Executive Summary
The German-Canadian consortium NephroCAGE is cooperating to demonstrate the added value of artificial intelligence (AI) using the concrete clinical example of kidney transplantation. Inadequate kidney function requires regular dialysis: there are currently around 100,000 dialysis patients in Germany, and around half that number in Canada. Dialysis costs approximately 30-40k EUR per patient per year. In comparison, a kidney transplant costs around 15-20 thousand euros. In 2019, more than 2,100 kidney transplants were performed in Germany (German Organ Transplant Foundation) and more than 1,700 in Canada (Canadian Institute for Health in Canada). However, suitable donor organs are rare: in Germany, for example, there are more than 7,000 patients on a waiting list, and in Canada more than 3,000. Even after a transplant, there is a risk of complications that can lead to severe limitations in kidney function or, in the worst case, even total loss of the organ.
The consortium partners are creating a learning AI system to match organ donors and recipients even more precisely in advance (matching) and thus prevent risks in kidney transplants. To this end, clinical centers of excellence in both nations are contributing transplant data from the last ten years. They will be analyzed using AI learning techniques and combined together with a novel matching algorithm to create clinical prognostic models for kidney transplant patients. By using a federated learning approach, where the algorithms are executed at the location of the data, data protection is maintained and sensitive health data from both nations can serve as a common basis for clinical prognostic models for the first time. As a result, a clinical demonstrator will be created to serve the exploitation of the medical and technical innovations in the context of care, as well as a basis for exploitation and follow-on projects.
Project Partners
Canada
- The University of British Columbia, Faculty of Medicine, Department of Pathology and Laboratory Medicine, Vancouver, British Columbia
- Genome BC, Vancouver, British Columbia
- Genome Canada, Ontario, Ottawa
- McGill University Health Centre, Montréal, Québec
- Centre hospitalier de l’Université de Montréal, Québec
- Genome Quebec, Montréal, Québec
Germany
- PIRCHE AG, Berlin
- Charité – Universitätsmedizin Berlin, Medical Department,
Division of Nephrology and Internal Intensive Care Medicine, Berlin - Karlsruhe Institute of Technology, Critical Information Infrastructures, Karlsruhe
- Hasso Plattner Institute and AnalyzeGenomes.com
Publications
- Job Opportunity: Research Assistant (f/m/*) with focus on “Machine Learning und Clinical Prediction Models”
- Job Opportunity: Research Assistant (f/m/*) with focus on “Federated Machine Learning for Digital Health”
Research Publications
- Schapranow, M.-P.: NephroCAGE: Wie Künstliche Intelligenz bei Nierenversagen unterstützen kann. Gesundhyte: Forschung neu vernetzen. 15, 100—102 (2023).
- Schapranow, M.-P., Bayat, M., Rasheed, A., Naik, M., Graf, V., Schmidt, D., Budde, K., Cardinal, H., Sapir-Pichhadze, R., Fenninger, F., Sherwood, K., Keown, P., Günther, O., Pandl, K., Leiser, F., Thiebes, S., Sunyaev, A., Niemann, M., Schimanski, A., Klein, T.: NephroCAGE—German-Canadian Consortium on AI for Improved Kidney Transplantation Outcome: Protocol for an Algorithm Development and Validation Study. JMIR Res Protoc 2023. 12, (2023).
- Borchert, F., Llorca, I., Schapranow, M.-P.: Cross-Lingual Candidate Retrieval and Re-ranking for Biomedical Entity Linking. In: Arampatzis, A., Kanoulas, E., Tsikrika, T., Vrochidis, S., Giachanou, A., Li, D., Aliannejadi, M., Vlachos, M., Faggioli, G., en Ferro, N. (reds.) Experimental IR Meets Multilinguality, Multimodality, and Interaction. bll. 135–147. Springer Nature Switzerland, Cham (2023).
- Llorca, I., Borchert, F., Schapranow, M.-P.: A Meta-dataset of German Medical Corpora: Harmonization of Annotations and Cross-corpus NER Evaluation. Proceedings of the 5th Clinical Natural Language Processing Workshop. bll. 171–181. Association for Computational Linguistics, Toronto, Canada (2023).
- Kämmer, N., and Borchert, F., and Winkler, S., and de Melo, G., and Schapranow, M.-P.: Resolving Elliptical Compounds in German Medical Text. The 22nd Workshop on Biomedical Natural Language Processing and BioNLP Shared Tasks. bll. 292–305. Association for Computational Linguistics, Toronto, Canada (2023).
Events
- Join the NephroCAGE symposium on Nephrology Disease Coop. b/w Canada & Germany for Applied AI on May 20, 2021 4:30 pm
- 2021 German-Canadian Symposium on the Potential of AI for Health Innovations on October 20, 2021 5:00 pm
- Join us for AI in Medicine and Care on December 14, 2021 4:30 pm
- Join us for the 2nd International NephroCAGE Symposium on August 16, 2022 3:15 pm
- Join us for 2023 Data4Health Conference on June 20, 2023 9:00 am
Sponsors
The project is generously sponsored by the German Federal Ministry for Economic Affairs and Climate Action (2021-2022) as project of strategic interest.