Congratulations to the 2024 Hackathon winners

The 5th edition of the Health Systems Innovation Hackathon took place in 8 cities around the world and 20+ winning teams were selected by each local Innovation Hub. The winning teams received two weeks of mentoring by members of the Harvard Health Systems Innovation Lab on how to refine their innovation and improve their pitch.


The final winners have been selected, and they will partake in a month-long Venture Incubation Program.

We want to congratulate the winning teams on their innovations:

  1. Winner: SNIFF

  2. 2nd:

  3. 3rd equal: SweetAudio and AI Pathology


Who are the Health Systems Innovation Hackathon winners of 2024? 


S.N.I.F.F. (Smart Neoplasm Identification by Fume Fingerprinting) is a non-invasive test that aims to revolutionize breast cancer surgery by decreasing the rate of incomplete surgical excision of positive margins. The test utilizes multiple parameters derived from gases emitted by the electrocautery device during tissue ablation. These parameters are used to discriminate between malignant and benign tissue in the resection zone using artificial intelligence algorithms. The test is a Point of Care Test that provides real-time intraoperative feedback to the surgeon about the margin status in the area being excised, enabling immediate modification of the surgical plan and wider excision if indicated.

The S.N.I.F.F. team consists of five medical students from the Pontificia Universidad Católica de Chile, two of whom are pursuing parallel studies in engineering and one of whom has a degree in chemistry. is a support tool for diagnosing and evaluating mental disorders through speech. It uses AI enhanced by Geometric Analysis to reach into the hyperspectrum of mental health data. The solution aims not to replace mental healthcare professionals but rather to equip them with assertive screening tests to improve diagnosis accuracy. Through speech recognition, uses geometric and physical analysis methods in addition to classical AI. This allows it to characterize speech patterns assertively even with little data collected. is the result of the combined efforts by team members affiliated with the Federal University of Pernambuco in Recife, Brazil with expertise in Medicine, Mathematics, and Statistics/Informatics.



SweetAudio is a groundbreaking, non-invasive method for glucose monitoring utilizing voice analysis. Leveraging the power of AI and the correlation between vocal biomarkers and blood glucose levels, the platform offers a personalized, accessible, and affordable solution that simplifies diabetes management worldwide.

SweetAudio team members include professionals and students at Harvard and MIT with expertise in biomedical engineering, biotech, AI/ML, medicine, public health and epidemiology.


AI Pathology

AI Pathology Tech harnesses innovative AI technology to address the growing incidence of skin cancer. Nevo, their flagship web application, facilitates early skin lesion screening using smartphone photography. Tailored to accommodate diverse ethnic backgrounds, it focuses on the broad range of skin tones found in Brazil. The application employs a deep learning model to analyze skin lesion images, thus streamlining the screening process and fast-tracking consultations with dermatologists.

The AI Pathology team includes pathologists, developers, an AI specialist, and a lawyer, all dedicated to tackling this significant health issue in Brazil.


Nour Sharara and Caroline Bulstra