AI in medicine

The following selection of projects with significant Fraunhofer involvement represents our role in transforming medical care. This is where the Fraunhofer-Gesellschaft is addressing the challenges of the healthcare sector — with the aim of consistently improving people's quality of life.

In recent years, there has been an exponential growth in the generation of data that could be harnessed for use in healthcare delivery and research. These data include readouts generated by digital technologies, patient reported outcome and experience measures, and results from clinical trials and routine clinical care. However, accessing, integrating and analysing these data to maximize their value for patient care and research is extremely challenging.

IDERHA aims to create a scalable platform for the seamless integration or linkage of diverse data at scale to support healthcare professionals, patients, and researchers with new capabilities to improve patient outcomes by better treating and manage disease, enable personalized care, bolster research and innovation, based on the development of common standards and practices – reducing current disconnected information silos. This enables more personalized health care along the patient's journey via data-driven tools and solutions.

https://www.iderha.org/

https://www.isst.fraunhofer.de/de/abteilungen/healthcare/projekte/IDERHA.html

https://www.itmp.fraunhofer.de/en/press/IDERHA_start.html

 More and more people have Alzheimer's disease (AD) and related dementia. AD is associated with a high level of suffering for those affected and rising costs for the healthcare system. By 2023, more than seven million people will be living with dementia in the European Union. Some progress has been made recently in the search for effective treatments, and it has been proven that treatment in the early stages of the disease is most effective. Therefore, there is a need for scalable, cost-effective diagnostic markers, tools, and procedures to identify individuals at increased risk to select the most promising personalized measures to prevent or delay dementia.

https://www.helse-stavanger.no/en/predictom/predictom-secures-21-million-investment-to-pioneer-early-alzheimers-detection/

https://www.scai.fraunhofer.de/de/projekte/PREDICTOM.html

Does a SARS-CoV-2 infection elevate the risk of dementia? This pressing medical question forms the crux of the EU-funded COMMUTE project, which stands for "COMmorbidity Mechanisms UTilized in HealthcarE." The project aims to unravel the underlying mechanisms linking COVID-19 infections with neurodegenerative diseases such as Alzheimer's and Parkinson's.

Fraunhofer SCAI coordinates the COMMUTE project, backed by a grant from the European Commission. Over the next four years, an interdisciplinary team of top-tier experts will explore whether COVID-19 infections increase the risk of acquiring neurodegenerative diseases. An innovative AI-driven system is being developed to provide tailored risk assessments for individuals who have recovered from COVID-19.

https://www.commute-project.eu/en/about.html

https://www.scai.fraunhofer.de/de/projekte/COMMUTE.html

Der demografische Wandel stellt unsere Gesellschaft vor enorme Herausforderungen. Insbesondere das Gesundheitssystem wird durch die Verrentungswelle der »Babyboomer« und die gleichzeitig steigende Nachfrage nach medizinischen Dienstleistungen doppelt belastet. Schon jetzt herrscht ein akuter Fachkräftemangel, der sich in den kommenden Jahren dramatisch verschärfen wird. Prognosen zufolge werden bis 2035 etwa 1,8 Millionen Stellen im Gesundheitswesen unbesetzt bleiben.

»Ohne Gegenmaßnahmen steuern wir auf eine buchstäblich lebensgefährliche Überlastung des Gesundheitssystems zu«, warnt Prof. Thorsten Buzug, Direktor des Instituts für Medizintechnik von der Universität zu Lübeck und Sprecher der Hanse Innovation Community GRANNI. »Unser partizipativer Forschungsansatz entwickelt zusammen mit klinischen Partnern Methoden für die Gerontologie 2.0, um den Druck auf das Gesundheitssystem nachhaltig zu senken.«

Das Projekt GRANNI verfolgt einen integrativen Ansatz, der sowohl technologische als auch gesellschaftliche Aspekte berücksichtigt. »Auf der Ebene des Gesundheitssystems müssen die Prozesse konsequent digitalisiert werden, um dem Fachkräftemangel durch Automatisierung und den Einsatz von KI effektiv zu begegnen«, erklärt Prof. Philipp Rostalski, Direktor am Fraunhofer IMTE. »Auf der Ebene der alternden Patientinnen und Patienten besteht ein großer Bedarf an innovativen Lösungen zur Förderung eines Alterns in Würde und Autonomie.«

https://www.imte.fraunhofer.de/de/presse-medien/pressemitteilungen-aktuelles/pm---luebecker-forschungslandschaft-erhaelt-foerdergelder-in-hoe.html

https://www.bmbf.de/bmbf/de/forschung/datipilot/datipilot.html

An international team started the research project CERTAINTY in December 2023. Together with partners from science, industry and the healthcare sector, the project team led by the Fraunhofer Institute for Cell Therapy and Immunology IZI aims to develop a virtual twin that will improve treatment with personalized cancer immunotherapies in the future.

In recent years, cancer immunotherapies have established themselves as a further pillar of medical oncology alongside traditional treatment options (surgery, radiotherapy and chemotherapy). The advantages of personalized treatment approaches, such as CAR-T cell therapy, also include more precise phenotyping of individual patients.

