Fraunhofer Institute for Cognitive Systems IKS
Fraunhofer Institute for Cognitive Systems IKS
© iStock.com/alexkar08

© iStock.com/alexkar08

Trustworthy Digital Health

Medical AI solutions you can trust

The safety and trustworthiness of intelligent health applications is at the heart of our research on Digital Health at Fraunhofer IKS. AI applications must be reliable, transparent, and robust for patients, doctors, healthcare providers and other important stakeholders. Only then can the much-discussed potential of digitalization and Artificial Intelligence (AI) in the safety-critical healthcare sector be exploited to provide the best possible patient care.

Our guiding theme, Trustworthy Digital Health, brings together various key areas of research, interdisciplinary expertise, and solutions that we are constantly developing in line with the state of research, technology, and the healthcare industry. Together with clinics, medtech & health IT companies, networks, universities and many other partners from research and industry, we research and develop intelligent healthcare solutions, ranging from basic AI models to cutting-edge generative AI approaches. Our focus is specifically on applications for clinical prediction and decision support.

Our AI solutions

Fraunhofer IKS offers a wide range of solutions across medical specializations and applications in healthcare. Our customers include clinicians and clinics, companies, and universities.

Expand all Close all

  • Bring AI ideas to clinical implementation

    You have an idea for how AI can facilitate your clinical work but don't know how to get there or where to start? Fraunhofer IKS can help bring AI from a first idea all the way into your clinical practice.

    Our process:

    1. Joint project ideation & scoping: Collaborate with us to brainstorm and develop innovative, feasible AI concepts that fit your clinical needs.
    2. Identification of funding opportunities: Our team will assist you in discovering the most suitable funding sources to bring your projects to life.
    3. Development & implementation partnership: Count on us as your trusted partner in the development and execution of your AI projects, ensuring fast success and a trustworthy AI solution every step of the way.

    Transform your AI ideas into reality!

    Contact us now

  • Fast track AI for your products with Fraunhofer IKS

    Do you have an idea for how AI can be used in your company but lack the resources or experience to make it a reality? Fraunhofer IKS boosts your AI projects with a holistic competence stack in (safe) AI development – whether you aim for a first proof of concept, rapid prototype or market-ready AI product.

    Our support offers:

    • Proof of concept: Small-scoped proof of the technical feasibility of the intended AI solution.
    • Rapid prototyping: Agile AI development, iteration and testing of a working prototype as a first milestone on your product roadmap.
    • Development of custom AI solutions: Crafting innovative AI models tailored to your unique needs.
    • Data analysis with AI: In-depth AI analysis of your data to deliver critical insights.
    • Joint R&D teams: Sparring with our AI (safety) experts to fill competence, resource or scientific gaps in your team and speed up your AI development.

    Unlock your AI project's potential!

    Contact us now

  • Build a consortium with Fraunhofer IKS

    Are you seeking a reliable AI development and implementation partner with a proven track record in successful funding proposals and projects?

    As a neutral applied research partner, we contribute:

    • Proposal preparation expertise: Leverage our experience to secure the funding you need for your projects.
    • Experienced leadership: We are ready to step in and lead essential work packages or take on pivotal roles in large-scale projects.
    • Strong network of international partners from industry, academia, healthcare, political and regulatory associations, etc.
    • AI and digital health competence of the Fraunhofer-Gesellschaft, one of the leading applied research organizations.

    Join forces for AI success!

    Contact us now

  • Independent data and AI model review

    Are you in need of neutral, expert feedback on your AI model? Fraunhofer IKS supports you in addressing the specific challenges of preparing AI models for regulatory approval and clinical use. 

    Our extensive consulting services include:

    • Data Quality review: Assessment of the accuracy, consistency, and completeness of training and validation data.
    • Bias detection: Identification of sources of bias that may affect the model’s generalizability.
    • Model selection: Review of implemented algorithms to ensure the most state of the art methodologies are employed.
    • Explainability and interpretability: Evaluation and suggestion of mechanisms for interpretability to increase trust and ease of validation in clinical use.
    • Performance and error analysis: Evaluation of performance metrics and types of errors to assess significance in the context of the medical application.

    Get expert insights on your AI model!

    Contact us now

Discover the future of reliable healthcare!

Partner with us to drive innovation in medical AI. We specialize in developing trustworthy and reliable AI models based on various types of medical data, designed to enhance your platforms and applications.

Prediction & decision support

Elevate and personalize patient care with our targeted AI solutions tailored to your needs – in Cardiology, Women‘s Health and other areas of healthcare.

Trustworthy AI

Our interpretable and explainable AI models give you the confidence to trust and understand every decision made, we help you to handle data challenges, and facilitate your clinical workflow using safe LLM agents.

