Pre-Clinical Hemodynamic and Imaging Analysis for Cardiogenic Shock Phenotyping

Heart failure and cardiogenic shock, a life-threatening condition in which the heart suddenly fails to pump enough blood to sustain the body, carry high mortality even with the most advanced treatments available. Managing cardiogenic shock often requires mechanical devices to temporarily support the heart, but decisions about which device to use, when to act, and how to adjust treatment are currently made with limited quantitative guidance. This project, conducted through a collaboration between the MIT Edelman Laboratory (Institute for Medical Engineering and Science) and King Abdulaziz City for Science and Technology (KACST), develops AI-driven tools to address this gap. Using pre-clinical cardiogenic shock model with hemodynamic waveform data and cardiac imaging, the team is building machine learning models that extract standardized, quantitative metrics from echocardiography, characterise the progression of cardiogenic shock, and predict responses to mechanical circulatory support. The project aims to produce validated AI phenotyping models and a shared analytical pipeline that will form the technical foundation for a broader clinical validation program in Saudi Arabia.