Numerous clinical, imaging, molecular and cell analytical data are collected and processed for each patient for diagnosis, treatment decisions and follow-up. The totality of all patient data within a clinical picture harbors enormous potential for improving diagnosis and therapy for future patients. One approach to realizing this potential is the concept of the virtual twin. This involves merging certain molecular and cellular characteristics of a person and their clinical progression data into a digital representation, which is regularly updated using a series of data variables. Based on comparative data from patients with similar characteristics, the virtual twin can then be used to simulate prognoses regarding the course of the disease or various treatment options.

https://www.certainty-virtualtwin.eu/

https://www.izi.fraunhofer.de/en/press/press-releases/virtual-twin-to-improve-treatment-with-cancer-immunotherapies.html

In einem von der Universitätsmedizin Essen angeführten Konsortium erarbeitet ein Team aus Wissenschaftler*innen der Fraunhofer-Institute für Intelligente Analyse- und Informationssysteme IAIS und für Digitale Medizin MEVIS, der RWTH Aachen und der TU Dortmund zusammen mit Expert*innen der m.Doc GmbH und der GSG Consulting GmbH Konzepte und Lösungen, wie Krankenhäuser aus NRW in sogenannte »Smart Hospitals« transformiert werden können. Der Förderbescheid in Höhe von rund 14 Millionen Euro wurde heute von Digitalminister Prof. Dr. Andreas Pinkwart an Dr. Anke Diehl, Chief Transformation Officer der Universitätsmedizin Essen, und Prof. Dr. Stefan Wrobel, Institutsleiter des Fraunhofer IAIS, überreicht.

https://smarthospital.nrw/

https://www.iais.fraunhofer.de/de/presse/presseinformationen/presseinformationen-2021/presseinformation-210225.html

Im Projekt wird eine Software mit offenen Schnittstellen entwickelt, um Schmerzen bei Versuchstieren erfassen und beurteilen zu können. Dies erfolgt auf Basis der sogenannten Grimace Scale (Schmerzgrimassen) auf Basis von KI-Methoden. Hierzu muss ein relevantes Tiermodell erstellt und mit verschiedenen Mausstämmen (mit verschiedenen Fellfärbungen) simuliert werden. Dabei werden einzelne Beobachtungen in eine konsistente, durchgängige und objektiv überprüfbare Methode überführt.

Parallel wird der Grimace Scale von geschulten Wissenschaftlern am Fraunhofer Institut für Zelltherapie und Immunologie analog erhoben, um die Funktionalität der Software zu überprüfen. Grundlegend ist hier, die visuellen Daten von Mäusen zu erfassen, die Schmerzgrimassen zeigen. Anschließend werden diese Merkmale mit einem vorgegebenen Satz von Schmerzindikatoren verglichen und abgestimmt und die Schwere der Schmerzen auf der Grundlage ihrer Relevanz für die identifizierten Schmerzgrimassen eingestuft.

https://www.softwaresysteme.dlrpt.de/media/content/01IS23038_Projektblatt_KIGRIMACE.pdf

Funded by the Innovative Medicines Initiative (IMI), the Fraunhofer Cluster of Excellence Immune-Mediated Diseases CIMD, together with 25 European partners from research, pharmaceutical companies, SME’s and patient organizations, is researching a poorly understood disease that affects millions of people. By looking into the disease mechanisms of psoriatic arthritis, the 26 European partners collaborating in the new research project HIPPOCRATES aim at improving diagnostic and therapeutic options for patients living with this condition. Through gaining a better understanding of the complex interplay between clinical and environmental factors, genotype and molecular pathways, the team aims to enable earlier diagnosis and a more accurate prediction of disease progression. This will revolutionize treatment and deliver profound patient benefits.

https://www.hippocrates-imi.eu/

https://www.itmp.fraunhofer.de/en/press/hippocrates.html

As one of the largest university hospitals in Germany with eight sponsors and over 600,000 treatments annually, the University Hospital of the Ruhr University Bochum (UK RUB) has enormous potential for data-driven clinical research. Therefore, under the leadership of the Faculty of Medicine of the RUB, the responsible bodies of the UK RUB are striving for a cooperative research data management within the framework of a joint IT umbrella strategy in order to strengthen research and teaching through structured and interoperable data preparation and provision.

https://www.isst.fraunhofer.de/en/departments/healthcare/projects/RUB-HCMS.html

The »Trusted Ecosystem of Applied Medical Data eXchange (TEAM-X)« project researches and implements solutions to make health data, which is sometimes difficult to access, more easily available to patients, doctors and nursing staff.

The primary goal is to establish a protected and trustworthy digital data ecosystem based on the Gaia-X infrastructure for the development of data-driven business models, products and services as the basis for forward-looking healthcare.

Patients retain control over their data at all times and decide for themselves who has access and for what purpose.

https://www.iis.fraunhofer.de/en/ff/sse/health/mobile-health-lab/team-x.html

https://project-team-x.eu/

Damit Krankenhäuser ihre Leistungen, die bei der Behandlung von Patient*innen anfallen, bei den Krankenkassen abrechnen können, müssen diese kodiert werden. Für die Erlössicherung der Kliniken ist dieser Prozess essenziell – er ist jedoch personalintensiv, zeitaufwendig und mitunter fehleranfällig. Das Fraunhofer-Institut für Intelligente Analyse- und Informationssysteme IAIS hat gemeinsam mit der GSG Consulting GmbH die KI-gestützte Software RICO entwickelt, die das Personal im Kodierprozess unterstützt. RICO wurde mit höchsten Datenschutz-Standards entwickelt, ist ohne zusätzlichen Aufwand sofort betriebsbereit und befindet sich bereits in mehreren Krankenhäusern im Einsatz.

