The goal of machine learning is to program computer systems by providing them with example data so that the program can adapt its behavior by learning from examples, much like a child learns from the world around him or her. In other words, machine learning concerns the recognition of complex patterns in observed data in order to automatically make decisions about patterns hidden in this data. Practice of medicine is filled with countless examples of this process. However, currently the task of pattern recognition is performed by humans. Medical systems (e.g., CT and MRI scanners, EKG machines, EEG and other physiologic monitors) produce huge amounts of data that often contain abundant information; only a fraction of this information is important for the diagnosis. Can machine learning automate and help to extract the informative parts in medical data? What are the various paradigms for organizing, extracting relevant information, and teaching a program to make automated decisions after a session of example-driven learning? These are the central questions that we will be addressing in this course.


This course will help a student acquire knowledge, skills and insight in machine learning in the domain of medical imaging and sensor data.

Examples of polyp, ischemic stroke, retrained surgical hardware (sponge) after surgery, and subarachnoid haemorrhage. Can a computer learn to automatically detect them after looking at thousands of such examples?


The course addresses the machine learning paradigm for classification and pattern recognition in vast amounts of data. For example, how can we program a system in such a way that anomalous frames in an endoscopic video stream are detected automatically? In the machine learning paradigm, the solution is found by providing the system with a large number of examples from which, in a training stage, the system tries to generalize. Once trained, the system is then ready to be used on any individual image in the operational stage.

This course is an introduction to machine learning. The topics are structured as follows:

  1. Linear and quadratic classification applied to pixel classification
  2. Non-parametric (distribution free) classification
  3. Overfitting and cross validation
  4. Feature Extraction and Selection
  5. Cluster analysis
  6. Decision trees, Random Forests, Deep learning (convolutional neural networks)

The course is organized in a daily pattern consisting of:

  • Clinical context and clinical relevance of the topic of that day
  • Introduction of the theory
  • Matlab hands-on exercise
  • Re-cap, closing remarks, and closure of the day

The students will be expected to program extensively by themselves. This will enable them to quickly apply their newfound knowledge and learn the material from a practical stand point.


This course is intended for PhD students and young scientists with a background in Biomedical Technology, Technical Medicine, and related fields. It is primarily intended for students who have a need for more insight in machine learning with application to medical image data.

The course assumes a strong background in programming and students will be driven by a set of well selected programming exercises in addition to theoretical exposure.

Some experience and affinity with Matlab and with vector-matrix calculus is a prerequisite for this course.


The course is organized by:

  • Prof Rajiv (Raj) Gupta, MD, PhD – Massachusetts General Hospital and Harvard Med School
  • Dr Ferdinand (Ferdi) van der Heijden, PhD – University of Twente

under the auspices of the Technical Medicine department of the University of Twente in collaboration with the Medical Imaging group of the MIRA Institute for Biomedical Technology and Technical Medicine.


For UT MSc students (BME and TM):

  • Attending the course is free of charge.
  • The course will represent 1,5 EC.
    • Note for TM students: if you have already enrolled yourself for the course “3D Computer Vision for Medical Applications” (5EC), then you are automatically enrolled for this Machine Learning course. There is no need to enrol yourself again.

For other participants:

  • The price for this five-day course is €550.
  • Accreditation has been requested from the NVvTG.

The upcoming course dates and the enroll form can be found here.

In case you want to be contacted in order to answer any further questions, please fill out this information request form. We will contact you as soon as possible.