Inspiration
Over 1.5 million people have been diagnosed with the coronavirus. In many countries, hospitals are out of capacities and doctors are overwhelmed with patients. There is still no reliable information about the true number of infected people. Many people with minor and mild symptoms cannot get access to medical help and health monitoring, some people with severe symptoms are dying because of delayed hospitalization.
What it does
TrackCovid is an automated symptom tracking tool for disease with long-standing or rapidly developing symptoms that require regular monitoring (including COVID-19). Patients complete an initial registration and receive regular automated calls from Voice Assistant which checks the health conditions of the patient and assesses the risks of complications and needs for immediate hospitalization using ML algorithms. If high risk for health is predicted by the system, the patient is connected with emergency medical services. As a result, patients do not feel left alone in an emergency, the medical system is less overwhelmed and acute cases are noticed in time. In addition, we collect reliable information about symptoms which is not subject to the discretion and punctuality of the patients.
How I built it
We use voice to text machine learning algorithms to recognize the answers of the patients and extract meaningful insight out of these answers.
Challenges I ran into
We leverage our experience in machine learning and telecom to create the solution which is the most needed.
What's next for TrackCovid
We are going to develop a functional prototype of a Voice Assistant for symptoms tracking.
Log in or sign up for Devpost to join the conversation.