Shivanshu Mathur and Raghav Jain, B Tech students of Lovely Professional University, developed ‘Medicare’, software that predicts disease with which a patient might be afflicted with. The software utilizes machine learning algorithms on the reported symptoms and test reports of the patient to provide suggestions.
The software helps save the doctor time that goes into reading reports and suggesting a diagnosis. Additionally, it provides the doctors with a larger list of potential diseases that a patient may be suffering from, some of which the doctors might not have considered.
This innovation of theirs has fetched them a second prize at NEC India Hackathon 2019, bagging prize money worth Rs 1.5 lakhs.
Shivanshu and Raghav explain their innovation in the interview below.
Shivanshu Mathur: I am a 3rd-year undergraduate student pursuing B.Tech. in Computer Science and Engineering from Lovely Professional University. I belong to Lakhimpur-Kheri, Uttar Pradesh. I am performance-driven, strategic thinker and very passionate to find a solution to real-life problems.
Raghav Jain: I am a Computer Science Engineering student at Lovely Professional University. I belong to Ludhiana in Punjab. I have done my schooling from Ryan International School. My hobbies are to browse the internet and to play cricket.
2. Describe your software.
The software predicts the disease from the symptoms and test reports of a patient. Currently, it can predict 148 diseases from 404 symptoms including Liver Diseases, Cardiovascular Diseases, Diabetes and major chronic diseases using NEC’s SX-Aurora (TSUBASA) which is a vector Supercomputer. The software uses two-step verification of disease. First, the symptoms are analyzed for the prediction which is further verified from the test reports of a patient from our pre-trained models.
3. How accurate are the results given by the software?
The disease prediction is ~91% accurate from the symptoms and from test reports it varies from ~70% to ~93% depending upon the type of the disease. The software uses machine learning algorithms and has the ability to learn and improve from the test reports without being explicitly programmed. Also, its accuracy will be enhanced automatically in the upcoming time.
4. What kind of support have you received from your college? What financial expenses have you incurred in the development stage of the software?
The university has helped us by providing us the platform to enhance our technical skills. Our mentor Gurpreet Singh provided us continuous support and guidance in winning the Hackathon.
All the tools that we have used are open source and therefore there are no financial expenses in the development stage of the software.
5. How do you plan to deploy the software?
The software has been deployed on NEC’s SX-Aurora (TSUBASA) which is a vector Supercomputer. NEC is expected to consider the idea for commercialization.
6. What are your future career plans?
Shivanshu Mathur: I want to be a Data Scientist in a product-based company and will do a Masters in Technology in Data Science. Apart from this, I have a keen interest in Cloud Computing as well. I have a clear goal of finding solutions to real-life problems and would like to continuously work towards bringing a significant change in the world with my contributions.
Raghav Jain: I would like to first gain experience with a job in a multinational company in my relevant field of study i.e. Big Data. I am very keen to pursue Cloud Computing and learn both technologies. I look forward to taking up higher education in the same later on. I see myself as a future cloud and big data architect.
7. Is there anything else that you would like to talk about your innovation?
We are training our software to predict more diseases and the data collected can be used for drug discovery and genetic testing in future.