People

Atif Khan

Project: mitoML: Machine learning approaches to understand mitochondrial disease pathology 

Mitochondrial diseases are currently untreatable, one of the reasons for this is that we don’t completely understand their causes and effects (i.e. pathology). To study these diseases, scientists use observational data from various patient groups and controls, find patterns in this data and determine thresholds that can classify/quantify these diseases. One type of observational data relevant in mitochondrial diseases is the imaging of protein expression in muscle and neuronal cells that manifest the function/dysfunction of these cells. In my PhD I train various CNN and GAN models on this high dimensional data to classify/quantify mitochondrial diseases and extend this to multimodal ML analysis to include other clinical and omics data.