Medical data

From time series to medical images, tabular data, and text, we harness the power of data to deliver reliable insights, enabling you to provide better diagnoses and to optimize healthcare processes.

Expand all Close all

Prediction & decision support

The digital transformation of healthcare opens the potential of using AI along the patient journey to analyze patient data, to support medical decisions in ways that were previously not possible. More individualized risk or disease prediction, suggestion of treatment plans, or AI-based event forecasting can contribute to more personalized and preventive healthcare. 

We specialize in developing explainable and reliable AI models for clinical prediction and Decision Support Systems (DSS) that support healthcare professionals in data analysis, improved decision-making and health process optimization. From Cardiology and Women’s Health to hospital management and patient documentation, we develop solutions for the diverse areas of healthcare and medical specializations together with healthcare experts and industry. Besides "classical" AI methods, we are also exploring future technologies like quantum computing.

Our research approaches

Medizin im Labor
© iStock.com/gorodenkoff
  • AI-based Decision Support Systems, e.g., decision support for a multi-organ failure device
  • Individual risk prediction models, e.g., prediction of complications in stent procedures
  • Reliable AI forecasting models
  • Generative AI solutions for the analysis of complex or multimodal health information, e.g., EHR data, diagnoses, patient documentation, medical guidelines
  • AI solutions for resource planning and process optimization, e.g., AI-based workforce planning in hospitals
  • Hybrid quantum-enhanced AI models, e.g., breast cancer classification in medical images

Trustworthy AI

In the safety-critical healthcare sector, where lives are at stake, ensuring trustworthy and robust AI solutions is crucial. Regulations such as the EU AI Act or Medical Device Regulation (MDR) underline the importance of safety in medical AI. Our extensive trustworthy AI expertise from previous research projects includes handling the real-world healthcare challenges, such as non-transparent AI-driven decision processes, incomplete, unstructured, little or distributed data, biases in datasets, or hallucinations of Large Language Models (LLMs). In close collaboration with healthcare experts and industry, we develop medical AI solutions that you can trust and understand.

Our research approaches

arzt patient
© iStock.com/AmnajKhetsamtip
  • Explainable (XAI) and human-interpretable-by-design AI models to provide transparency of AI-driven decision-making, e.g., Prototype Learning, Concept Learning
  • Reliable AI prediction models 
  • Independent evaluation of medical AI models, e.g., model explainability, accuracy, dataset bias, distribution shift
  • Confident AI solutions for use cases with “little data”, e.g., Feedback-guided Automation of Sub-Tasks (FAST) 
  • Trustworthy LLMs, e.g., for women’s health
  • Safety frameworks for Generative AI solutions, e.g., Retrieval-Augmented Generation (RAGs) or our Safety Companion
  • Transferability of AI models (domain generalization) across applications and healthcare facilities
  • Federated Learning, enabling collaborative model training with data distributed across multiple medical institutions without data centralization 

Medical data

From medical images to time series, tabular or text data, the digital transformation makes an increasing amount of data available for AI-driven analysis that can support improved decision-making and relieve healthcare professionals from time-consuming, manual tasks. In medical imaging, such as Pathology and Radiology, AI already supports doctors in the evaluation of X-ray, MRI or CT images. AI models analyzing time series data can identify patterns and trends over time, for example to predict risks or allow for more efficient resource planning. Modern Generative AI (GenAI) approaches allow us to analyze unstructured or multimodal health data, enabling the (partial) automation of time-consuming administrative or documentation tasks. With our expertise and variety of research approaches, we help you to use the potential of AI-driven analysis for the different types of health and medical data.

Our research approaches

Ärztinnen begutachten medizinische Bildaufnahmen
© iStock.com/vm
  • Explainable and human-interpretable image classification and segmentation models, e.g., early detection of osteoporosis in CT images
  • Time-series analysis, e.g., monitoring disease progression in sequential patient data such as vital signs and lab values
  • Generative AI solutions, e.g., Retrieval-Augmented Generation (RAG) for information extraction and summarization
  • Safe LLMs, e.g., for information extraction and summarization from health reports and guidelines 

More information about our solutions and offerings

Expand all Close all

Digital Health projects

 

© iStock.com/sudok1

INDICATE: Connecting Data in European Intensive Care

INDICATE aims to advance patient-centered care and promoting ethically responsible data use and the development and implementation of trustworthy AI models.

 

© iStock.com/vm

Using AI for facial fracture detection

Consultation projects play a crucial role in fulfilling the Fraunhofer mission of translating cutting-edge research into industry applications. Recently, Fraunhofer IKS cooperated with the South Korean company ZIOVISION on AI-based facial fracture segmentation from medical images. The successful outcome of the project demonstrates the potential benefits such collaborations offer to both partners.