https://www.dedalus.com/dach/de/our-offer/products/rico/

https://www.iais.fraunhofer.de/de/presse/presseinformationen/presseinformationen-archiv/presseinformationen-2020/presseinformation-200714.html

Künstliche Intelligenz (KI) gewinnt im Gesundheitswesen an Aufmerksamkeit und Bedeutung. KI kann dabei helfen, mühsame Routineaufgaben zu erleichtern – auch die Personalbedarfsprognose im Krankenhaus? In einem gemeinsamen Projekt mit ATOSS Software, der Universitätsmedizin Mainz und dem Ökosystem für innovative Gesundheit Flying Health entwickelt das Fraunhofer-Institut für Kognitive Systeme IKS einen KI-basierten Lösungsansatz, um den zukünftigen Personalbedarf auf einer Station der Universitätsmedizin Mainz vorherzusagen. Dabei kommen hochmoderne State-of-the-Art-Modelle aus dem Bereich der Zeitreihen-Prognose zum Einsatz, um die Genauigkeit und Zuverlässigkeit der Vorhersagen sicherzustellen.

https://safe-intelligence.fraunhofer.de/en/articles/healthcare-hackathon-2023-ai-and-hospital-workforce-planning

The goal of this EU-funded research project is to revolutionize AI in healthcare through the use of Swarm Learning (SL).It aims to overcome the obstacles of data collection in healthcare by utilising SL, where partners work together to train AI models without the need to share any personal patient data. Thereby, ODELIA will break data sharing boundaries and accelerate the scale-up of medical AI in Europe for the benefit of European citizens, patients, and clinicians.

https://odelia.ai/

https://odelia.ai/consortium/mevis/

Hospital staff usually suffer from enormous workloads and staff shortages, especially in intensive care units. This has consequences for patients, particularly when facing the prospect of multi-organ failure. In the future, Artificial Intelligence (AI) aims to enhance the application of a clinical multi-organ support system to further optimize patient treatment.

https://safe-intelligence.fraunhofer.de/en/articles/ai-intensive-care

Bei der Lösung komplexer medizinischer Fragestellungen gewinnen intelligente, vernetzte Implantate immer mehr an Bedeutung. Derzeit zur Verfügung stehende Systeme sind für den Nutzer oft nicht transparent und können von diesem nicht selbst bedient werden. Zukünftige Systeme könnten neben einem lebenslangen Einsatz die Möglichkeit der unmittelbaren Einflussnahme der Patienten auf ihre individuelle Situation stärker in den Vordergrund stellen.

https://www.interaktive-technologien.de/projekte/intakt

http://intakt-projekt.de/

https://www.ibmt.fraunhofer.de/de/ibmt-presse-uebersicht-2017/presse-ausgezeichnete-orte-INTAKT-2017-06-26.html

Point-and-click prevention, diagnostics and treatment: That is the vision behind the MED²ICIN lighthouse project. Digital twins have already been widely adopted in many industries. The development of digital models of patients has the potential to revolutionize the healthcare sector. The adoption of innovative digital solutions throughout the treatment process can improve patient care, and also pave the way for treatment that is more targeted, efficacious – and therefore more cost-effective.

The support system for decision-making developed as part of the MED²ICIN project should increase the treatment success rate. It supports physicians in their decision-making process by pooling all of the information on an individual patient and comparing this to that of cohorts made up of similar individuals. As well as helping to select the best option for therapy, this solution reduces treatment time and costs.

https://www.igd.fraunhofer.de/de/media-center/presse/digitales-patientenmodell-unterstuetzt-behandelnde-bei-entscheidungsfindung-und-reduziert-kosten.html

https://websites.fraunhofer.de/med2icin-en/about-the-project/

Trend-setting topics in medical technology are robot-assisted and (partially) automated surgical interventions with networked systems and fused information. Increasing automation and increased use of AI-based assistance systems are predicted for medical interventions. Furthermore, an operating theatre is already a high-tech place, but it is equipped with many isolated devices from different manufacturers. This creates technical and regulatory hurdles that make the networked collection, documentation and utilisation of all accruing information difficult. As a result, a great potential for improved individual patient care is currently being lost.

Within the framework of LIROS, a unique research centre for robot-assisted surgery is therefore being created at Fraunhofer IMTE with a realistic operating theatre environment, modern high-end equipment and individual anatomical patient models. The focus is on the optimisation and personalisation of training, the use of imaging techniques for intraoperative navigation, the networking of medical technology devices and the investigation of usability aspects to increase user-friendliness and safety in the operating theatre.

https://www.imte.fraunhofer.de/en/researchfields/liros.html

https://www.imte.fraunhofer.de/de/Kompetenzfelder/Medizintechnik/Medizinische-Robotik-und-Training.html

Quantum computing has the potential to train artificial intelligence in medical diagnostics more efficiently. This could make diagnoses more accurate even when there is little data available. In the future, medical professionals expect to see improvements in the screening, diagnosis and progress monitoring of brain tumors, for example.

https://www.iks.fraunhofer.de/en/projects/bayqs-quanten-security-data-science.html

https://safe-intelligence.fraunhofer.de/en/articles/quantum-computing-in-medical-diagnostics

