IBM's Watson artificial intelligence technology will be used to support physicians at a hospital in Germany to analyze large volumes of clinical data and help doctors diagnose patients.
In an agreement announced on 18 October, IBM said it will perform a pilot project with Watson at Rhön-Klinikum's Centre for Undiagnosed and Rare Diseases facility at University Hospital Marburg. Many patients being seen at the center have long medical histories that include unstructured data, like lab tests, clinical reports, drug prescriptions, and radiology and pathology reports.
The amount of patient data that needs to be reviewed can be overwhelming, so medical and technical scientists from Rhön-Klinikum and IBM will work together to develop a cognitive assistance system for diagnosing rare diseases, with the goal of helping physicians make decisions quickly and safely.
Patients will fill out digital questionnaires that will anonymize their data and send it to IBM servers in the cloud, where relevant data will be extracted for clinicians. The algorithm will use a natural language processing algorithm for medical terminology developed by IBM Global Business Services to correlate the German-language questionnaire to English-based medical data to provide a differential diagnosis.
Experience gained from the pilot project at University Hospital will be analyzed and used across other hospitals throughout the group. In addition to helping doctors diagnose patients, the system is expected to support the patient admission process and provide direct patient routing, preventing sending patients to the wrong specialists or providing wrong treatment plans.
