The problem Vedanta solves

In the medical domain, committing errors can cost someone's life. With the wide array of diseases and symptoms, the probability of a wrong diagnosis by a doctor is huge. Even after a right diagnosis, cases have been reported where the prescription was wrong, or the patient recieved the wrong medicine due to clerical error. Another problem in healthcare is maintaince and transfer of health records. The patient has an extensive medical history which has to be taken itno account for a diagnosis to be completely accurate.

Hence, solving the few problems mentioned above, we present VEDANTA.

Features

Vedanta app
1: Disease Recommendation System
This is an ML trained model which recommends the diagnosis on the basis of the symptoms entered by the doctor. It acts as an assistant to the doctor in making a daignosis.
2: Medicine Recommendation System
As per the disease entered by the doctor, it recommends the best medicine that's used to cure the said disease. This again is to provide assistance to the doctor.
3: Virtual Referrals
This is a special feature which allows one doctor to refer a patient to another doctor and transfer all of the patient's medical records along with it. It ensures smooth flow of records and reduces unecessary paperwork.
4: Digital prescription
Around 70% of prescription errors are due to the sloppy handwriting of doctors, to reduce this percentage, we have integrated a feature by which a doctor can generate a digitally signed prescription which can be mailed to the user too. This is a step towards digitalization of healthcare.\

Vedanta web-console
A platform to manage appointment scheduling at the receptionist end to notify doctors as well as patients. This console will automatically mail the patients as well as the doctors reminding them of the upcoming appointments.

Challenges we ran into

We had problem implementing the firebase authentication part and we also faced problems while training our ML model for recommnedation system.
We tried writing the code snippets in javascript and kept debugging it and we used doctors ID for authentication which simplified the process.
And for ML we used datasets kaggle which solved the issues.