The human body is a complex entity: problems rarely occur without any early warning signs, and learning to detect them prematurely might be the difference between life and death for many. With the growing tide of wearables such as the Apple Watch and FitBit, monitoring patient lifestyle and health parameters outside of the clinic becomes an all too easy task. What if such devices could be used to detect and identify heart disease?
Cardya is an iOS application designed to collect and aggregate annotated heart data from its users, in an effort to develop models capable of detecting and identifying heart conditions such as atrial fibrillation and ventricular tachycardias. Its development can be divided into 2 distinct phases:
- Data Collection and Annotation: Heart rate data, previously diagnosed heart conditions and various health parameters (i.e. height, weight, age, etc.) are collected from the user and stored on our servers.
- Model Development and Testing: Data is cleaned, triaged and fed into a neural network in the hopes of developing a model capable of detecting various heart conditions with accuracies upward of 0.90.
Participation is voluntary. After downloading the app, youll be asked to give your consent for participation. The heart rate sensor data you provide over time will be collected for research analysis after being anonymized.