Why Some Women Respond to Treatment, but Others Don't
Research and Innovations Heart disease is different for everybody. The Peter Munk Cardiac Centre is using artificial intelligence to figure out why.
Imagine this scenario.
Two women, both 45-years-old and leading a similar lifestyle, arrive at the hospital severely fatigued and exhibiting shortness of breath. They are both found to have heart failure; a condition where your heart is not effectively pumping blood throughout your body.
A diagnosis of heart failure is severe for anyone, but treatment is designed to manage the symptoms. In this case, one woman responds well to the treatment and is able to be discharged from the hospital the next day, while the other woman continues to experience shortness of breath and has to be transferred to the intensive care unit. Despite a heightened level of care, the woman's kidneys begin to fail. Soon, other organs follow suit. She dies within a matter of days.
Why did the women respond so differently to treatment?
Today's medical landscape
The current standard in medicine is a "one treatment fits all" approach. "Right now, treatment plans include guidelines on how to manage the average patient. The plans are infrequently stratified into groups, let alone tailored to individuals," explains Dr. Heather Ross, Division Head of Cardiology at the Peter Munk Cardiac Centre, Director of the Ted Rogers Centre for Heart Research, and Loretta A. Rogers Chair in Heart Function. "This can be problematic for heart disease patients, particularly women," Dr. Ross continues, "since the guidelines are based on clinical trials where women are underrepresented."
When prescribing a treatment plan for heart disease, there is a wealth of data, imaging, laboratory testing, and medical history that should be taken into account to have a better chance of success. The problem is the sheer volume of data and time-sensitive nature of treatment which make it impossible to interpret the data in a meaningful way. Without assistance, the amount of data available now exceeds the capacity of the human mind. That's where artificial intelligence comes in.
Tomorrow's medical future
The team at the Peter Munk Cardiac Centre is learning more about the intersection of artificial intelligence and cardiac care. Artificial intelligence takes a tremendous amount of patient data and uncovers patterns the human brain cannot see.
"We currently ignore some data because it's so complex," notes Dr. Cedric Manlhiot, Director of the Cardiovascular Data Management Centre at the Ted Rogers Centre for Heart Research. "For example, right now, after an exercise test, clinicians focus on a handful of data points out of the thousands that are generated by the test. We have created an artificial intelligence algorithm that interprets the data that has traditionally been left out. For the first time, clinicians are now provided with unique insights allowing them to select the most appropriate therapy for each individual patient," he explains.
The future of artificial intelligence does not stop at developing a more responsive treatment plan, however. "By making sense of the personalized data that is available, we will be able to offer patients the right treatment at the right time, before heart disease occurs," says Dr. Carolina Alba, a Cardiologist and Scientist at the Peter Munk Cardiac Centre. By developing specialized algorithms, cardiologists can move towards preventative monitoring as opposed to reactive treatment.
Thanks to advancements in artificial intelligence, a new scenario is possible. In it, two women visit their doctor's office for a routine check-up and their physicians calmly explain that based on their age, sex, ethnic background, genetic composition, tests, and medical history, they are both at risk of heart disease. Undeterred, the physician proceeds to develop a personalized treatment plan for each woman that successfully prevents the disease from developing in their lifetime.