the level where we can say our they are one of the first groups to be using computer vision to look at anomalies in big data. The team has several target groups for this information. Insurance companies selling hedging products to hospitals and surgeons can use it to understand the risks in the operating room . Hospitals will have an idea of the correlation between anomalies and errors in the operating theatre to support them in mitigating risk, increasing efficiency, and reducing costs . Universities and future clinicians will be able to use it to reproduce surgical scenes for their work . And industry can use the The Team