Innovative AI solutions that will change the Healthcare sector

In today’s world, healthcare is facing new and unprecedented challenges. At the same time, technology has never played such a significant role when it comes to assisting healthcare professionals and their patients. In partnership with Microsoft, an online hackathon was organised and hosted by Stratiteq, from April 21st to May 12th, 2021. This hackathon was about coming up with solutions, trying completely new and different approaches that can help the healthcare sector.

Main partners of this hackathon were NVIDIA, SAS and Women in AI. The challenges in the hackathon were presented by leading Healthcare Institutions across Europe: Foundation 29, HM Hospitales, Hospital San Raffaele, Sanitas and University Hospital Basel. The hackathon was supported by many companies, universities, and organizations: Live AI, Microsoft for Startups, NVIDIA Inception, Microsoft Reactor, Faculty of Organisation and Informatics Croatia, Hackathons Australia, and Global AI Student Conference.

Prior to the hackathon two panel discussions were organised on topics “How will AI change healthcare in the next years?” and “Open Data & Data Collaboratives to tackle the COVID-19 challenge”. The panels featured experts from hospitals and leading healthcare institutions across Europe, as well as Massachusetts Institute of Technology, Microsoft, NVIDIA, SAS, Women in AI, and moderated by representatives from Stratiteq.

Hackathon challenges were divided in three tracks:

  • AI applied to medical imaging: how AI can help extract information from medical images that represent a pathological tissue or an organ?
  • Predictive medicine: how it’s possible to obtain predictive models of evolution of the disease, epidemiological models, information on the response to the various treatments applied, knowledge of virus for the creation of a vaccine, and sociodemographic data on the impact on the population of the covid-19 virus, or predict the likelihood of a rare disease?
  • Empower patients: how to help patients that often struggle to understand their medical reports or other medical documents because they don’t have the necessary knowledge to understand the terminology?

Developers of all backgrounds and skills registered to participate in this hackathon, and the hackathon had over 300 developers registered from over 50 countries worldwide. The hackathon took place online over 3 weeks, and during the hackathon teams were supported by mentors from partners, hospitals, and organisations. Each week we made a follow-up with hackathon participants via live streams (April 21st, April 28th, May 5th, May 12th) with guests from our partners, showcasing possibilities with AI and providing other useful info to the hackathon participants.

The hackathon was organized via Devpost, a platform that helps software engineers to participate in software competitions. Participating teams, until the end of the hackathon, submitted 37 projects that were reviewed by the hackathon council and the jury made the final choice on the best solutions. The winning solutions were presented in the winner announcement stream.

The winner of the “AI applied to medical imaging” track was the “AI Wonder Girls” team that developed the “Medical Imaging Communication Tools” application that detects wrist fractures, automatically generates reports and displays an analytics dashboard from adaptable configuration capabilities.

The winner of the “Predictive medicine” track was the “AnalysisMode” team that created a model which can solve both predictive medicine and hospital resource allocation.

The winner of the “Empower patients” track was the “CaReD: Care for Rare Diseases” that worked on the patient-centred application with which patients and their families can track their daily symptoms, meet with others exhibiting similar symptoms with the same disease in real-time to provide each other with emotional support, while also empowering them with trustworthy and up-to-date medical information tailored to their specific symptoms.

Microsoft for Startups awarded the prize to the “Carotid Plaque Risk Assessment” team that worked on track “AI applied to medical imaging”. Given a carotid ultrasound sequence the software segments plaque volume in carotid and other arteries to extract volumetric parameters like volume, volumetric stenosis, 3D greyscale median and greyscale median distribution. The original ultrasound is visualized in a 3D view where it is clearly shown the carotid and the plaque.

NVIDIA awarded the prize of expedited access to NVIDIA Inception, an acceleration platform for AI, data science, and HPC startups, to the “AI Wonder Girls” team that also had the best solution in the “AI applied to medical imaging” track.

SAS awarded the prize to the “Butterfly Effect” team that worked on the track “Predictive medicine”. Their solution was using previous COVID-19 patient data to create a model that predicts whether a patient is likely to be a short or long term patient in the hospital based on readings of vital signs taken in the first few days of hospital. Their dashboard assesses patient data and can show the user the details about median results informing clinical decision making and resource allocation.

The hackathon solutions are visible in the hackathon project gallery. This hackathon was a great experience, with a lot of amazing solutions built that tried out completely new and different approaches which will for sure help the healthcare sector in the future.