Comparison and validation of different implementations of ML-based analysis framework Comparison and validation of different implementations of ML-based analysis framework

Different versions of the same algorithms, for the analysis of complex biomedical data, have been implemented, using different programming languages, solvers and parallelisation mechanisms. The central algorithm is Unsupervised Multiple Kernel Learning, but a whole analysis framework building upon it has been developed.

The main objective of this project is to design and implement a testing framework for a thorough version evaluation and comparison, in terms of performance, convergence and results. A secondary objective would be to replicate a recently-developed block of the analysis framework, that is currently only implemented in a Matlab environment, in the python language.

Preferred skills: comfortable with Matlab and python.

 

References:

[1] Sanchez-Martinez S, Duchateau N, Erdei T, Fraser AG, Bijnens BH, Piella G. Characterization of myocardial motion patterns by unsupervised multiple kernel learning. Med Image Anal. 2017 Jan;35:70-82. doi: 10.1016/j.media.2016.06.007. Epub 2016 Jun 11. PMID: 27322071.