RetnoScan: AI-Based Retinopathy Diagnosis Won the D2K Sponsored Data Science Challenge

Second-Place Winner of the HackRice X 2020

HackRice Data Science Track Winner

HackRice X 2020 took place virtually from September 18 to 20 with nearly 350 participants and 83 teams submitting projects from all over the world. The data science challenge, sponsored by the D2K Lab, provided participants various challenges to work on real-world data science challenges, brainstorm innovative solutions, and transform data into knowledge and actionable insights.

Four high school students from Plano East Sr High: Shrey Joshi, Ishaan Javali, Madhu Sharma, and Kanishk Garg submitted their project "RetnoScan: AI-Based Retinopathy Diagnosis", which won the D2K challenge and the second-place prize in the overall competition.

"HackRice X was a really great experience that encouraged us to innovate a solution to a worldwide problem in just 36 hours. We definitely pushed ourselves to create RetnoScan and learned a lot about machine learning and app development in the process. We’re ecstatic to have won the D2K challenge because this is our first virtual hackathon and we weren’t expecting to win. Thank you for sponsoring the challenge and for giving us the amazing opportunity to present our project!" said Ishaan Javali.

RetnoScan is a mobile application that uses deep learning to diagnose eye images as being indicative of diabetes retinopathy. Students leverage machine learning concepts and advanced regularization methods to process transfer learning and build Android apps. "We used machine learning to train a model to detect healthy and Retinopathic eye images and packaged it into an android app which can be used by anyone anywhere. After taking a picture of the eye, the model predicts if your eye is retinopathic or healthy. You can also share the results with your doctor if needed."

HackRice X 2020 covered various topics ranging from biomedical, finance, to climate change and data science. Students practiced problem-solving skills while working on social impact projects. Besides the data science track, three of the D2K Industry Affiliate members including, Chevron and Energy Transfer have sponsored challenges in the HackRice X. The winners for each challenge are:

  • Challenge Winner: "Barbell2Go" by Albert Zhong, Hitesh Boinpally, Aditya Balasubramanian and Eric Marnadi.
  • Chevron Challenge Winner: " Work Order Tracking System for Managers and Employees" by Agna Parikh, Gedeon Pil, Shayar Shah and Mark Pepperl.
  • Energy Transfer Challenge Winner: " Schedule Transfer" by Andrew Feng, Christina Zhou, Colin King and Camsy Huang.

Read more about the HackRice X and winners from other challenges.

Related Articles

Team’s computer vision system tracks waterfowl, counts them from the air

Rice D2K Lab students build a predictive maintenance model to improve operations with machine learning.