 

© iStock.com/AndreyPopov

Reinforcement Learning Shift Planning Agent Set to Transform Hospital Staffing

Faced with cost pressures and a shortage of healthcare professionals, organizations are challenged to increase efficiency. The integration of artificial intelligence (AI) into workforce management offers promising approaches. In a joint project, Fraunhofer IKS and ATOSS Software have developed an AI-controlled shift planning agent that automates staff scheduling while demonstrating remarkable scalability.

 

 

© iStock.com/eggeeggjiew

AI-supported personnel planning in hospitals

AI can help make tedious routine tasks easier, such as forecasting staffing requirements in hospitals. Fraunhofer IKS is addressing this issue in a current project - the findings were presented at the Healthcare Hackathon 2023 in Mainz, Germany.

 

© iStock.com/Jan-Otto

AI helps where humans get stuck

Artificial intelligence can help in the treatment of coronary artery disease using stents. It was possible to predict complications and reduce their occurrence.

 

© iStock.com/sudok1

AI assists in making treatment decisions

Clinicians, for the most part, suffer from enormous workloads. With the help of AI, a clinical multi-organ support system can be used even better for treatment.

 

© iStock.com/alvarez

Online tool checks reliability of AI models

In modern perception applications, such as in medical engineering, models based on artificial intelligence (AI) are increasingly being used due to their strong performance. However, this increasing performance often comes at the cost of the transparency of results. An online tool developed by the Fraunhofer Institute for Cognitive Systems IKS can help here.

Publications

Jahr
Year		Titel/Autor:in
Title/Author		Publikationstyp
Publication Type
2025 Application of Infrared Thermography and Artificial Intelligence in Healthcare: A Systematic Review of Over a Decade (2013-2024)
Vicnesh, Jahmunah; Salvi, Massimo; Hagiwara, Yuki; Hah, Yan Yee; Mir, Hasan; Barua, Prabal Datta; Chakraborty, Subrata; Molinari, Filippo; Acharya, Rajendra U.		 Zeitschriftenaufsatz
Journal Article
2025 A View on Vulnerabilites: The Security Challenges of XAI
Pachl, Elisabeth; Langer, Fabian; Markert, Thora; Lorenz, Jeanette Miriam		 Konferenzbeitrag
Conference Paper
2025 Evaluation of Active Feature Acquisition Methods for Time-varying Feature Settings
Kleist, Henrik von; Zamanian, Alireza; Shpitser, Ilya; Ahmidi, Narges		 Zeitschriftenaufsatz
Journal Article
2024 Analysis of Missingness Scenarios for Observational Health Data
Zamanian, Alireza; Kleist, Henrik von; Ciora, Octavia; Piperno, Marta; Lancho, Gino; Ahmidi, Narges		 Zeitschriftenaufsatz
Journal Article
2024 Zukunftsteam KI und medizinisches Personal
Lorenz, Jeanette Miriam; Pachl, Elisabeth		 Zeitschriftenaufsatz
Journal Article
2024 Delineating morbidity patterns in preterm infants at near-term age using a data-driven approach
Ciora, Octavia; Seegmüller, Tanja; Fischer, Johannes; Wirth, Theresa; Häfner, Friederike; Stoecklein, Sophia; Flemmer, Andreas W.; Förster, Kai; Kindt, Alida; Bassler, Dirk; Poets, Christian F.; Ahmidi, Narges; Hilgendorff, Anne		 Zeitschriftenaufsatz
Journal Article
2024 KI-Entwicklung - von der Vorschrift zum Computercode
Zamanian, Alireza; Pachl, Elisabeth; Lancho, Gino		 Zeitschriftenaufsatz
Journal Article
2024 Intelligente Gesundheit
Lorenz, Jeanette Miriam; Schmidhuber, Johanna		 Zeitschriftenaufsatz
Journal Article
2024 The quest for a practical quantum advantage or the importance of applications for quantum computing
Lorenz, Jeanette Miriam		 Zeitschriftenaufsatz
Journal Article
2023 Proteomics reveals antiviral host response and NETosis during acute COVID-19 in high-risk patients
Bauer, Alina; Pachl, Elisabeth; Hellmuth, Johannes C.; Kneidinger, Nikolaus; Motaharehsadat, Heydarian; Frankenberger, Marion; Stubbe, Hans C.; Ryffel, Bernhard; Petrera, Agnese; Hauck, Stefanie M.; Behr, Jürgen; Kaiser, Rainer; Scherer, Clemens; Deng, Li; Teupser, Daniel; Ahmidi, Narges; Muenchhoff, Maximilian; Schubert, Benjamin; Hilgendorff, Anne		 Zeitschriftenaufsatz
Journal Article
2023 Quantum-enhanced AI in medicine
Lorenz, Jeanette Miriam		 Vortrag
Presentation
2023 Post-hoc Saliency Methods Fail to Capture Latent Feature Importance in Time Series Data
Schröder, Maresa; Zamanian, Alireza; Ahmidi, Narges		 Konferenzbeitrag
Conference Paper
2023 What about the Latent Space? The Need for Latent Feature Saliency Detection in Deep Time Series Classification
Schröder, Maresa; Zamanian, Alireza; Ahmidi, Narges		 Zeitschriftenaufsatz
Journal Article
2023 Künstliche Intelligenz im Gesundheitswesen
Ahmidi, Narges		 Zeitschriftenaufsatz
Journal Article
2023 Toward Safe Human Machine Interface and Computer-Aided Diagnostic Systems
Hagiwara, Yuki; Espinoza, Delfina; Schleiß, Philipp; Mata, Núria; Doan, Nguyen Anh Vu		 Konferenzbeitrag
Conference Paper
2023 Assessable and interpretable sensitivity analysis in the pattern graph framework for nonignorable missingness mechanisms
Zamanian, Alireza; Ahmidi, Narges; Drton, Mathias		 Zeitschriftenaufsatz
Journal Article
2021 Early-, Late-, and Very Late-Term Prediction of Target Lesion Failure in Coronary Artery Stent Patients: An International Multi-Site Study
Pachl, Elisabeth; Zamanian, Alireza; Stieler, Myriam; Bahr, Calvin; Ahmidi, Narges		 Zeitschriftenaufsatz
Journal Article
Diese Liste ist ein Auszug aus der Publikationsplattform Fraunhofer-Publica

