Clinical integration of deep learning models for medical image analysis

Open Project: Clinical integration of deep learning models for medical image analysis

Type: Master Thesis / IDP

Contact Person: Fabian Bongratz, Christian Wachinger

This student project aims to apply deep learning models for the analysis of medical images in the radiology environment. The project’s primary focus is on executing pre-trained deep learning models (in PyTorch) to analyze medical images from a picture archive. The network communication for accessing scans will be done via DICOM query/retrieve. Students will utilize Python for the core programming tasks, Bash for scripting, and functions from the DICOM library. The image analysis task will focus on the segmentation of CT and MRI scans. 

By the end of the project, students will not only gain hands-on experience in applying MLOps techniques to real-world medical imaging tasks but also contribute to the improvement of clinical diagnosis through innovative technological integration.


  • Proficiency in Programming and Scripting: Advanced skills in Python and Bash for automation and scripting.
  • Experience with Machine Learning Operations (MLOps): Practical knowledge of MLOps practices, including environment setup, model deployment, and monitoring. 
  • Familiarity with Medical Imaging: Understanding of medical images. Optional is familiarity with DICOM. 
  • Experience with Deep Learning: Experience with executing deep learning models (PyTorch).
  • Familiarity with Windows and Linux platforms.

What we offer to you:

  • We offer a project (master thesis or interdisciplinary project) on the application of deep learning models on radiology data at Klinikum rechts der Isar.
  • You are an active member of our international and interdisciplinary research team working on the transfer of cutting-edge technology into clinical environments.
  • You work in close collaboration with computational scientists from TUM as well as medical experts from radiology. 
  • You will learn to know and integrate the latest deep-learning architectures, medical file formats, and ML workflows.