Content

The course will be organized around three main pillars 

• Advanced Statistics in Medical Physics: statistics are heavily used in medicine in general and in medical physics in particular. This includes statistical significance of laboratory and clinical experiments; quantification of risk; estimation and propagation of uncertainties (type A and type B uncertainties); probabilistic problem solving.  
• Monte Carlo techniques. Monte Carlo engines are often used as black box in clinical practice and R&D. The goal is to provide insights in the theoretical grounds of Monte Carlo simulations and also in the practical specificities of modern implementations. This includes: random number generation; sampling techniques (inverse transform, rejection technique); variance reduction; statistical error estimation (direct or batch technique); problem definition (geometry and materials); use of specialized hardware (many-core processors and GPU). Practical examples and important results are illustrated in radiotherapy, nuclear medicine and radiology.
• Introduction to Machine Learning:
  • Context and purpose of Artificial Intelligence, Machine Learning, and Deep Learning.
  • The various types of learning problems (supervised, non-supervised, reinforcement, transfer).
  • The various types of data sets and their purpose (training, validation, test).
  • Short introduction to optimization.
  • A few techniques:
    • Principal component analysis, linear discriminant analysis.
    • Decision trees and random forests.
    • Support Vector Machines.
  • Neural networks, from single artificial neuron to deep (convolutional) networks.
  • Interpretation of the results (ROC curve/sensitivity/specificity/…).
  • Specifics of data collection for AI/ML/DL in medical physics (access to patient data and the importance of consistent patient data; how to guide efforts in structured reporting).
  • Big data and data preprocessing (images, radiomics,.. ).

Course material

The course associates regular theoretical lectures and practical sessions.  

All theoretical lectures are either pre-recorded or recorded (if pre-record is not available). Therefore, in-class teaching can be adapted depending on the requests of the students present in class. When a pre-record is available, we favor a dynamic teaching with large developments on the black board on specific parts of the course. The students are encouraged to vision the pre-recorded courses before the in-class session so that they can ask specific questions and developments.

Physical presence is mandatory for the practical sessions. The schedule will be given during the first course. No streaming nor recording are foreseen for the practical sessions (one session for dose calculation lab; one session for margin lab).

The introduction to the course (course schedule; presentation of summary and teaching material; evaluation methodology; practical considerations) will be streamed and recorded.

After the introduction, no streaming is foreseen for the courses when a pre-record is available. This is the default format (no streaming, but a pre-recorded course). In the case a  pre-record is not available, the course will follow a classic format with a power point presentation. In the latter case (no pre-record), and only in that case, the courses will be streamed as well. It will be made clear to all students when a streaming option will be made available. But the students should assume there is no streaming option.  There will be many possibilities for the students having difficutlies to come to the course to ask their questions. Specific (streamed) sessions could be envisaged for answering questions.  

The contents that will be subject to evaluation are the ones and only the ones available in recorded material (slides and explanations).

Content

For each part of the course (statistics, Monte Carlo techniques, Machine Learning) a series of exercises is proposed. They should be solved by using the adequate computing material.

Course material

Series of exercises

Format: more information

There will be 2 challenges (one individual report per challenge)

–Monte Carlo simulations

–Machine learning

Python langage is supposed to be known. Please inform us if you have issues with Python!