ML-Driven Localization of Infection Sources in the Human Cardiovascular System

Short Description of the Paper

In vivo localization of infection sources is essential for effective diagnosis and targeted disease treatment. In this work, we leverage machine learning models to associate the temporal dynamics of biomarkers as detected at static gateway positions with different infection source locations. In particular, we introduce a simulation that models infection sources, the release of biomarkers, and their decay as they flow through the bloodstream. From this, we extract time-series biomarker data with varying decay rates to capture temporal patterns from different infection sources at specific gateway positions. We then train a stacked ensemble model using LightGBM and BernoulliNB to analyze biomarker time-series data for classification.

Data and Code

We publish our simulation data and the Python code to process it as described in the paper here on Zenodo and in a GitHub repository under the CC BY and MIT licenses, respectively.

Supplementary_Data.pdf provides an overview of biomarker distributions from different infection sources and classification results for the 70/30 split, complementing the 80/20 results in the main manuscript.

Contact

If you have any questions or suggestions for improvements, feel free to contact me.

• Saswati Pal
  
  •  Email: pal@ccs-labs.org