Design and Evaluation of Interfacing Concepts Between MATLAB and Python for Analysing Medical and Telehealth Data

Python and MATLAB are common tools used for predictive modelling applications not only in healthcare. At the AIT Austria Institute of Technology (AIT) the predictive modelling group works with both tools. None of the two tools is optimal for all tasks along the value chain of predictive modelling in healthcare. However, a combination of both tools would help a lot. The aim of this thesis was to nd di erent concepts to combine MATLAB and Python. These concepts should be used to extend the MATLAB-based \Predictive Analytics Toolset for Healthcare" with Python functions. To design interfacing concepts, pre-existing interfaces between MATLAB and Python have been evaluated. Thereafter, more comprehensive interfaces have been designed to exchange even complex data formats such as MATLAB tables. As a result of this thesis, concepts were developed to exchange data between MATLAB and Python without any losses. The interface concepts have successfully been validated using the PhysioNet/CinC Challenge 2021 evaluation function, as well as in a real-world natural language processing scenario. As a conclusion of this thesis, it is possible to integrate Python modules in the MATLAB toolset of the AIT. It could be shown that even complex data formats such as MATLAB tables can be exchanged. Further improvements especially in terms of performance are required if large datasets need to be exchanged.