Cardiovascular diseases (CVD) are a major global health concern. In Switzerland, they are the third most common reason for hospitalisation and the leading cause of death. Echocardiography is the most common method to diagnose CVD, but it is complex and time-consuming, involving several manual steps. Moreover, it is fraught with a misdiagnosis rate of 16%. This creates a large demand for efficient and accurate diagnostic tools.
EchoRapp supports cardiologists in making detailed diagnoses from echocardiographic data and generating rule-based full-text reports. It has to date one of the most rigorous solutions to apply evidence-based diagnostic criteria from international guidelines and peer-reviewed studies. The project goal is to complement these capabilities with deep learning-based image analysis and better text generation to match cardiologists’ need for an efficient and accurate solution that covers the whole process. We aim to automatically extract quantitative measures from ultrasound imaging data, improve the sensitivity to detect rare and early conditions, and generate comprehensive full-text reports based on language modeling.
With the proposed project and partnership with ZHAW, MediRapp obtains the resources and know-how to add the necessary capabilities to EchoRapp in a short time, while leveraging its unique basis. This would significantly increase market acceptance and accelerate market entry.
EchoRapp is designed to integrate seamlessly into clinical practice of small and large institutions. Its integration into the Swiss healthcare ecosystem will help to streamline the echocardiography process and enable clinicians to dedicate more time and attention to patients. It also offers an effective solution to the expected shortage of qualified professionals. The efficiency increase in combination with the state-of-the-art diagnostic support will lead to significant cost savings and contribute to optimal patient outcomes.