Project

RadioEye

RadioEye Logo
Keywords

The Autopilot in Diagnosing Misleading Radiology Cases Correctly

Pre Spin-Off

Project Website

Interpretation of radiological images is core in diagnostic radiology. However, the medical world has grown in terms of complexity and broadness. Furthermore, the workload of radiologists increased three-fold compared to the workforce (Statement by the BRC Radiologists, 2020), resulting in less time to interpret each radiological case.

In 10-35 % of daily cases, the radiologist consults various sources for reference to be able to interpret the image correctly. In most cases the radiologist ends up searching on radiological websites. These solutions are mainly text-based search functions and offer only a small amount of images that display only the standard appearances of the respective diseases. Finding the correct diagnosis is an extremely time-consuming and laborious process in daily clinical practice.

Katharina Erb-Eigner
Katharina Erb-Eigner
(Charité)

Project Lead

Finally, if the radiologist is not able to diagnose the lesion on the image, the radiology report just describes image features of the lesion. This will lead to a costly and potentially follow up examination, e.g. a risky biopsy – that delays the diagnosis, is a burden for the patient and results in high costs for the healthcare system due to hospitalization, surgery, and medical staff costs.

RadioEye closes this gap by offering diagnosis support with the help of a reference tool designed as an interactive case collection, providing curated information and a vast database of reports and radiological images of cases. RadioEye offers an AI-based image-search functionality to find similar cases based on image features alone and by ranking narrowing potential diseases down. The radiologist can swipe through images and compare them to the case at hand.

RadioEye has started with an eye and eye socket module and will expand its database to contain radiological images of the whole spectrum of diseases in the entire human body. A curated radiology database that covers the real-world variance of disease presentation together with a powerful image-search functionality is unique and will improve quality and efficiency in diagnostic radiology worldwide.

RadioEye is powered by an interdisciplinary team with years of experience in radiology and specialized radiology, AI and software development, as well as UX/UI design. The team also works closely with expert advisors in regulatory affairs, business and market access.

DHA

Project Lead

Katharina Erb-Eigner

„Well, the next step is going to be money acquisition. And to go internationally.“

Katharina Erb-Eigner

(Charité)