Das Projekt CanConnect verfolgt das Ziel, Krebsregisterdaten mit vielfältigen Diagnostikdaten der meldenden Kliniken zu verknüpfen. Im Zentrum steht die Entwicklung eines allgemeinen Verknüpfungskonzepts, das umfangreiche, detaillierte Falldaten für die Forschung nutzbar macht und gleichzeitig den Schutz der Patientendaten gewährleistet. Die Machbarkeit und der Nutzen des entwickelten Verknüpfungskonzepts werden anhand des Anwendungsfalls »Glioblastom«, eines bösartigen Hirntumors, demonstriert. Hierfür wird das entwickelte Verknüpfungskonzept angewendet, um Krebsregisterdaten mit weiteren Diagnostikdaten aus der Pathologie anzureichern. Die verknüpften Daten werden mit Methoden der Künstlichen Intelligenz (KI) analysiert, um daraus als exemplarische Anwendung neue Diagnoseparameter (sogenannte Biomarker) für das Glioblastom abzuleiten.

https://www.bundesgesundheitsministerium.de/ministerium/ressortforschung/handlungsfelder/forschungsschwerpunkte/krebsregisterdaten/canconnect

In the new EU project “SafePolyMed" an international research team sets out to provide physicians and pharmacists with innovative tools to increase drug treatment safety and educate patients on how to actively manage their own healthcare. The Fraunhofer Institute for Biomedical Engineering IBMT is contributing its many years of expertise in the field of health information systems to the new project.

https://www.safepolymed.eu/

https://www.ibmt.fraunhofer.de/en/ibmt-press-releases-2022/press-ibmt-safepolymed-2022-07-06.html

Das PROSurvival-Projekt zielt darauf ab, das Überleben von Patienten mit Prostatakrebs (PCa) präziser vorherzusagen. Langfristig soll ein umfassender, standortübergreifender, digitaler Datensatz von PCa-Proben generiert werden, um die gemeinschaftliche Entwicklung von KI für die Präzisionsmedizin bei PCa zu unterstützen. Bisherige Forschungsarbeiten haben gezeigt, dass für das Training von KI-Modellen Daten von mehreren Standorten erforderlich sind. Oft können diese jedoch aufgrund von Datenschutzbestimmungen nicht gemeinsam genutzt werden. Daher werden föderierte KI-Modelle entwickelt. Solche Modelle nutzen die Patientengeschichte und die klinischen Daten in Kombination mit öffentlich verfügbaren Daten. PROSurvival wird eine datenschutzkonforme föderierte Infrastruktur einrichten, um den Fundus an klinischen Routinedaten zu nutzen. Die Bilddaten sollen mithilfe von klinisch relevanten Musterinformationen verdichtet werden, was die Komplexität des Datensatzes reduzieren und die Analyse mit handelsüblicher Hardware erleichtern wird.

https://www.offis.de/en/offis/project/prosurvival.html

https://www.gesundheitsforschung-bmbf.de/de/prosurvival-uberlebensvorhersage-fur-prostatakrebspatienten-mithilfe-von-foderiertem-15926.php

A project at Fraunhofer IGD called Click2Print Artificial Eyes (C2PAE) is devoted to developing 3D-printed eye protheses. Researchers there have created software to control a completely digitalized 3D printing process for artificial eyes. A printer driver dubbed Cuttlefish, which the institute has been developing since 2014, is used. FIT AG, a German enterprise, is also involved in the C2PAE project; it is focusing on additive production of prosthetic eyes with PolyJet technology. The company also performs quality management for the 3D-printed prosthetics. It is now working to develop a process for series production of prosthetic eyes. Within the scope of C2PAE, trials are currently being conducted with patient-specific data at Moorfields Eye Hospital in London. Ocupeye, a startup, is also involved; it licenses and operates the software, deals with regulatory issues, and handles marketing and sales.

https://www.igd.fraunhofer.de/en/research/public-projects/healthcare/3d-printed-prosthetic-eyes.html

https://www.cuttlefish.de/

The AIOLOS project (Artificial Intelligence Tools for Outbreak Detection and Response) has received approval from the French and German governments to develop a digital platform designed to allow for early detection of new respiratory pathogens epidemics, monitor their spread and inform decisions on appropriate counter measures. AIOLOS will provide insights for private and public decision-making in a web-based dashboard, which leverages real-time data from multiple data sources, advanced artificial intelligence (AI), and predictive modeling. Sanofi in France and the Fraunhofer Institutes SCAI and ITMP in Germany lead the consortium, which includes four other French and German partners: CompuGroup Medical, Quinten Health, Impact Healthcare and umlaut, part of Accenture. The project is supported by the French State and the German Federal Ministry for Economic Affairs and Climate Action in the context of the Franco-German call on Artificial Intelligence technologies for risk prevention, crisis management and resilience. This call for projects operated by Bpifrance, on behalf of the French government as part of the France 2030 investment plan, and by DLR Projektträger, aims to support innovation projects between France and Germany on artificial intelligence technologies.

https://www.itmp.fraunhofer.de/en/press/AIOLOS.html

http://www.aiolos-project.org/

Seltene Erkrankungen zu diagnostizieren, stellt Ärztinnen und Ärzte oft vor Herausforderungen. Die Digitalisierung kann dabei helfen, eine Diagnose zu stellen – das Universitätsklinikum Frankfurt zeigt wie: Es ist federführend am Forschungsprojekt Smartes Arztportal für Betroffene mit Seltenen Erkrankungen (SATURN) beteiligt. Gefördert vom Bundesministerium für Gesundheit entsteht eine Plattform, die Künstliche Intelligenz (KI) nutzt, um bei der Diagnosestellung zu helfen. Als Softwareexperte untersucht das Fraunhofer IESE im Projekt, wie mit Hilfe von KI bei geringen Datenmengen nachvollziehbare und transparente Verdachtsdiagnosen für Seltene Erkrankungen gestellt werden können.

