My group focuses on improving the quality of care for cardiovascular patients, in particular, those with heart failure and cardiomyopathy (enlarged and thickened heart muscle). We’re investigating novel drug targets and better ways to select patients who will truly benefit most from cardiovascular therapies by analyzing big data from genetic sequencing, real world data (information collected continuously about a drug’s effectiveness within a patient population in their normal living environment) and machine learning.  

Even with the same genetic mutation, not everyone presents the same symptoms

Every patient responds slightly differently to a disease and treatment, calling for a shift in care towards precision medicine. My group focuses on why some patients are at high risk for end-stage heart disease (dilated cardiomyopathy) and why we see variability in disease severity, even within family members who carry the same mutation. Understandably, family members also want to know what their risk of heart disease is.
 
Using genome-wide association studies (GWAS), whole genome sequencing (WGS) and real world data, we’ve discovered new genes that contribute to end-stage heart disease and have observed combinations of affects, for example in females, chemotherapy may induce an increased risk of heart disease. In particular, we’re making disease maps of patients that include exposure effects (for example, environment, sports, medication, occupation) and genetic modifiers (genes or mutations that can alter the expression of another gene) in order to study how and why a disease presents differently in patients. We expect that these deep genetic interrogations, once thought of as futuristic, may one day become standard care, not only for patients, but also for healthy individuals. 
 

A learning healthcare system

Have you ever reserved a hotel on Booking.com? Or bought something from Amazon? You may or may not have noticed that these websites “learn” what your preferences are and push why don’t you try this or places you may be interested in as suggestions for your next purchase, making your search more accurate for your profile. Such a tool could greatly enhance treatment plans. A patient and physician could add to a patient’s profile during care, and based on the information, can accurately adjust the patient’s treatment, outlook, prognosis and lifestyle in a timely manner. 
 

Clinical trials in the clinic

It’s no secret that the timeline of 10-12 years for a clinical trial is long. During that time, technologies change, methods advance and machine learning re-defines how we examine patients. For many patients, even if a new drug or therapy reaches the market, it’s often too late or not beneficial. What if we re-define clinical trials and integrate them into daily practice? This immediately involves patients, allows physicians to randomize during care (using patient electronic medical records), and no patient is unfortunate enough to receive a placebo. Every patient is treated and adjusted immediately according to their disease progression during this more personalized style of a clinical trial.
 
We have the technology and computational potential to shape the future of healthcare. Digital tools are facilitating closer collaboration between patients and healthcare providers and generate an enormous amount of real-time health-related data, giving us a more complete overview of genetic and non-genetic factors influencing a person’s well-being.