From patient data to use cases of health AI application and integration

mibe000315 10.3205/mibe000315 urn:nbn:de:0183-mibe0003153 Editorial From patient data to use cases of health AI application and integration Von Patientendaten bis hin zu Anwendungsfällen für den Einsatz und die Integration von KI im Gesundheitswesen Hübner Hübner Ursula U Prof. Dr.

Hochschule Osnabrück, Albrechtstr. 30, 49076 Osnabrück, GermanyResearch Center for Health and Social Informatics, Hochschule Osnabrück University of Applied Sciences, Osnabrück, Germany

u.huebner@hs-osnabrueck.de author German Medical Science GMS Publishing House

Düsseldorf

610 ISCB GMDS 2026 20260922 engl This is an Open Access article distributed under the terms of the Creative Commons Attribution 4.0 License. Dieser Artikel ist ein Open-Access-Artikel und steht unter den Lizenzbedingungen der Creative Commons Attribution 4.0 License (Namensnennung). 1860-9171 22 GMS Medizinische Informatik, Biometrie und Epidemiologie GMS Med Inform Biom Epidemiol 17 EditorialThis special issue combines six papers that were submitted to the joint ISCB and GMDS conference 2026 in Freiburg and got accepted in the combined review process for the conference plus the MIBE journal. Following the conference theme “BIG DATA, small data, Your Data – transitions in the age of biomedical AI” one paper addresses health data modelling that allows for technical, syntactic and semantic data re-usability. The other five papers focus on the application of AI methods and their integration to support patient care and research use cases. The study of Elgert et al. shows that health data modelling for improved interoperability requires a thorough strategic model planning and implementation that cover the entire data lifecycle including data re-use. Profound domain and standards knowledge supported by guidelines such as eHealth European Interoperability Framework (ReEIF) could aid project planners and data stewards the authors argue.While patient data re-usage plays an undoubted role in AI model training, AI model application needs to be explored and evaluated to demonstrate its feasibility and the conditions under which AI applications can work properly. Reichenpfader and Dennstädt tested Semantic Retrieval-Augmented Translation (S-RAT) for translating clinical texts from radiology. The ontology-guided framework applied in the study made use of concept retrieval from the Unified Medical Language System (UMLS) and SNOMED CT to improve large language model prompts. The authors conclude that S-RAT demonstrated the potential as a locally deployed, privacy preserving methodology given the multilingual limitations of the ontologies integrated were overcome. Diaz explored the feasibility of coding texts for qualitative data analysis with the help of ChatGPT employing p</PlainText></TextGroup>rompting strategies of increasing complexity. She found the less structured the text was the better more complex prompting performed. She recommended a combination of AI-based and classical tools to achieve the best results.</Pgraph><Pgraph>Yüzüncüoglu and co-authors <TextLink reference="5"></TextLink> developed an information extraction pipeline for unstructured documentation of liver cancer cases to support decision making in tumour board discussions. The performance of rule-based methods (regular expressions), transformer-based models, and an exploratory large language model (LLM) were contrasted. The findings hint at a hybrid approach – so the authors – combining transformer-based and rule-based methods to yield the best results.