https://www.saturn-projekt.de/en/

https://www.iese.fraunhofer.de/de/media/presse/pm-2024-02-29-saturn.html

https://www.iese.fraunhofer.de/blog/ki-zur-diagnostik-von-seltenen-erkrankungen/

In the project SATURN, which is funded by the German Federal Ministry of Health, Fraunhofer IESE is collaborating with its project partners to research methods for using AI technologies from the areas of rule-based systems and machine learning for traceable and transparent decision support / tentative diagnosis in the context of unclear diseases.

https://www.saturn-projekt.de/en/

https://www.iese.fraunhofer.de/en/reference/saturn-ai-based-tentative-diagnostics.html

ADIS stands for »Early Diagnosis of Alzheimer's Disease by Immune Profiling of Cytotoxic Lymphocytes and Recording of Sleep Disturbances.« The project is funded by the EU Joint Programme for Neurodegenerative Diseases Research (JPND). JPND is the largest global research initiative to address neurodegenerative disease challenges. The ADIS project (start: July 2022) will run for three years and has a budget of 1.3 million euros, of which 300,000 euros will go to Fraunhofer SCAI.

Alzheimer's disease (AD) and related dementias are heterogeneous, multifactorial diseases in which several etiopathogenic mechanisms lead to neuronal cell death and loss of cognitive function. The disease is thought to begin decades before diagnosis, posing a significant treatment challenge. Therefore, the identification of prognostic biomarkers for AD is of great importance. 

https://adis-project.eu/

https://www.scai.fraunhofer.de/en/projects/ADIS.html

In the dental field, image data (X-ray images) represent the initial source of information for a first assessment of a patient's health status and also serve as the basis for further planning of the treatment process. Both extraoral images, such as the orthopantomogram (OPG) or lateral cephalometric radiographs (LCR), and intraoral images, such as bitewing radiographs, are used. The OPG is the typical initial image because all teeth including roots are clearly visible.

https://www.igd.fraunhofer.de/en/research/core-competencies/dental-image-analysis.html

REMEDi4ALL launched with the aim of making a major leap forward in drug repurposing. This promising approach to drug develop-ment consisting in the identification, testing, and validation of new therapeutic indi-cations for existing medications, is a developing field but faces numerous barriers and systemic inefficiencies. Still, its potential to significantly bring down times and costs of drug development -it focuses on already approved, discontinued, shelved or investigational therapeutics- makes this novel strategy attractive for rare and ne-glected conditions, cancer, emerging public health threats such as COVID-19 or new drug combinations. It also translates into more sustainable health systems. 

To advance knowledge in this field and address substantial obstacles -fragmented and siloed research; non-standardised datasets; heterogenous quality of computa-tional tools; poor patient engagement or lack of incentives and policies to support and enhance drug repurposing- the European Union (EU) through the Horizon Europe (HE) programme will invest 23 million euros in REMEDi4ALL over the next 5 years. It is expected that, due to REMEDi4ALL, more (and better) repurposed therapeutics will be widely available thanks to more agile, cutting-edge development processes, ultimately contributing to increased sustainability of health systems. 

https://www.itmp.fraunhofer.de/de/presse/REMEDi4ALL.html

https://remedi4all.org/

With an aging population and accompanying increase in the number of cancer cases, as well as an increasing number of complex diagnostic procedures for new therapies in cancer treatment, the workload in pathology is unceasingly increasing. At the same time, there is a shortage of specialists. Digitization together with artificial intelligence methods offer new opportunities for support in pathological diagnostics and help to close the gap in demand.

https://www.scs.fraunhofer.de/en/focus-projects/ada-center/robust-ai-for-digital-pathology.html

Today’s imaging technologies in ophthalmology are so advanced that retinal and vascular structures in the eye can be resolved with unprecedented precision in two, three and even four dimensions. However, interpreting the image material and deriving a therapy decision taking into account the patient's history is a complex task that requires a lot of experience. Treatment errors may have severe consequences for patients. The recently launched joint project »Ophthalmo-AI«, coordinated by the Fraunhofer Institute for Biomedical Engineering IBMT, aims to create an intelligent and interactive assistance system that supports ophthalmologists with methods of explainable artificial intelligence to provide comprehensible diagnoses and treatment suggestions.

https://www.interaktive-technologien.de/projekte/ophthalmo-ai

https://www.ibmt.fraunhofer.de/en/ibmt-press-releases-2021/press-ophthalmo-ai-2021-05-04.html

We support elderly people by providing technological and social end-to-end solutions that enable them to live self-sufficiently and autonomously in their own home for a longer time. To this end, we use a wide range of technologies, such as telematics or sensor technology, and combine these with social and organizational concepts to create an overall solution.

https://www.iese.fraunhofer.de/en/reference/smart-sssistance-systems-for-senior-citizens.html

In many application areas, a detection of affective and cognitive states can be beneficial. For example, in areas such as usability testing, state detection can provide better insight into the effect of a product on the user and provide information about their possible overload with the product.

