Context
Cardiovascular diseases are the prime cause of death in the world – in the EU, it makes up for 48% of deaths, and in Romania 60%. eCuore asked QUALITANCE to help develop an affordable system that monitors patients outside hospital area and delivers alerts about their cardiovascular activities in real time to their doctors.
Challenge
The client needed a system that could provide complete diagnoses by analyzing thousands of electrocardiograms records. These contain sensitive data and can be easily misinterpreted. We built a component for doctors to make annotations on electrocardiograms to create an exhaustive training data store for the machine learning system.
Solution
We developed a user-friendly platform that facilitates data collection, storage, and sharing, and can display every minute of an electrocardiogram by analyzing tens of thousands of data.
Enriched data received from the doctor’s annotations is structured and normalized as input data for training of the learning system. This approach increases the detection for real-time diagnosis of cardiovascular disorders.
We built the prototype in 2 weeks and the core UI in 5 weeks.
Technologies
Python NLP, Numenta nupic, Angular, node.js.