</Pgraph><Pgraph>Pelka and colleagues presented results from two studies on integrating AI into clinical workflows. In the first study, Pelka, Kamzol and colleagues <TextLink reference="6"></TextLink> employed two different methods to mediate between heterogeneous clinical PACS implementations and federated AI services. Tested in a clinical environment, they concluded that a mediation layer integrating standards was necessary to bridge the gap between the different systems sufficiently well. Addressing the need for transparency and traceability according to the EU AI Act, Pelka, Eil and colleagues <TextLink reference="7"></TextLink> tested the Open Medical Inference (OMI) Gateway for FHIR and DICOM environments to connect with AI inference backends. They argued that the OMI integrated governance mechanism and the audit trail generated via the FHIR based validation pipeline provide an AI Act compatible infrastructure for medical AI inference.</Pgraph><Pgraph>These studies highlight the opportunities, conditions and current limitations of approaches harnessing AI to improve patient care and research. While technical barriers still need to be overcome, questions on tangible benefits for patients and healthcare providers remain to be investigated. </Pgraph></TextBlock> <TextBlock name="Note" linked="yes"> <MainHeadline>Note</MainHeadline><SubHeadline>Competing interests</SubHeadline><Pgraph>The author declares that she has no competing interests.</Pgraph></TextBlock> <References linked="yes"> <Reference refNo="1"> <RefAuthor>Anonym</RefAuthor> <RefTitle></RefTitle> <RefYear></RefYear> <RefBookTitle>47th Annual Conference of the International Society for Clinical Biostatistics, 71st Annual Conference of the German Association for Medical Informatics, Biometry and Epidemiology (ISCB GMDS 2026); 2026 Sep 27 - Oct 1; Freiburg, Germany</RefBookTitle> <RefPage></RefPage> <RefTotal>47th Annual Conference of the International Society for Clinical Biostatistics, 71st Annual Conference of the German Association for Medical Informatics, Biometry and Epidemiology (ISCB GMDS 2026); 2026 Sep 27 - Oct 1; Freiburg, Germany.</RefTotal> </Reference> <Reference refNo="2"> <RefAuthor>Elgert L</RefAuthor> <RefAuthor>Richter J</RefAuthor> <RefAuthor>Flessner J</RefAuthor> <RefAuthor>Liehr D</RefAuthor> <RefAuthor>Katzensteiner M</RefAuthor> <RefAuthor>Pape M</RefAuthor> <RefAuthor>Schulz N</RefAuthor> <RefAuthor>Krefting D</RefAuthor> <RefAuthor>Koppelin F</RefAuthor> <RefAuthor>Bott OJ</RefAuthor> <RefAuthor>Wolf KH</RefAuthor> <RefTitle>Von (Daten-)Standards zu Interoperabilität: Empirische Einsichten zu guter Gesundheitsdatenmodellierung</RefTitle> <RefYear>2026</RefYear> <RefJournal>GMS Med Inform Biom Epidemiol</RefJournal> <RefPage>Doc16</RefPage> <RefTotal>Elgert L, Richter J, Flessner J, Liehr D, Katzensteiner M, Pape M, Schulz N, Krefting D, Koppelin F, Bott OJ, Wolf KH. Von (Daten-)Standards zu Interoperabilität: Empirische Einsichten zu guter Gesundheitsdatenmodellierung. GMS Med Inform Biom Epidemiol. 2026;22:Doc16. DOI: 10.3205/mibe000314</RefTotal> <RefLink>https://doi.org/10.3205/mibe000314</RefLink> </Reference> <Reference refNo="3"> <RefAuthor>Reichenpfader D</RefAuthor> <RefAuthor>Dennstädt F</RefAuthor> <RefTitle>Semantic retrieval-augmented translation for radiology reports using small language models: Algorithm development and evaluation</RefTitle> <RefYear>2026</RefYear> <RefJournal>GMS Med Inform Biom Epidemiol</RefJournal> <RefPage>Doc15</RefPage> <RefTotal>Reichenpfader D, Dennstädt F. Semantic retrieval-augmented translation for radiology reports using small language models: Algorithm development and evaluation. GMS Med Inform Biom Epidemiol. 