However, some states are expressed extremely subtly, making their detection a major challenge. For example, one modality, e.g. video, is not sufficient to robustly detect cognitive overload. This can only be made possible by combining different modalities, such as gaze detection and different biosignals.

https://www.scs.fraunhofer.de/en/focus-projects/ada-center/explainable-ai-for-multimodal-state-detection.html

https://www.scs.fraunhofer.de/en/focus-projects/ada-center.html

Bei Erkrankungen wie entzündlichen rheumatischen Erkrankungen (z. B. axiale Spondyloarthritis), Spondylolisthesis, Skoliose, Morbus Bechterew, Morbus Scheuermann oder Bandscheibenvorfällen sind chronische Rückenschmerzen ein häufiges Symptom. Meist erhält der Patient oder die Patientin Physiotherapie, alleine oder ergänzend zu den Medikamenten. Die Physiotherapie soll dabei eine Stärkung der Muskeln des unteren Rückens fördern. Unter normalen Umständen wird die Muskulatur des unteren Rückens nicht bewusst kontrahiert. Deshalb wird das gezielte Training dieser Muskulatur als schwierig empfunden. Um das Training zu erleichtern, wollen die Verbundpartner ein tragbares Ultraschall-System entwickeln. Dieses soll ein Biofeedback über die Genauigkeit des Trainings liefern. Die Grundlage dafür soll die Analyse der Muskelkontraktion sein. Die Muskelkontraktion wird mit Hilfe von Ultraschall gemessen. Die aufgenommenen Ultraschallsignale werden an ein Elektronikmodul übertragen. In diesem werden die Signale mit sogenannten Deep-Learning- Ansätzen analysiert. Daraus wiederum wird dann ein Biofeedback generiert und an den Patienten oder die Patientin gegeben. Diese können dann das Training entsprechend anpassen.

https://www.gesundheitsforschung-bmbf.de/de/ultrawear-ultraschall-basiertes-wearable-als-biofeedback-system-fur-ein-effektives-12518.php

The 24-hour ECG as a standard method of diagnosing heart disease can miss any irregularities that occur with lesser frequency. For sufferers, this often results in a lengthy odyssey until they finally obtain a proper diagnosis and are put on a suitable course of treatment. A team of researchers at Fraunhofer IGD in Rostock is seeking to plug this gap in diagnostics. The Guardio® health app enables users to perform multichannel ECGs without electrodes being attached to their skin and to detect heart disease much earlier than with conventional methods.

https://www.igd.fraunhofer.de/en/media-center/press-releases/app-by-fraunhofer-detects-heart-disease-earlier.html

KIPeriOP is a research project funded by the German Federal Ministry of Health (BMG) with the aim of improving perioperative risk management and reducing perioperative mortality and permanent damage. Clinical guidelines already support perioperative decision-making and will be complemented in the project by the trustworthy use of artificial intelligence, including the prediction of postoperative risks based on preoperative risk factors. The project consortium combines outstanding clinical, technical, ethical and economic expertise and is led by the University Hospital of Würzburg (clinical coordination) and the Fraunhofer Institute for Digital Medicine MEVIS (technical coordination).

https://www.kiperiop.de/en/home.html

https://www.mevis.fraunhofer.de/en/press-and-scicom/press-release/2021/making-operations-safer-with-digitized-data-acquisition-and-ai.html

Recent advancements in image-based diagnostics, driven by artificial intelligence (AI) methods, have been significant. EMPAIA International e. V. is committed to facilitating routine use of validated and certified AI solutions by healthcare professionals. Additionally, the promotion of AI usage involves the proactive elimination of regulatory, legal, technical, and organizational obstacles.

The inadequate standardisation of interfaces in digital pathology is an important obstacle on the way to more digitalisation. Building on established standards such as HL7 and DICOM, we develop and disseminate open standards for the rapid dissemination of modern methods in clinical diagnostics. This also serves to improve patient access to their data.

https://www.empaia.org/

https://www.mevis.fraunhofer.de/en/press-and-scicom/institute-news/2020/kickoff-for-large-scale-ai-project-empaia.html

The »Translational Brain Imaging Training Network« (TRABIT) is an interdisciplinary and intersectoral joint effort of computational scientists, clinicians, and the industry in the field of neuroimaging. Its aim is to train a new generation of innovative and entrepreneurial researchers to bring quantitative image computing methods into the clinic, enabling improved healthcare delivery to patients with brain disease.

Since brain imaging often visualizes disease effects with much greater sensitivity than clinical observation, it holds great promise to help diagnose patients at the earliest stages of their disease, when treatment is most effective; and personalize their treatment by evaluating their response to a specific intervention. A fundamental bottleneck in translating the wealth of information contained in medical images into optimized patient care is the lack of patient-specific computational tools to help analyze and quantify the torrent of acquired imaging data. The last two decades of medical image computing research have matured to allow robust and automatic assessment of carefully homogenized scientific studies of mostly healthy brain scans. Yet analyzing the »wild« type of neuroimaging data arising in the standard clinical treatment of brain disorders remains a hard and unsolved problem.

https://trabit.eu/

https://www.mevis.fraunhofer.de/de/press-and-scicom/institute-news/2017/kickoff-of-translational-brain-imaging-training-network.html

Ein zentrales Element bei der Psychotherapie von Angststörungen ist es, die Betroffenen den Angst auslösenden Situationen auszusetzen. Im Projekt wird ein System entwickelt, das die emotionalen Reaktionen der Klientinnen und Klienten während solcher Expositionen multimodal erfasst und daraus durch Datenfusion Parameter extrahiert, die für den weiteren Verlauf relevant sind. Aus diesen Parametern werden Mikrointerventionen abgeleitet, die den Patientinnen und Patienten durch einen virtuellen Agenten in einer intuitiven Interaktion zur Verfügung gestellt werden. Durch die maschinelle Verarbeitung der in zahlreichen Expositionen gewonnenen Datensätze werden Prognosen für erfolgreiche Mikrointerventionen abgeleitet.

