OIC Fall 2022 Info Session Recap

Sep 9, 2022

On September 8th at 5 pm PST we hosted an information session with RTI International and the All of Us Research Program stakeholders to provide context and answer any questions about our Fall Open Innovation Challenges.

This hackathon is intended to be a challenge-based learning exercise where teams will compete to develop a proof of concept that best solves the challenges each stakeholder have presented below.

If you have any questions, would like to mentor, or become a sponsor please contact [email protected]

TLDR

* Register by September 14th 11:59:59 PT.
* Kick-Off: September 16-18th in Los Angeles (at least 1 team member needs to attend).
* Pick a challenge: 1. RTI Rarity AI Challenge OR 2. the All of Us Research Program Dashboard Design Challenge.
* You have 5 weeks to build a proof of concept and demo on October 20th.
* ​Winning teams will receive cash prizes and publicity while helping improve the human condition.

Register: https://airtable.com/shrFnT1ZrhnxlAOco

Kickoff Weekend Schedule

* Friday, September 16th: 6pm-8pm Team building and stakeholder presentations.
* Saturday, September 17th: 10am-4pm Office hours with stakeholders and workshops on Design, EDA, and Machine Learning.
* Sunday, September 18th: 10am-4pm Office hours to get feedback from stakeholders.

PRIZES

Each challenge will have three awards with the following prizes.

First Place

* $5,000 USD 
* Featured on the AI LA Blog and RTI International and the All of Us Research Program publications
* Social network amplification from our partners 
* Office Hours with AI LA and BioscienceLA

Second Place

* $1,000 USD 
* Featured on the AI LA Blog and RTI International and the All of Us Research Program publications
* Social network amplification from our partners 
* Office Hours with AI LA and BioscienceLA 

Third Place

* $1,000 USD 
* Featured on the AI LA Blog and RTI International and the All of Us Research Program publications
* Social network amplification from our partners 
* Office Hours with AI LA and BioscienceLA

CHALLENGE #1: RTI RARITY

PREDICTING UNDERDIAGNOSIS AND UNDERUTILIZATION FOR CHRONIC DISEASES

BACKGROUND: RTI Rarity is an AI solution developed by RTI International, an independent, nonprofit research institute dedicated to improving the human condition. One of RTI’s areas of focus is helping  government and private-sector clients understand social risks in their populations by leveraging data, analytics and advanced scientific knowledge. The RTI Rarity approach to social risk adjustment has, to date, focused on predicting long-term outcomes like life expectancy and drug overdose mortality.

With this challenge, we are delighted to partner with AI LA to bring great minds to bear on shorter-term, pressing problems. With better neighborhood hot-spotting, payers, providers and policymakers can focus their attention and funding on the highest-risk areas, potentially resulting in better returns on investment in SDoH.

PROBLEM: Type 2 diabetes mellitus (T2DM) and asthma are enormous public health challenges that result in high levels of health-related spending and potentially preventable utilization. Many cases of T2DM and asthma are underdiagnosed for years, leading to higher healthcare costs and poorer outcomes. Additionally, T2DM and asthma are often under-treated, leading to suboptimal disease management and higher costs. The problems of underdiagnosis and underutilization are known to be exacerbated in communities of color. Both diabetes and asthma are also known to be sensitive to social determinants of health (SDoH) – the conditions in the places where we are born, grow, live, work, and age that influence our health and well-being.

OUTCOMES: This challenge aims to identify neighborhoods (defined as Census tracts) in California with a high risk of either underdiagnosed or poorly managed T2DM and/or asthma. The winning team will develop a highly accurate, AI-driven approach to predicting the outcome(s) of interest using RTI Rarity™ and other data.

DATASETS PROVIDED:

* RTI Rarity database of over 200 variables in 10 domains of social drivers of health, demographics, and more for California
* AHRQ SDoH database for California
* RTI SynthPop™
data for California, if needed
* Medical Expenditure Panel Survey (MEPS) data for the entire US

REQUIRED SKILLS: Data science; data management; predictive modeling. Social epidemiology expertise is useful but not required. Healthcare background helpful but not required.

CHALLENGE #2: THE ALL OF US RESEARCH PROGRAM

BACKGROUND: The All of Us Research Program is working to enroll one million or more participants across the U.S. to share data that can fuel new insights into human health. By engaging the participation of people and communities who have been left out of medical research in the past and making their data safely available to a diverse cohort of researchers, the program seeks to accelerate health and medical research to enable individualized prevention, treatment, and care.

PROBLEM: The All of Us Research Program is committed to returning value to our participants and to the research community. As part of this commitment, we would like to engage the research community in creating an interactive, public tool using All of Us data of interest to participants.

OUTCOMES: Develop an interactive, public tool that All of Us participants and the broader community can use to explore and understand issues of health equity related to COVID-19 or social determinants of health.

DATA: Must use data from the All of Us Research Program’s Registered Tier.

* Data from more than 372,000 PARTICIPANTS, nearly 80% of whom identify with groups historically underrepresented in medical research, including 45% who identify with a racial or ethnic minority group.
* The Researcher Workbench now includes records from nearly 20,000 PARTICIPANTS who had SARS-CoV-2, opening new opportunities to study COVID-19 disease prevention, progression, and recovery through clinical and participant-reported data.
* The COVID-19 Participant Experience (COPE) Survey, with responses from more than 100,000 PARTICIPANTS, provides insights into participants’ mental and physical health during the first year of the pandemic.
* More than 132,600 RESPONSES to the Minute Survey on COVID-19 Vaccines, offering insight into participants’ perspectives on vaccines, and information about their vaccination status and plans.
* More than 57,600 INITIAL RESPONSES to the program’s new social determinants of health survey are now available for analysis.
* You can explore various EHR and Survey data types (summary statistics available here) as well as data from wearables (Fitbit) available in the Registered Tier.

If you have any questions, would like to mentor, or become a sponsor please contact [email protected]

ABOUT THE AUTHOR
Todd Terrazas
AI LA, Executive Director

As an innovator in the tech information and education space, Todd focuses on creating a robust ecosystem of resources and collaborative partnerships that can enhance LA's AI literacy, catalyze innovation, and advance workforce development opportunities.