Tweet This
An artificial intelligence (AI) designed by researchers in the UK beat doctors at predicting heart disease. They have developed a model for coronary artery disease using an AI and a relevant set of 600 variables, that outperformed a model built using 27 variables chosen by medical experts (such as "age", "gender" and "chest pains). The AI also identified certain non-obvious factors such as “home visit from their GP”, as a good predictor of patient mortality.
Although AI research is still in its infancy, these early studies already establish how AI is set to revolutionize cardiac care. This is particularly relevant today as cardiovascular diseases are still the number one killer in the world, resulting in 31% of all global deaths, and is also the most expensive condition to treat.
Here are 4 different ways in which AI is revolutionizing cardiovascular care ,
1. AI aided diagnostics
The biggest impact of AI in cardiac care will be in diagnosing cardiovascular diseases. Typical diagnostic pathways involve three stages. The first stage is measuring an electrocardiogram (ECG) at rest. Anomalies in this stage results in a combination of semi-invasive tests such as ECG stress test, stress echocardiography, and chest CT scan. Anomalies in this tests lead an invasive angiography. Researchers and companies, are already using AI to predict the anomalies quickly, cheaply and accurately without using the third invasive step.
Cardiologs’ AI-powered platform is CE Marked and FDA approved and uses ECG data to detect atrial fibrillation (most common abnormal heart rhythm) with 91 percent accuracy as compared to 59 percent using standard methods.
A team of researcher led by Dr. Partho Sengupta, developed an AI that could potentially diagnose diastolic dysfunction in patients from two-dimensional cardiac ultrasound images.
2. AI aided cardiac imaging
AI is enhancing live visualization of the heart, by color coding the different heart chambers in real time from low-resolution grayscale echocardiography images. This technology was not available a couple of years ago and is now vastly improving the efficiency of clinical workflow for both cardiologists and radiologists. Philips' echocardiography uses an AI called HeartModelᴬ⋅ᴵ⋅ to additionally build a 3D model of the patient's heart from echocardiography images.
Zebra's AI-powered radiology assistant recently received FDA 510(k) clearance for its coronary calcium scoring algorithm which can predict the likelihood of coronary artery disease from a non-contrast chest CT scans. HeartFlow offers a similar AI-powered FDA cleared software solution for identifying coronary artery disease using chest CT scans but provides a more comprehensive 3D output for the cardiologists.
Arterys’ AI-powered Cardiac MR Suite is FDA 510(k) approved and allows cardiologists to view the patient’s heart in 4D, by color coding the blood flow in the heart in real time from magnetic resonance imaging (MRI) images.
3. AI aided therapy selection
One of the biggest challenges for cardiologists, hospital systems, patients and their families is to determine the risk and cost of care pathways recommended by the cardiologists. KenSci reportedly uses machine learning to predict patient risks of acquiring diseases including heart disease.
Babylon Health’s Healthcheck is a chatbot that uses AI to give patients a quick assessment and understand their health. Corti labs perhaps take this a step further and use AI to recognize out-of-hospital cardiac arrests and help emergency dispatchers make critical life-saving decisions.
4. AI aided continuous monitoring
Consumer grade wearable devices such as Fitbit and Apple Watch that continuously monitors a consumer’s heart rate, activity and location, serves as a good platform for building AI tools that can predict early warning signs of lifestyle diseases including cardiovascular anomalies. Cardiogram’s DeepHeart that works with Apple Watches is a semi-supervised AI learning for cardiovascular risk prediction. As consumer medical devices start using more accurate single lead ECG sensors, such as Apple Watch 4 and AliveCor’s Kardia Band the output from the AI will hopefully be even more reliable.
Smart-clothing companies such as Hexoskin, OMsignal and Think Biosolution is also developing AI for monitoring cardiovascular and other lifestyle diseases form tracking biometric and user behavior.
AI is truly at the verge of redefining how cardiovascular care is delivered to patients. Companies and researchers have implemented AI in every step of the process, from continuous monitoring of basal heart rate for early warning signs to quick and efficient noninvasive diagnosis of cardiac conditions. AI is also making later stages of the care pathway more efficient, such as real-time visualization of the cardiac anomaly and subsequent therapy selection. However, the question that remains to be answered is will this advanced technology, in the long run, be able to bring down cost and time of cardiovascular care for patients.
