Computational Physiology for Medical Image Computing

1) Physiological imaging: Computational models for extracting local physiological information from imaging modalities such as magnetic resonance, computed tomography, or positron emission topography (e.g., MR and PET models for blood flow, metabolism, tissue microstructure)

2) Organ models: Computational models describing anatomy and function at the organ level as well as its variation across the population (e.g., 3D and 4D population atlases describing structure and shape)

3) Disease and diagnostics: Computational models describing pathophysiological processes and the progression of diseases (functional models from biophysics and theoretical biology, empirical models for clinical decisions)

4) Use cases, for example, from cardiac imaging, tumor analysis, and neuro-degenerative diseases

Upon completion of the module, the participants will be able to use computational models for extracting diagnostic information from different types of clinical image data sets. These data sets may provide information, for example, about blood perfusion or microstructural tissue properties, about metabolic processes, or patterns of disease progression. The participants will understand the physiological concepts underlying the computational algorithms employed, and will know of advantages and shortcomings of different modeling strategies. This will allow them to analyze clinical imaging protocols with respect to the underlying physiological information, and to propose diagnostic algorithms that combine anatomical and physiological information of different imaging modalities.

CAMP 1, or other introductory lecture on image processing; proficiency in a computing language such as Matlab, Python, or CPP.

Weekly lecture, discussion of project work; final presentation (written and oral)

Slides, additional reading material (review papers, book chapters); publicly available clinical data sets and software tools.

An updated list will be maintained on the homepage of the lecture

During the course of the module the participants will define and pursue a small project dealing with the analysis of clinical image data. They will implement and analyze computational algorithms for processing these data using the methods and tools presented in the lecture. They will pursue this project work in small teams and will actively participate in the discussion of results from other teams.
Each team will document their project work and the contributions of the different team members. They will present their results at the end of the course in an oral presentation with subsequent discussion. They will be evaluated by their model implementation, their model analysis, and their presentation. Moreover, they will be evaluated based on their participation in the discussion of the projects of other participants.