The healthcare sector generates countless data every day – information about diagnoses, treatments and disease progression as well as information about molecular details and metabolic processes. “We have a vast treasure trove of data,” says Professor Sylvia Thun, who has been heading the BIH Core Unit eHealth and Interoperability since 2018. “It would be unethical not to use these data.”
Inconsistent data hinders insights
Yet putting such data to use remains challenging. Data from research and healthcare are recorded differently in every laboratory and every hospital. They are formulated differently, formatted differently and stored in different software systems, if not on paper. “That’s why it’s not possible to compare the data of all breast cancer patients or all diabetes patients,” regrets Thun. However, such a comparison could help identify the most effective personalized treatment, discover rare severe side effects of drugs, or reveal correlations between genetic information and disease symptoms.
Communication standards in healthcare
“We need communication standards in the healthcare system,” demands the physician and medical IT specialist. “So we have made it our mission to process data from medical care, from molecular biological findings, from tissue and blood examinations, and from pathology reports in a structured way.” This requires physicians and scientists to enter their readings and diagnoses in a standardized “global language.” “To make this possible we must carry out research projects and create innovative software so that no additional work has to be done. Data scientists will process the data in a beneficial way so patients should feel the added value very soon.”
Uniform data handling also has the support of policymakers: through the Hospital Future Act of the German Federal Ministry of Health (BMG), whose DigitalRadar project to evaluate the digital maturity of hospitals is led by Thun – and through the German Federal Ministry of Education and Research (BMBF). “Under the BMBF’s Medical Informatics Initiative, almost all German university hospitals are setting up data integration centers tasked with jointly developing concepts for documenting and sharing data in compliance with data protection regulations. We need to seize this opportunity,” says Thun. To achieve this the parties involved will introduce the FAIR principles, which state that scientific data should be findable, accessible, interoperable and reusable. And Thun and her team are going even further: Next they want to standardize data via smartphones. This should enable patients to regularly send information to their doctors so that they can monitor how successful the therapy actually was and how their patients’ health is changing over time.
Standardized data as a prerequisite for translation
Going forward, Professor Christopher Baum, Chair of the BIH Board of Directors and Chief Translational Research Officer of Charité – Universitätsmedizin Berlin, plans to integrate the Professorship in Digital Medicine and Interoperability into the BIH thematic area of Medical Data Science. “With the appointment of Sylvia Thun, we have gained an outstanding expert in the field of biomedical data standardization for the BIH. Standardized data is a prerequisite for the benefit driven personalized medicine that we are pursuing at the BIH. This will support in particular the joint Health Data Platform (HDP) of the BIH and Charité.”
Thun is currently paying special attention to data from research into the SARS-CoV-2 coronavirus and to data from the treatment of COVID-19. Scientists are studying how to keep the infection rate low, why some people get severely ill but others have only mild symptoms, how to best treat the COVID-19 disease, and how long vaccination protection lasts. Institutes and universities as well as start-up companies and government agencies are gathering data, results, and information, which are most valuable when they are shared. “By using standardized languages like FHIR, SNOMED, and LOINC, the data can not only be clearly interpreted, but also pooled internationally and used for research purposes.”
Consensus data set of COVID-19 patients
The team led by Thun created a “consensus data set” of COVID-19 patients for the National Research Network of University Hospitals Against COVID-19. It contains all relevant information, starting with personal data like age, gender, height, and blood pressure, followed by lab results like creatinine and D-dimer levels, risk factors, medication use, as well as symptoms and therapeutic procedures performed.
“Standardization is of course just as important in oncology, cardiovascular medicine or diabetes,” says Thun. “After all, such diseases did not disappear during the coronavirus pandemic.” But she is especially passionate about so-called orphan diseases, which are rare diseases for which there are few or no treatments available. “Digital networking can be extremely helpful,” explains Thun. “Especially for medical diagnoses that occur perhaps just 100 times throughout Germany.” In the CORD-MI project, the team is working to ensure that advances in digitalization also benefit the centers for rare diseases at university hospitals.
More women needed in digital medicine
And then there’s another issue that’s very important to the busy physician-scientist: gender equality in digital medicine. Women are underrepresented both in the data on which AI algorithms are based and among AI developers. So together with the President of the German Association of Women Physicians, Dr. Christiane Groß, she founded the network #SHEALTH with the aim of increasing the visibility of women in medicine, especially digital medicine. “We need more female medical IT professionals and female health data scientists; the sector definitely has some catching up to do,” explains Thun, who has even co-edited a book on the subject.
Sylvia Thun studied physical and biomedical engineering as well as human medicine at RWTH Aachen University, where she also earned a medical doctorate in radiological imaging in 2001. In addition, she received the title of medical informatics specialist from the North Rhine Medical Association and is certified by the German Association for Medical Informatics, Biometry and Epidemiology (GMDS). She initially conducted research for the German Federal Ministry of Health and for the German Institute for Medical Documentation and Information (DIMDI) in Cologne before being appointed in 2011 to a Professorship in Information and Communication Technologies at the Niederrhein University of Applied Sciences (HSNR) in Krefeld. In 2014 Sylvia Thun was named one of Germany’s “digital minds” by the Federal Ministry of Education and Research (BMBF). In 2017, while helping to launch the BMBF’s Medical Informatics Initiative (MII), Thun was appointed Visiting Professor at the BIH. Since then she has been Director of the Core Unit eHealth and Interoperability, which includes research groups and about 20 research fellows.