2026;22:Doc15. DOI: 10.3205/mibe000313</RefTotal> <RefLink>https://doi.org/10.3205/mibe000313</RefLink> </Reference> <Reference refNo="4"> <RefAuthor>Diaz M</RefAuthor> <RefTitle>Prompting strategies for coding in AI-assisted qualitative data analysis</RefTitle> <RefYear>2026</RefYear> <RefJournal>GMS Med Inform Biom Epidemiol</RefJournal> <RefPage>Doc14</RefPage> <RefTotal>Diaz M. Prompting strategies for coding in AI-assisted qualitative data analysis. GMS Med Inform Biom Epidemiol. 2026;22:Doc14. DOI: 10.3205/mibe000312</RefTotal> <RefLink>https://doi.org/10.3205/mibe000312</RefLink> </Reference> <Reference refNo="5"> <RefAuthor>Yüzüncüoglu I</RefAuthor> <RefAuthor>Chen Y</RefAuthor> <RefAuthor>Lüser AA</RefAuthor> <RefAuthor>Ramasetti NS</RefAuthor> <RefAuthor>Oehring R</RefAuthor> <RefAuthor>Krezien F</RefAuthor> <RefAuthor>Thomas P</RefAuthor> <RefAuthor>Möller S</RefAuthor> <RefAuthor>Roller R</RefAuthor> <RefTitle>Automatic extraction of clinically relevant parameters for tumor board decision support</RefTitle> <RefYear>2026</RefYear> <RefJournal>GMS Med Inform Biom Epidemiol</RefJournal> <RefPage>Doc13</RefPage> <RefTotal>Yüzüncüoglu I, Chen Y, Lüser AA, Ramasetti NS, Oehring R, Krezien F, Thomas P, Möller S, Roller R. Automatic extraction of clinically relevant parameters for tumor board decision support. GMS Med Inform Biom Epidemiol. 2026;22:Doc13. DOI: 10.3205/mibe000311</RefTotal> <RefLink>https://doi.org/10.3205/mibe000311</RefLink> </Reference> <Reference refNo="6"> <RefAuthor>Pelka O</RefAuthor> <RefAuthor>Kamzol NA</RefAuthor> <RefAuthor>Manjunatha K</RefAuthor> <RefAuthor>Singh N</RefAuthor> <RefAuthor>Omeirat J</RefAuthor> <RefAuthor>Nensa F</RefAuthor> <RefTitle>A standards-based infrastructure for distributed AI inference in medical imaging</RefTitle> <RefYear>2026</RefYear> <RefJournal>GMS Med Inform Biom Epidemiol</RefJournal> <RefPage>Doc12</RefPage> <RefTotal>Pelka O, Kamzol NA, Manjunatha K, Singh N, Omeirat J, Nensa F. A standards-based infrastructure for distributed AI inference in medical imaging. GMS Med Inform Biom Epidemiol. 2026;22:Doc12. DOI: 10.3205/mibe000310</RefTotal> <RefLink>https://doi.org/10.3205/mibe000310</RefLink> </Reference> <Reference refNo="7"> <RefAuthor>Pelka O</RefAuthor> <RefAuthor>Eil J</RefAuthor> <RefAuthor>Manjunatha K</RefAuthor> <RefAuthor>Singh N</RefAuthor> <RefAuthor>Omeirat J</RefAuthor> <RefAuthor>Sigle S</RefAuthor> <RefAuthor>Mathes G</RefAuthor> <RefAuthor>Schweizer ST</RefAuthor> <RefAuthor>Stump SR</RefAuthor> <RefAuthor>Girdziunaite G</RefAuthor> <RefAuthor>Nensa F</RefAuthor> <RefTitle>Interoperable clinical data integration for distributed medical AI: The Open Medical Inference Gateway</RefTitle> <RefYear>2026</RefYear> <RefJournal>GMS Med Inform Biom Epidemiol</RefJournal> <RefPage>Doc11</RefPage> <RefTotal>Pelka O, Eil J, Manjunatha K, Singh N, Omeirat J, Sigle S, Mathes G, Schweizer ST, Stump SR, Girdziunaite G, Nensa F. Interoperable clinical data integration for distributed medical AI: The Open Medical Inference Gateway. GMS Med Inform Biom Epidemiol. 2026;22:Doc11. DOI: 10.3205/mibe000309</RefTotal> <RefLink>https://doi.org/10.3205/mibe000309</RefLink> </Reference> </References> <Media> <Tables> <NoOfTables>0</NoOfTables> </Tables> <Figures> <NoOfPictures>0</NoOfPictures> </Figures> <InlineFigures> <NoOfPictures>0</NoOfPictures> </InlineFigures> <Attachments> <NoOfAttachments>0</NoOfAttachments> </Attachments> </Media> </OrigData> </GmsArticle>