Cartography of the heart
Like everything now infused with technology, health care is becoming more precise, allowing practitioners to see and treat illnesses with heightened levels of exactness. Researchers and clinicians are developing and using technology to better diagnose conditions and more accurately assess protocols and recovery. These technological advances — like 4D ultrasound combined with 3D strain mapping to chart regional heart motion — identify granular markers, rather than general changes, to better pinpoint pathology and improve diagnostics and treatment.
Take the example of heart disease. Coronary artery disease (CAD) is the leading cause of death in the United States, responsible for roughly one-third of all deaths in people over the age of 35. CAD occurs when plaques form and rupture, leading to thrombus formation (blood clots) that blocks the downstream flow of blood. Patients with atherosclerosis — a disease associated with plaque buildup — can experience complications such as heart attacks and strokes, which account for 84.5 percent of all cardiovascular deaths worldwide and 28.2 percent of mortality from all causes. Diagnosing, treating and managing cardiovascular disease is projected to cost more than $800 billion annually by 2030 in the United States alone.
After a heart attack, the heart heals by “remodeling” itself in response to the injury — a complex, multifaceted process involving cellular, molecular, mechanical and neurohormonal changes. Early remodeling begins with an inflammatory response, followed by scarring. While both are important in healing the dead issue, a poorly structured scar can lead to expansion of the dead tissue and detrimental long-term effects.
The most popular imaging technique for assessing cardiac function is echocardiography, involving an ultrasound scan of the heart. Different tissues have unique acoustic properties; echocardiography takes advantage of this fact to obtain heart measurements. This helps researchers locate necrotic (dead) tissue, identify variations in wall thickness of the heart muscle, and detect discrepancies in the ability of the heart to contract and function.
Technology we have developed integrates noninvasive 4D ultrasound with 3D mapping of the heart strain to create an enhanced “cartography” of the heart. This technique characterizes the cardiac mechanics within and around dead tissue during the remodeling process, improving upon conventional 2D analyses. This approach could better predict remodeling outcomes, as well as allow physicians to identify risk and better manage treatment. Purdue researchers are working to apply machine learning and artificial intelligence (AI) to the tracking techniques to reduce further-analysis times and enhance accuracy.
Our technology also is applicable to other types of cardiac disease — cardiac hypertrophy (abnormal thickening of the heart), cardiac valve dysfunction (when the internal valves do not open or close properly), and right-sided heart failure associated with pulmonary hypertension.
We envision a world where patients can get fast heart scans. The person taking the images will not have to be highly trained, thanks to assistance from an imaging system that acquires everything needed within a few minutes. The images can then be automatically processed and analyzed with AI and machine learning so clinicians can make treatment decisions with confidence, eliminating costly and invasive follow-up diagnostic tests and their complications.
Craig J. Goergen
Leslie A. Geddes Associate Professor
Weldon School of Biomedical Engineering
Principal Investigator
Cardiovascular Imaging Research Laboratory
College of Engineering, Purdue University
Related Links:
The power of 4D technology advances care for heart patients
Video: Cardiac hypertrophy in 4D
Video: 4D strain analysis of cardiac and vascular pre-clinical image data
Multi-modality method offers precise characterization of atherosclerosis progression
