Geisinger was named runner-up in the Centers for Medicare & Medicaid Services (CMS) Artificial Intelligence Health Outcomes Challenge, out of more than 300 participants.
Geisinger collaborated with Medial EarlySign, a leader in machine learning-based solutions for the early detection and prevention of high-burden diseases, to use artificial intelligence (AI) to predict unplanned hospital admissions, readmissions soon after hospital discharge, healthcare-associated risks, and mortality. The two organizations partnered to create models that predict the risk of these outcomes using Medicare administrative claims data, as well as novel visualization to explain the results in a clinician-friendly way, which is an important component of AI implementation.
CMS recognized Geisinger and EarlySign for "consistently strong performance in all competition elements while delivering the highest prediction accuracy results" due to their shared vision of innovation and collective focus on patient-centered care. Their ability to successfully communicate predictions to clinicians, known as AI explainability, played an important role in their selection as runner-up.
In 2019, the CMS AI Health Outcomes Challenge was launched, with over 300 organizations offering AI solutions for predicting patient health outcomes. Submissions aimed to forecast a variety of outcomes, including unplanned admissions for heart attack, asthma, chronic obstructive pulmonary disease, and other high-risk conditions, as well as adverse effects such as hospital-acquired infections, sepsis, and respiratory failure.
In November 2020, Geisinger was selected as one of seven finalists. CMS used a systematic assessment process backed by a team of AI scientists to determine the winner and runner-up. The explainability of the models was tested and scored by clinicians from the American Academy of Family Physicians, a CMS partner in the AI Challenge. A jury of CMS senior leadership reviewed the submissions and chose the winners.
About Geisinger
Geisinger is dedicated to making better health more accessible to the more than one million people it serves. Abigail Geisinger founded the system more than 100 years ago, and it currently comprises nine hospital campuses, a health plan of over 500,000 members, a research institute, and the Geisinger Commonwealth School of Medicine. Geisinger, with almost 24,000 employees and over 1,600 employed physicians, adds billions of dollars to the Pennsylvania economy each year.
About Medial EarlySign
Medial EarlySign assists healthcare systems with keeping patients healthier for longer periods. EarlySign accelerates the successful implementation of AI and Machine Learning solutions in clinical workflows by including insights derived from real-world data, allowing for the early detection and prevention of high-burden complications and conditions. EarlySign works closely with clients to implement one or more of its pre-built, outcome-focused solutions (AlgoMarkers) for augmenting early detection of tumors, indicating chronic disease progression and deterioration, stratifying risks for unplanned admissions, and adverse events and complications from infectious diseases, with a focus on improved results and lower costs.