In 2012, Karina Gasbarrino‘s grandfather passed away from an ischemic stroke. Since then, Gasbarrino, a graduate of McGill’s PhD program in experimental medicine, has dedicated her career to understanding and developing early stroke prevention methods. In 2019, she succeeded, launching the digital health startup PLAKK, a cutting-edge tool that helps predict ischemic strokes. Globally, ischemic strokes are the second leading cause of death and the third leading cause of disability, impacting millions of families worldwide.
An ischemic stroke is caused by blood clots in the arteries leading to the brain. This results from the rupture of an atherosclerotic plaque, an accumulation of fat and cholesterol that builds up in carotid arteries in the neck. For now, atherosclerotic plaques cannot be detected with a blood test and, instead, require an ultrasound. Since atherosclerosis is a progressive disease, it needs early screening to provide proper treatment, which does not start until a patient displays symptoms. The most common symptoms include dizziness, mobility issues, and numbness in the facial muscles.
In an interview with The McGill Tribune, Kashif Khan, McGill medical student and co-founder and CEO of PLAKK, explained that an inaccurate diagnosis is one of the many hurdles to proper stroke care.
“Inaccurate diagnoses [cause] many patients to be missed by current clinical standards, putting them at a larger risk for stroke,” Khan said. “The limitations cause approximately three million preventable strokes each year.”
In other words, current diagnostic procedures do not go into the depth required for accurate stroke diagnoses and often are only performed when the patient’s atherosclerotic plaque accumulates to dangerous levels.
As the chief operating officer of PLAKK, Gasbarrino and her team have worked tirelessly to create a tool that uses artificial intelligence (AI) to accurately identify high-risk atherosclerotic plaques. They have also developed a “stroke-risk score” which would help doctors proceed with the form of treatment that best suits the patient’s needs. The use of such high-tech image analysis, in combination with histology—the study of cell structure—helps with the early detection and prevention of strokes.
SonoPlaque, the revolutionary AI technology that Gasbarrino developed at McGill, overcomes the problems of current diagnostic methods for ischemic strokes by using a variety of clinical parameters, like the plaque composition and flow dynamics of the affected artery, to create an accurate stroke-risk assessment. A combination of blood biomarkers—molecules found in patients’ blood that correlate with high stroke risk—annotated medical images, clinical characteristics, and omics data are also used to create patient-specific algorithms.
The program also uses ultrasound technology to monitor levels of arterial blockage.
According to Khan, the road to developing SonoPlaque as a patient-ready tool was a collective effort that included contributions from a variety of industry professionals.
“Transitioning from academia to industry was a complicated, but exciting journey,” Khan said. “We sought out to build a multidisciplinary team of clinicians, surgeons, AI engineers, data scientists, and experts in business strategy.”
Supported by numerous donors and government grants, PLAKK is now being used in clinical settings in Quebec to start testing and validating its software. The company hopes to get Food and Drug Administration (FDA) and Health Canada approval next year.
As for the future, Khan foresees two important milestones for SonoPlaque: The first would be publishing the results of PLAKK’s clinical studies, and the second, to close a seed round of two million dollars for funding future works.
“Together, the success of these milestones will allow us to achieve regulatory approval in the U.S.,” Khan said. “FDA approval puts SonoPlaque on the map as a valuable stroke-detection technique, and funding helps to propel more essential research forward.”