Visualization, interpretability and trustworthiness of AI-based features for the characterization of musculoskeletal disorders

  • Heinrich, Mattias (Principal Investigator (PI))
  • Glüer, Claus Christian (Consortial Partner)
  • Koch, Reinhard (Consortial Partner)
  • Engelke, Klaus (Consortial Partner)
  • Meyer, Carsten (Consortial Partner)

Project: Projects with Federal FundingProjects with Federal Ministry Funding: BMBF

Project Details


The ARTEMIS network has set itself the goal of using the enormous potential of artificial intelligence together with advanced image processing methods for imaging and modeling in the field of musculoskeletal disorders. The methods of artificial intelligence are used in particular for osteoporosis, but also for degenerative diseases of the lumbar spine and the hip joint, which are assessed using computer tomography (CT). In this subproject, the following advanced AI learning methods will be developed and evaluated: 1) visualization techniques that highlight areas of attention and neural network activation to improve their interpretability, 2) robust AI models that can learn from limited annotations using partial self-supervision in large image databases, and 3) reduction of computational costs to train lighter models applicable to high-resolution 3D CT scans in clinical practice.
Effective start/end date01.10.2031.03.24

UN Sustainable Development Goals

In 2015, UN member states agreed to 17 global Sustainable Development Goals (SDGs) to end poverty, protect the planet and ensure prosperity for all. This project contributes towards the following SDG(s):

  • SDG 3 - Good Health and Well-being
  • SDG 9 - Industry, Innovation, and Infrastructure

Research Areas and Centers

  • Centers: Center for Artificial Intelligence Luebeck (ZKIL)

DFG Research Classification Scheme

  • 205-07 Medical Informatics and Medical Bioinformatics
  • 409-05 Interactive and Intelligent Systems, Image and Language Processing, Computer Graphics and Visualisation