AI+ML in Healthcare
This was the key message from decision makers, users, and developers at the German AI + ML Summit organized by InterSystems earlier this year: clean, healthy data and interoperability provide the foundation for the creation of new applications in healthcare as well as across further industries. Interoperable data platforms and integrated tools act as enablers, with data privacy and ethical questions playing important roles. Key topics discussed at the Summit included making data available from clinical practice for research, decentralization and GDPR compatibility, and clinical use cases for AI at SMITH (Smart Medical Information Technology for Healthcare) Consortium.
“To bridge the fields of IT and Medicine – this is one of the key challenges in the context of achieving progress in healthcare, with the support of technologies”, underlined Dr. Danny Ammon when speaking about patient data and its use for AI applications. Dr. Ammon is a member of the Board of HL7 Germany and Caretaker of Quality, Research, and Public Health of IHE Germany. He also serves as Head of the Interoperability Working Group at SMITH and is the Head of the Data Integration Center at Jena University Hospital.
Speakers discussed the importance of AI applications in healthcare to promote effective decision-making. “The volume of data in healthcare is growing 60 percent faster than in other industries. Humans can only consider around 7 data points for every decision; therefore, we also need AI algorithms in order to support decision making processes in the delivery of care as well as in research. The SMITH use cases are perfect examples for the rapid delivery on requirements from clinical practice,” underlined Volker Hofmann, Manager of Healthcare at InterSystems. “We consider ourselves fortunate to be working jointly with our implementation partner, März Internetwork Services AG, in this challenging ecosystem.”
At the core of the SMITH reference architecture is InterSystems HealthShare. HealthShare provides a foundation which allows for the integration of data curators, metadata, identification, and authentication as well as patient consent. As an implementation partner, März is in the process of configuring and expanding existing HealthShare functions in order to reach the specific goals set by SMITH consortium data integration centers. Researchers will be able to access data in those centers via smart service platforms to suggest research projects and proofs-of-concept and evaluate whether defined sites should be part of a trial depending on their number of patients with respective conditions.
SMITH is one of four consortia funded by the German Medical Informatics Initiative (MII) to create an alliance of universities, university hospitals, research institutions and IT companies. SMITH’s goals are to establish Data Integration Centers at each SMITH partner hospital and to implement use cases which demonstrate the usefulness of the approach such as HELP, a decision support system for blood stream infections or ASIC which focusses on patients in ICUs who have severe respiratory conditions. All German university hospitals are part of this nationwide medical informatics initiative.
“This information-based bridging from bedside to bench and back to enable innovation is a task which can only be mastered through the commitment to cross-sector communication by everybody involved,” stated Andreas Kumbroch, Software Development Board Member at März. He continued, noting that “the initiative builds on firm commitments, creating sustainability for innovation – phased according to specific milestones.”
This story originally appeared the HealthShare Connections News Flash No.6 – June 21, 2021