What can I bring to an initial discussion?

A short description of the application, the technical challenge and the outcome you want to achieve is enough to start. If available, describe the types of data involved and your current stage of development.

Do we need a complete dataset before contacting DFKI?

Not necessarily. An initial discussion can help clarify the data requirements and whether the proposed research direction is a suitable fit. Feasibility will depend on the specific project.

Can we discuss proprietary or sensitive data?

Describe the data categories and project requirements first. Do not submit patient records, personal health information or confidential datasets through this contact form. Any data-sharing arrangements must be agreed separately.

Does prototype development mean a market-ready medical product?

No. A research prototype or proof of concept demonstrates an approach within an agreed scope. Productisation, clinical evaluation, regulatory requirements and deployment are separate considerations.

Can the collaboration focus on one specific technical challenge?

The initial discussion should establish whether DFKI's expertise matches the challenge and what a suitable project scope might be.

Medical AI Research & Development

Your Medical AI R&D Partner — From Research to Prototype

Develop and validate medical AI with DFKI. Bring together expertise in medical imaging, sensor fusion and human movement analysis to turn a complex research challenge into a practical prototype.

For medtech, healthcare, pharmaceutical and industrial R&D teams.

DFKI researcher and clinician reviewing an AI-supported patient overview on a hospital workstation.

Where Projects Get Stuck

A Promising Idea. A Difficult Technical Next Step.

You may have a product concept, a research question or valuable data. The next challenge is understanding what is technically feasible—and how to demonstrate that it works.

Validate feasibility

Explore whether AI can address your specific challenge before committing to a larger development programme.

Make complex data useful

Work with medical images, physiological signals or multimodal sensor data that standard tools do not handle well.

Move beyond static analysis

Investigate motion, longitudinal change and relationships across time, patients or sensing modalities.

Build a demonstrable prototype

Translate a research approach into something your team can evaluate, test and develop further.

Capabilities

Specialist Expertise for Medical AI Development

Four areas of applied research expertise that most often shape a medical AI development programme.

Medical Image Analysis & Registration

Develop methods for analysing medical images and aligning scans across time or modalities. Relevant challenges include motion correction, longitudinal monitoring and quantifying morphological change.

Medical Image Analysis Medical Image Registration

Multimodal Sensor Fusion

Explore how different sensor streams can be combined to provide a more complete picture than individual signals alone.

Medical Sensor Fusion Biomedical Signal Processing

Human Movement Analysis

Investigate movement patterns and functional change for applications such as rehabilitation research, wearable sensing and monitoring.

Biomedical Signal Processing

AI Prototyping & Technical Evaluation

Test feasibility, define evaluation criteria and develop prototypes that help teams assess an approach before further investment.

AI Pilot Program

Project requirements may also include privacy-preserving methods, explainability and physiological plausibility. The relevant approach depends on the data, intended application and research question. Technical evaluation and research validation are not regulatory approval, clinical certification or a guarantee of patient outcomes.

DFKI researcher and clinician evaluating an AI decision-support prototype in a hospital emergency department. — Applied research in a clinical environment: evaluating an AI prototype with clinicians.

Collaboration Framework

A Clear Path From Challenge to Prototype

An illustrative collaboration framework — not a fixed commercial package.

Define the challenge

Clarify the application, available data and what a successful outcome would demonstrate.

Assess the approach

Explore feasibility, data requirements and relevant AI or signal-processing methods.

Develop and evaluate

Build the research approach and evaluate it against agreed technical criteria.

Demonstrate the next step

Create a prototype, proof of concept or research output to support the next development decision.

The scope and outputs are agreed for each collaboration, based on the organisation's needs and technical maturity.

Case Examples

Research Connected to Real-World Development

Selected industrial research collaborations documented by DFKI.

DFKI × Xenoma

AI Wearable Technology

Challenge: Analyse human movement in real time from wearable sensor technology rather than from laboratory equipment.
DFKI's contribution: DFKI contributed AI research for intelligent wearable sensor technology capable of analysing human movement in real time.
Why it is relevant: Shows how an industrial partner can work with DFKI on sensing and movement-analysis research inside a product development context.

Read the DFKI announcement

DFKI × Hitachi

AI for Occupational Health

Challenge: Measure physical workload and ergonomic risk from human movement in real working environments.
DFKI's contribution: DFKI contributed AI-powered wearable sensing research to assess physical workload, ergonomic risk and movement.
Why it is relevant: Relevant for organisations exploring health-related sensing applications built on multimodal physiological and motion data.

Read the DFKI announcement

DFKI × Hitachi × Xenoma

Intelligent Sensor Suit

Challenge: Combine AI, wearable sensing and smart textiles for real-time movement analysis and activity recognition.
DFKI's contribution: DFKI contributed applied AI research within a multi-partner industrial research collaboration.
Why it is relevant: Illustrates consortium-style R&D collaboration across several industrial partners and research topics.

Read the DFKI announcement

DFKI researcher and clinical partner in a hospital resuscitation room during a joint research project. — DFKI researchers working with clinical partners on site.

Project Fit

Is This the Right Fit for Your Project?

This conversation is for organisations exploring a concrete medical AI research or development challenge.

You are developing a medical, healthcare or sensing application.
You have relevant data, or need to understand the data requirements.
You need specialist research expertise beyond a standard software implementation.
You want to assess feasibility, evaluate an approach or build a prototype.
You are exploring an industrial or funded research collaboration.

This page is for research and development enquiries. It does not provide personal medical advice or downloadable medical software.

FAQ

Frequently Asked Questions

Start a Conversation

What Are You Trying to Develop or Validate?

Tell us about your application, the technical challenge and the outcome you are aiming for. This helps establish whether DFKI's research expertise is a relevant fit.

Or email us directly: philipp.koch@dfki.de · marja.fleitmann@dfki.de