https://www.interaktive-technologien.de/projekte/optapeb

https://www.iis.fraunhofer.de/en/ff/sse/machine-learning/affective-sensing.html

Some things that most people do entirely at the unconscious level pose enormous challenges for autistic children: correctly recognizing and interpreting the emotions of the person they are interacting with and responding accordingly. Using robots as an interactive communication tool in therapy allows children on the autism spectrum to get the help they need early on.1

The joint ERIK project therefore aims to develop a robotic platform that addresses new interaction strategies in treatments for children with impaired socioemotional functioning, such as those on the autism spectrum.

https://www.iis.fraunhofer.de/en/ff/sse/machine-learning/affective-sensing/emotionssensitive-robotik.html

https://www.interaktive-technologien.de/projekte/erik

Before integrating new technologies into the healthcare system, it is crucial to comprehensively test their safety, robustness, and reliability. In the case of AI and robotics, quality requirements in the European Union are high, but the current testing infrastructure for developing standards, testing innovations, and certifying new products remains insufficient.

The TEF Health consortium aims to improve this process and accelerate the validation and certification of AI and robotics in medical devices. To this end, the research team is developing a test infrastructure (both virtual and physical) that can evaluate various technologies in realistic environments, including hospitals and laboratories. For instance, software for patient care or diagnostics, as well as surgical or nursing robots can be tested by users.

https://www.hhi.fraunhofer.de/en/news/nachrichten/2023/fraunhofer-hhi-develops-european-test-infrastructure-for-ai-and-robotics-in-healthcare.html

https://tefhealth.eu/home

The aim of the platform is to make the drug class of proximity inducing drugs (PiDs) usable for the treatment of immune-mediated diseases. PiDs are bifunctional molecules which, by transient binding to two target structures, bring them into spatial proximity to each other and thereby trigger a biological effect. In PiDs in the narrow sense, one of the two structures is a ubiquitin E3 ligase, and the other is the target protein, which is tagged by this E3 ligase for degradation in the proteasome. These PiDs thus lead to the complete loss of the target protein with all its functions, e.g. catalytic, structural or regulatory, which distinguishes PiDs from classical inhibitors

https://www.proxidrugs.de/

https://www.cimd.fraunhofer.de/en/Platforms/ProxiDrugs.html

https://www.itmp.fraunhofer.de/en/press/BMBF-Proxidrugs.html

In the EU AIDPATH project, the partners from industry and research will build an automated and intelligent facility over a period of four years that is capable of producing targeted and patient-specific cell therapy directly at the point of treatment. In addition, the project addresses the integration of the facility into the hospital environment, taking into account logistics processes as well as data management and data security.

The fairly new CAR-T cell therapy is based on genetically modified T cells. These are the body's own white blood cells that make up part of the immune system. The T cells are taken from the patient's blood for treatment and equipped with a so-called chimeric antigen receptor (CAR). This receptor enables the cells to recognize and destroy tumor cells. CAR-T cell therapy has already been used in Germany for two years and is paving the way for completely new treatment approaches in hematology and oncology. However, much of the time spent on CAR-T cell therapy to date has still been taken up by complex logistics processes from central production facilities and inflexible manufacturing and application schemes. In addition, it has not yet been possible to take into account the individual cell characteristics of the patient, so that the success of the therapy cannot always be guaranteed.

https://www.ipt.fraunhofer.de/en/projects/aidpath.html

https://www.sciencrew.com/c/6499?title=AIDPATH

Biomedical research currently finds itself at a crossroads: despite it being technically possible to develop, manufacture and utilize innovative therapeutic procedures, their broad application would place a huge burden on the health care system – both financially and logistically.

The SaxoCell consortium has made it its mission to translate new technologies that could potentially treat previously incurable diseases into clinical application.

In Leipzig, Fraunhofer IZI has already gained experienced in manufacturing this kind of new drug in the form of CAR-T cell therapy, which is used to treat rare forms of leukemia. Application of the living drug is also proving successful at Saxony’s university hospitals in Dresden and Leipzig.

https://www.saxocell.de/en/home/

https://tu-dresden.de/tu-dresden/newsportal/news/saxocell-startet-mit-lebenden-arzneimitteln-in-die-zukunft

https://www.izi.fraunhofer.de/en/press/press-releases/saxocell-living-drug-made-in-saxony.html

https://www.izi.fraunhofer.de/en/departments/saxocell.html

Große Mengen hochwertiger Daten sind eine wichtige Grundlage für zukunftsweisende Forschung auf dem Gebiet der Gesundheitsversorgung. Das Forschungsdatenzentrum Gesundheit (FDZ) am Bundesinstitut für Arzneimittel und Medizinprodukte (BfArM) hat die Aufgabe, Daten von gesetzlich Krankenversicherten berechtigten Institutionen zu Forschungszwecken zur Verfügung zu stellen. Hierbei gibt es zwei Herausforderungen: Zum einen muss das FDZ organisatorisch und technisch fähig sein, auf Anfragen zeitnah und nutzerorientiert zu reagieren. Zum anderen handelt es sich um hochsensible, schutzbedürftige, persönliche Gesundheitsdaten. Das Vorhaben »Künstliche Intelligenz am Forschungsdaten­zentrum« (KI-FDZ) soll die vorhandenen Daten für die Forschung auch mit KI-Methoden erschließen und eine bestmögliche Nutzung erlauben, ohne dass Informationen über einzelne Personen abgeleitet werden können. Gemeinsam mit dem Fraunhofer-Institut für Digitale Medizin (MEVIS) wird ein »Sandbox«-System im FDZ aufgebaut, also ein virtueller Raum und nutzungsfreundlicher KI-Werkzeugkasten, der ein Austesten der Möglichkeiten in einer geschützten Umgebung möglich macht. Das FDZ will damit Anträgen für Forschungsprojekte, die KI-Methoden erfordern, den Weg bereiten.

