New AI Tech May Help Predict Heart Attacks Five Years in Advance
Keerthi Vedantam is a bioscience reporter at dot.LA. She cut her teeth covering everything from cloud computing to 5G in San Francisco and Seattle. Before she covered tech, Keerthi reported on tribal lands and congressional policy in Washington, D.C. Connect with her on Twitter, Clubhouse (@keerthivedantam) or Signal at 408-470-0776.
What if you could predict a heart attack?
Los Angeles’ Cedars-Sinai Medical Center announced this week the development of a tool that uses artificial intelligence to measure artery health and detect future cardiac risks in seemingly healthy patients as far as five years in advance.
Heart attacks are often caused by plaque deposits in arteries, the muscular-walled tubes that carry oxygenated blood throughout the body. These deposits constrict blood flow and raise the risk of potential heart problems. While doctors can use CTA scans to create 3D images of a patient’s arteries and measure the density and composition of such plaque, it can be a complicated and time-consuming process.
Cedars-Sinai’s new Autoplaque device uses AI to detect these potential threats in a matter of seconds, arming doctors with a better form of preventative care for their patients. The device’s algorithm was trained on CTA scans from more than 900 patients, featuring over 5,000 lesions already seen by doctors. When given a new set of data, investigators found that the device was able to measure how much arteries had narrowed down to an exact percentage point.
“The AI tool was able to quantify the plaque just as well as trained doctors, who have been trained to do so by painstaking analysis which takes a lot of time—like 25-30 minutes [per scan],” Dr. Damini Dey told dot.LA. As director of the quantitative image analysis lab at Cedars-Sinai’s Biomedical Imaging Research Institute, Dey witnessed firsthand the AI’s ability to quickly measure coronary plaque, allowing doctors to give patients a faster prognosis.
The device is already being used by Cedars-Sinai and other medical centers, but will require more testing and research before it gets into the hands of clinicians.
Autoplaque is among a slew of AI-enabled technologies that helps doctors find and diagnose complications in the body. Earlier this month, L.A.-based startup Pearl announced it had received FDA clearance for a device that uses AI to identify dental cavities, plaque and other oral conditions.- Moving Analytics Is Betting Telehealth Is Here to Stay - dot.LA ›
- Scientists Making Brain Computers Worry about Ethics, Safety - dot.LA ›
- Regard Medtech Startup Raises $15M for AI-Powered Software - dot.LA ›
- Regard Medtech Startup Raises $15M for AI-Powered Software - dot.LA ›
Keerthi Vedantam is a bioscience reporter at dot.LA. She cut her teeth covering everything from cloud computing to 5G in San Francisco and Seattle. Before she covered tech, Keerthi reported on tribal lands and congressional policy in Washington, D.C. Connect with her on Twitter, Clubhouse (@keerthivedantam) or Signal at 408-470-0776.
Image Source: Waymo
Image Source: Snap
Image Source: Dodgers 
Image Source: ChatGPT