This list has been generated from the publication platform Fraunhofer-Publica

Digital Health in our Safe Intelligence online magazine

 

© iStock.com/vm

AI in medicine / 7.10.2025

Using AI for facial fracture detection

Consultation projects play a crucial role in fulfilling the Fraunhofer mission of translating cutting-edge research into industry applications. Recently, Fraunhofer IKS cooperated with the South Korean company ZIOVISION on AI-based facial fracture segmentation from medical images. The successful outcome of the project demonstrates the potential benefits such collaborations offer to both partners.

AI in medicine
 

© iStock.com/vm

Artificial Intelligence / 21.8.2025

Can Generative AI Revolutionize Modern Healthcare?

Artificial intelligence, especially large language models (LLMs), are seen by many as a key resource for an overburdened healthcare system. AI-supported automation in particular could quickly relieve the burden of knowledge management tasks. Before this can happen, security and safety challenges as well as legal requirements must be taken into account. Fraunhofer IKS research is dedicated to both of these aspects.

Generative AI
 

© iStock.com/Erika Parfenova

Portrait Katie Fitch / 27.3.2025

"The interaction between research and industry inspires me"

Dr. Katie Fitch has been head of the department Trustworthy Digital Health at Fraunhofer IKS since November 2024. Katie's enthusiasm for mathematics led her to the engineering section early on. Then she discovered medical AI research for herself.

Portrait Katie Fitch
 

© iStock.com/AndreyPopov

AI in Workforce Management / 6.3.2025

Reinforcement Learning Shift Planning Agent Set to Transform Hospital Staffing

Faced with cost pressures and a shortage of healthcare professionals, organizations are challenged to increase efficiency. The integration of artificial intelligence (AI) into workforce management offers promising approaches. In a joint project, Fraunhofer IKS and ATOSS Software have developed an AI-controlled shift planning agent that automates staff scheduling while demonstrating remarkable scalability.

 

AI in Workforce Management
 

© iStock.com/Drbouz

Machine learning in medicine / 24.7.2024

Data-driven diagnostics improve the health of premature babies

Babies born prematurely, i.e. before their organs have fully developed, often suffer from various health problems, known as morbidities. These rarely manifest alone, but often occur simultaneously. Researching connections or even patterns in their co-occurrence helps to develop more effective and more personalized care for premature babies. A project report.

Machine learning in medicine
 

© iStock.com/Orientfootage

Safe Intelligence
online magazine

Would you like to find out more about the research of Fraunhofer IKS on AI in medicine? Then take a look at our Safe Intelligence online magazine:

To our Online Magazine

Let’s collaborate for a more trustworthy healthcare future!

Reach out to learn how our innovative solutions can elevate your R&D efforts. Together, we can transform healthcare.

Thank you for your interest in the Fraunhofer IKS.

We have just sent you a confirmation e-mail. If you do not receive an e-mail in the next few minutes, please check your spam folder or send us an e-mail to business.development@iks.fraunhofer.de.

* Required

An error has occurred. Please try again or contact us by e-mail: business.development@iks.fraunhofer.de