Proof of concept detects and pinpoints anomalies in ECG data in near real-time on edge devices

The German technology company ITK Engineering, headquartered at Rülzheim, Rhineland-Palatine, and the Japanese startup Hacarus have jointly developed a proof of concept for a personalized patient monitoring system, serving to evaluate ECG signals. It not only evaluates anomalies in a single patient heart’s ECG in near real-time, but also provides at-a-glance comparisons with the expected normal ECG curve. Missing interpretability has been one major acceptance hurdle for classic deep learning approaches in medical applications as the decision-making process of AI has been a black box up to now. The novel AI technology is visualized in a form that can be interpreted by doctors and supports them in fast and accurate judgement.

The proof of concept was co-engineered in Japan and Germany, integrated by ITK Engineering based on 25 years of systems engineering experience in medical grade connected systems including wearables, mobile medical apps, algorithms and secure interfaces.