Which segmentation tool should you use?
A nyone who annotates medical image data for the development of AI applications in medical technology knows that there is a wide range of segmentation tools to choose from. Therefore…
Proven medical AI development processes.
Selection of our clients:
MxDB Digital Biobank
Enable the full potential of AI in healthcare through a structured development process from clinical data acquisition to regulatory approval. We offer “gold standard” training datasets, high-quality annotations and feedback of clinical experts for the development and approval of AI-based medical technology.
MxDB Advisory
Benefit from proven processes that we implemented successfully in various projects:
- FDA and MDR regulatory best practices for market approval
- Best practice for medical AI development, verification and validation
- Technical and clinical know-how for a successful implementation of AI in the clinical workflow
Advisory & Solution
for AI development and approval
MxDB Solution
We provide support throughout the entire process of idea generation, development, validation and approval:
- Acquisition of clinical data for training of medical AI
- Ground truthing and annotations for validation and approval of medical AI
- Ensuring regulatory approval and clinical evidence by clinical performance testing
Our Services
For high quality clinical data,
better medical AI
and accelerated time-to-market
Clinical Workflow Analysis
Expert evaluation of the common clinical workflow (best practice)
Clinical Use Case Analysis
Testing of the planned clinical use case to optimize the standard of care
Integrating clinical data centers
Framework contracts enable data access
Acquisition of the clinical raw data
Ensure fast and reliable data exchange
Creating Annotation Guidelines
Standardizing clinical training data
Annotating Clinical Data
M3i ensures “gold standard” annotation with clinical experts
Data Quality Management
Highest quality through controlled processes
Validation & Approval
Ensure regulatory approval and clinical evidence by clinical performance testing
Discuss your project:
Benefit from proven medical AI development processes!
”We need a reliable partner who can provide us with clinical data in an ethical and legally compliant manner. We have found this partner in M3i.
Andreas GieseManaging Director, Snke OS
Case Study
AI-assisted spine surgery planning
The TRANSITION project is a collaboration between Brainlab AG and M3i. In the project, we are developing a software for the automated 3D-planning of dorsally instrumented surgeries on the spinal column in all sections. This software provides essential information about the individual patient anatomy. After learning from experts and gold standards, the software is able to automatically – without manual planning of the physician – suggest improvements of the trajectory and to choose the best type of implant. Patients benefit from shorter surgery times and a reduction of intraoperative X-ray radiation, which is also beneficial for the medical team.
The role of M3i in the project:
- quality checking, annotating and segmenting clinical data (25.000 labels)
- validating the algorithm with clinical experts
- optimizing its user-friendliness
Case Study
Improving tumor detection through AI-assisted confocal microscopy
The KONFIDENT project aims to reduce the adoption barriers for using confocal microscopy during tumour surgeries where skin tissue is removed. Currently, providing feedback of the pathologist to the surgeon takes from a few hours to several days. During this time, the patients surgical wounds cannot be closed. Multiple surgeries to remove the malignant skin tissue completely are necessary. With the help of AI, both the in-vivo and ex-vivo tissue analysis can be augmented and surgeons can more precisely define the tissue that needs to be removed during the surgery.
M3i’s role in the joint project with Vivascope, Munich Innovation Labs and Heidelberg University Hospital:
- quality checking, anonymizing, and annotating and segmenting clinical data
- enabling data exchange between hospitals and industry partners
- validating research results in other clinics
- perform realistic usability evaluations for regulatory approval of the AI
Case Study
Intelligent solution for structured medical reporting
In the KI-BETA project, an existing intelligent solution for structured medical reporting will be further developed. The first version of the platform helps radiologists to produce high-quality guideline-compliant reports. The aim is to extend the platform, so it can be used for joint radiological-histological findings in a structured way. With the help of AI algorithms, the radiological and histological images could be analysed (semi-)automatically. This increases the efficiency and quality, and thereby improves the treatment of patients.
M3i’s role in the joint project with Smart Reporting:
- quality checking, anonymizing, and annotating and segmenting clinical data
- validation of the developed algorithm by our broad network of clinical experts
- contribute with expertise about machine learning and knowledge on efficient and legally compliant development
- realistic usability evaluation with eye tracking
Get started: Our medical AI advisory
The development of data-based medical device algorithms can be extremely challenging. Discuss your project with us and see how we can support your development processes.
Why talk with us?
Benefit from proven medical AI development processes.
