In the health sector, one of the biggest roadblocks to effective solutions is the unavailability of past diagnosis and their respective remedies. For the most part, the health industry in India still uses conventional methods like paper files that have to be brought to the next visit voluntarily. This makes the margin of error huge and often this is the reason for the failure of the health industry. A few hospital chains do have sophisticated computerised systems but they are internalised which stops the widespread use of such systems. In cases such as these, a system that integrates all parties seamlessly is the need of the hour.
The main challenge was to find a way to handle two different categories of users: doctors and patients in the backend database.
This was solved by making a custom login interface that handled role-based authentication.
The other challenge was to deploy ML in the website to give accurate results.
This was solved by using a UK government database relating deaths and the PM 2.5 pollution in different locations and applying linear regression to it using the scikit-learn library in Python along with visualisation tools like Matplotlib.
The timeline UI to show medical history was a challenge in on the front-end side especially as it has to be dynamically updated according to the amount of data added previously.
This was solved by using a custom CSS package and tweaking it to suit our purposes.
Technologies used
Discussion