https://www.bundesgesundheitsministerium.de/ministerium/ressortforschung/handlungsfelder/digitalisierung/ki-fdz

Since 2018 we are an active member of the PIONEER consortium, a European Network of Excellence for Big Data in Prostate Cancer, consisting of 32 partners across 9 countries. PIONEER’s goal is to ensure the optimal care for all European men living with prostate cancer by unlocking the potential of Big Data and Big Data analytics. A key objective of PIONEER is to standardise and integrate existing ‘big data’ from quality multidisciplinary data sources into a single innovative open access data platform, to accelerate prostate cancer research. Within PIONEER, we contribute our strengths and expertise in data harmonisation of transcriptome-wide expression studies as well as statistical data analyses to identify and confirm biomarkers.

https://prostate-pioneer.eu/

https://www.izi.fraunhofer.de/en/departments/leipzig-location/diagnostics/bioinformatics/projects/biomarker-signatures-for-prostate-cancer-diagnosis-and-prognosis.html

A significant challenge facing the development of immunomodulating therapies is their preclinical evaluation in terms of efficacy and safety. The greatest problem here is the complexity of the human immune system. The EU consortium imSAVAR (Immune Safety Avatar: nonclinical mimicking of the immune system effects of immunomodulatory therapies) is addressing these challenges by coming up with new ways of examining immunomodulatory therapies. Existing model systems are to be improved and new ones developed in order to identify adverse side effects of new therapies affecting the immune system. Furthermore, new biomarkers for diagnosing and predicting immune-mediated pharmacology and toxicities will be developed. The focus is also on more detailed research into toxicity mechanisms and the potential for their mitigation via therapeutic interventions.

The interdisciplinary imSAVAR consortium is made up of 28 international partners from 11 nations and is being coordinated by the Fraunhofer IZI and Novartis. Partners include university and non-university research facilities, pharmaceutical and biotechnology companies, as well as regulatory authorities.

https://imsavar.eu/

https://www.izi.fraunhofer.de/en/departments/leipzig-location/diagnostics/bioinformatics/projects/medical-bioinformatics-in-immuno-oncology.html

Schlaganfallpatienten entwickeln in bis zu 10% aller Fälle eine extreme Hirnschwellung. Man spricht dann von einem »malignen Infarkt«, der aufgrund des Druckanstieges im Gehirn zu schwerwiegenden Folgeschäden führt und häufig tödlich verläuft. Die Schwellung des malignen Infarktes ist durch Medikamente kaum beeinflussbar und bedarf einer operativen Entlastung, die ihreseits risikobehaftet ist und deshalb häufig zu spät durchgeführt wird. Klinisch besteht somit die Herausforderung, schon frühzeitig diejenigen Patienten zu identifizieren, bei denen eine operative Therapie unerlässlich ist, um genau diese Patienten zeitnah zu operieren.

Ziel des Projektes ist es daher, mit Hilfe der Methode des »maschinellen Lernens« computertomografische Aufnahmen (CT-Bilder) und klinische Daten von 1.500 Patienten zu analysieren und ein Modell zu entwickeln, das hilft, die Hirnschwellung frühzeitig zu erkennen und ihren Verlauf vorherzusagen. Zunächst identifizieren computer-basierte Algorithmen automatisch charakteristische Merkmale aus den CT-Bildern und den klinischen Daten (Lernphase). In einer anschließenden Validierungsphase wird das so entwickelte Modell an neuen Datensätzen überprüft. Mit dem Einsatz des maschinellen Lernens soll die Hirnschwellung besser verstanden und frühzeitig erkannt werden.

https://apices-trial.de/das-projekt/

https://innovationsfonds.g-ba.de/projekte/versorgungsforschung/apices-automatic-prediction-of-edema-after-stroke-computergestuetzte-automatische-prognose-der-entwicklung-eines-malignen-hirnoedems-nach-mediainfarkt.250

The COVID-19 Pandemic Radiological Cooperative Network RACOON is a joint project of the radiology departments at all 36 German university hospitals. Experts at all sites segment and annotate a large number of lung CT scans in a structured and uniform manner. The resulting data is used to develop medical assistance systems and an early warning system based on AI. Fraunhofer MEVIS is one of the three technical partners in RACOON, in addition to the German Cancer Research Center (DKFZ) and the Technical University of Darmstadt.

RACOON applies MEVIS technology called SATORI for interactive data curation, AI, and radiomics analysis. 

https://racoon.network/

https://www.mevis.fraunhofer.de/en/research-and-technologies/fraunhofer-mevis-vs-corona.html

https://www.charite.de/en/service/press_reports/artikel/detail/joining_forces_to_fight_covid_19/