People living in the roots of the country are many a times monolingual and the amount of ethnic langauges in India are so many. The medicine descriptions and names are generally in English which a lot of people do-not have an understanding of. Since they do-not have an understanding of English they are dependent on signs and symbols put up by the doctor on the medicine slips or on some other person to tell them what medicine they need to take at what time. Also if they by mistake mix the medicines, then it might become a life threatening issue. What people do have even at the root level is access to a smart phone, and that is what we plan to use. We plan to come up with a deep learning based application deployed on android based system that scans the name on the medicine, recognises it and tells its use in the native language so that people are less dependent on others for utilising medicines. This application can be really helpful for even the elderly living in metropolitans who in emergencies might be alone at home or at some place where their mobile might be the source of help and this application could turn out to be the life saver where a minute or two can make a difference between life and death.
We plan to create a dataset of all known medicines that are legalised in India and develop a deep learning based application that, using Convolutional Neural Networks recognises the characters on the medicine palette, will connect with an offline or online database to retrieve the information about the medicine in the native or chosen language of the person. We plan to provide support for English, Hindi and Oriya at the start. This can be done by employing a Kotlin based android app which uses a google API to logically corrected text in the chosen language.
For people with eye sight issues we will also employ a text to speech converter so that it becomes more accessible to the people. Also if it turns out that the medicine is not found then we would add it to a separate table in the dataset and there would be a separate work view to include those medicines. Both offline and online support would be issued and the information provided will vary.