Data-Driven Process Modeling for Prediction of Outcomes for Cardiac Differentiation Experiments

Cardiovascular diseases (CVD) are the leading cause of death worldwide, meaning that more people die each year of CVD than of any other cause. These diseases can lead to heart attacks, which can result in the loss of more than one billion heart cells and lead to congestive heart failure. Engineered heart tissue produced by differentiation of human induced pluripotent stem cells may provide an encompassing treatment for heart failure due to CVD. However, significant difficulties exist in producing the large number of cardiomyocytes needed for therapeutic purposes through differentiation protocols. Data-driven modeling with machine learning techniques represents a potential tool for elucidating the relationship between experimental conditions and the outcomes for cardiac differentiation experiments. Using data from previously conducted cardiac differentiation experiments, we have developed data-driven modeling methods along with feature selection for determining which experimental conditions are most influential on and predictive of the final cardiomyocyte content of a differentiation experiment. With those identified conditions, we were able to build classification models that can predict whether an experiment will have a sufficient cardiomyocyte content to continue with the experiment on the seventh (out of 10) day of the differentiation with a 90% accuracy. This early failure prediction will provide cost and time savings, as each day the differentiation continues requires significant resources.