LIVING HEART: USING PREDICTIVE AI/VR MODELS TO REDUCE UNCERTAINTY IN CARDIOVASCULAR DIAGNOSIS AND TREATMENT

Abstract

The ability to create predictive, validated, multiscale models of the human heart holds promise to minimize speculation in preventing, managing, and treating heart disease by enabling clinicians to determine the optimal treatment for a patient with CVD. Toward this goal, The Living Heart Project has brought together a multidisciplinary team of experts representing research, industry, clinical practice, and regulatory authorities to realize this goal. A predictive heart model, released in 2016, is an anatomically accurate multiphysics model of a healthy human heart with dynamic behavior governed by sound physiological principles. Advanced imaging modalities 3DCTA and DT-MRI were used to define the physical attributes of the heart structure, while the conservation laws of continuum mechanics capture the physical response. The LHM contains well-defined anatomic details including internal structures (e.g., heart valves, chordae tendineae, coronary arteries, and veins) and proximal vasculature (e.g., aortic arch, pulmonary trunk, and SVC). Cardiac contraction is driven by waves of electrical excitation traveling across the heart to generate physiologically observed wave propagation patterns. A closed system of fluid cavities and fluid links models blood flow. The fluid and solid models are directly coupled and the systemic and pulmonary circuits are endowed with vascular compliances and flow resistances that can be modified to simulate exercise, hypertension, and other physiological states. The LHM is thus a dynamic model of a human heart that has been used as a powerful diagnostic and treatment planning tool. Due to the modular nature of the Living Heart Model, it can be modified or extended to simulate a wide range of clinical applications. Using patient-specific data, it can be modified locally or globally to simulate the effects of myocardial infarction or other forms of heart failure. Efficacy of medical devices for improving/restoring cardiac function can be evaluated. The external circulatory system can be enhanced to study peripheral circulation and organ perfusion phenomena using patient-specific flow velocity data, or determine the sensitivity of cardiac function to various peripheral treatment options. The LHM can include models of cellular and subcellular electromechanical phenomena, such as may be required to study arrhythmias and their induction by pharmacological or pathophysiological agents. It is possible to replicate the pre-treatment characteristics of patients with specific congenital or adult cardiac conditions and thus predict patient-specific treatment outcomes.View Large Image Figure ViewerDownload Hi-res image Download (PPT)