La Hacks
From left to right: Judge Onike Williams, Medibuddy team members Choidorj Bayarkhuu, Arnav Roy, Stanley Sha, Emma Wu and judge Phil Martie. (Photo courtesy of LA Hacks)
LA Hacks, Southern California's largest hackathon, was hosted by UCLA on April 19-21. With 1,000 students from across the country participating, the winning team, MediBuddy was selected by judges Phil Martie and Onike Williams. Each team member received Play Station 5 consoles as their prize. The prize was sponsored by the Patient Safety Technology Challenge with funding from the Pittsburgh Regional Health Initiative.
MediBuddy leverages machine learning models, to analyze symptoms, body measurements, and more inputs by doctors to generate accurate, personalized suggestions on diagnoses and treatments based on previous case data.
Team members included: Choidorj Bayarkhuu, primary frontend developer studying computer science at UCLA; Arnav Roy, primary backend logic developer, machine learning studying computer science at UCLA; Stanley Sha, backend Flask developer and presentation contributor studying computer science at UCI; and Emma Wu, data processing, secondary front-end developer, and presentation lead statistics and data science at UCLA.
MediBuddy team member Choidorj Bayarkhuu shared how a misdiagnosis affected him in his childhood and inspired the project by making him, “realize how important patient safety is because if I didn't do a blood test, they would've removed my completely fine appendix.”
Team member Emma Wu shared more inspiration for MediBuddy, “My team saw the issue of misdiagnoses especially addressable using a data-based solution and machine learning because it is an issue that arises often due to a lack of data. Patients who have complicated conditions that are difficult to diagnose are often recommended to see numerous physicians or specialists in order to get multiple opinions on their condition. In essence, this is a form of data collection and analysis, since data from multiple doctors is being collected to come to a conclusion about a diagnosis. However, this is very time consuming and can be very expensive, and patients may waste time on trying to find multiple doctors to diagnose them that could have been spent on immediate treatment. We came up with the idea of using a machine learning model to eliminate the need for patient- and doctor- conducted data collection by giving MediBuddy to assist doctors with their diagnoses by providing insights from a large database of health records.”
Bayarkhuu continued with more about his team’s work, “MediBuddy offers doctors a user-friendly website to aid in diagnosing patients. Using a Machine Learning model, it predicts the likelihood of diabetes, kidney failure, or cancer based on key inputs provided by the doctor. Once the diagnosis is complete, it communicates with Google's Gemini API to generate a concise report for the patient, which can be easily incorporated into their daily routine. Doctors have the option to save the patient's report or download it as a unique PDF.”
MediBuddy team member Arnav Roy added, “Ultimately, the purpose of MediBuddy is not to replace the job of a doctor, but to complement it. As an efficient tool that enhances diagnostic accuracy, accelerates recovery by providing immediate treatment suggestions, and saving money that otherwise would have been spent on unnecessary testing (for diseases that are highly unlikely), MediBuddy is a powerful tool that can help save lives and improve the world of healthcare.”
LAHacks organizer Julie Lam touted that LA Hacks loved working with the Patient Safety Technology Challenge and providing hackers the opportunity to create innovative health solutions and address medical errors.
“We are strong advocates of PSTC's initiative and aim to spread awareness of such critical causes through ou hackathon,” Lam said.