Subscribe to RSS
DOI: 10.1055/s-0042-1760595
Leveraging medical-AI to speed up Cold Agglutinin Disease detection
Introduction Timely identification of rare diseases is often challenging. In Germany, approximately 4 million people are affected by rare conditions [1]. Systems based on artificial intelligence (AI), such as symptom checker applications, can help to detect such rare diseases.
We integrated the rare disease model CAD into our system. Our approach was to abstract medical knowledge both from literature research and by conducting a guided interview with a medical expert.
Method Ada’s medical experts carried out extensive literature research. Also, a workshop was held with an external clinical expert in CAD, in order to confirm the modeling strategy and provided additional clinical insights to be integrated. Once the disease model was created and integrated, it was tested by Ada’s medical experts, before being made available to users worldwide ([Fig. 1]).
Results The CAD model has been created and incorporated in Ada’s reasoning engine, available to users of Ada’s products.
In the first 30 days of the model released, in 48 cases CAD has been suggested among first 3 candidate (possible conditions associated with the reported presentations). Among cases with a feedback, 8 out of 9 were positive feedback (helpful). 1 case was unhelpful.
Conclusion Integrating CAD in a symptom checker system can help speed up disease detection, potentially providing quicker access to healthcare systems to ultimately improve health journeys.
Publication History
Article published online:
20 February 2023
© 2023. Thieme. All rights reserved.
Georg Thieme Verlag
Rüdigerstraße 14, 70469 Stuttgart, Germany
-
References
- 1 Eidt D, Frank M, Dr. Reimann A, Prof. Dr. Wagner T, Dr. Mittendorf T, Prof. Dr. von der Schulenburg J.-M “Maßnahmen zur Verbesserung der gesundheitlichen Situation von Menschen mit Seltenen Erkrankungen in Deutschland”, Bundesministeriums für Gesundheit, page 1, Hannover, Germany 2009