A personalized approach to anti-arrhythmic therapies
Patients respond differently to anti-arrhythmic drugs depending on specific mutations in the Na+ channel gene. We use statistical models to predict these patient-specific responses to anti-arrhythmics. These models can then be used to provide more effective anti-arrhythmic therapies to patients.
The figure below demonstrates our model’s ability to predict UDB (A) and QTc shortening (B) by mexiletine from channel gating parameters.
The model was then tested on 5 LQT3 variants (A) not used in training the model. The partial least square regression model predicts QTc shortening by mexiletine (B).
Minimally-invasive transcatheter ablations are currently the primary therapeutic option for patients who suffer from chronic arrhythmias. A central difficulty for physicians performing these procedures is that they cannot see what they are doing in 3D. Our goal is to use recent advances in virtual and augmented reality to improve outcomes from these procedures.