follow us on facebook follow us on twitter email us

Patient Matching Algorithm Challenge

About the Challenge
Integrating patient information for better medical decision making

Posted By: Department of Health and Human Services
Category: Software/Apps
Skill: Algorithms Interest: Health Submission Dates: 12 a.m. ET, May 02, 2017 - 2 p.m. ET, Sep 12, 2017

HHS Names Patient Matching Algorithm Challenge Winners

Thousands of submissions received from more than 140 teams

The U.S. Department of Health and Human Services’ Office of the National Coordinator for Health Information Technology (ONC) today announced the winners of the Patient Matching Algorithm Challenge.

ONC selected the winning submissions from over 140 competing teams and almost 7,000 submissions using an ONC-provided dataset.  “Patient matching” in health IT describes the techniques used to identify and match the data about patients held by one healthcare provider with the data about the same patients held either within the same system or by another system (or many other systems). The inability to successfully match patients to any and all of their data records can impede interoperability, resulting in patient safety risks and decreased provider efficiency.

“Many experts across the healthcare system have long identified the ability to match patients efficiently, accurately, and to scale as a critical interoperability need for the nation’s growing health IT infrastructure.  This challenge was an important step towards better understanding the current landscape,” said Don Rucker, M.D., national coordinator for health information technology.

Winners include:

Best “F-score” (a measure of accuracy that factors in both precision and recall):

  • First Place ($25,000): Vynca
  • Second Place ($20,000): PICSURE
  • Third Place ($15,000): Information Softworks

Best First Run ($5,000): Information Softworks

Best Recall ($5,000): PICSURE

Best Precision ($5,000): Ocuvera

Each winner employed widely different methods.   PICSURE used an algorithm based on the Fellegi-Sunter (1969) method for probabilistic record matching and performed a significant amount of manual review. Vynca used a stacked model that combined the predictions of eight different models. They reported that they manually reviewed less than .001 percent of the records. Although Information Softworks also used a Fellegi-Sunter-based enterprise master patient index (EMPI) system with some additional tuning, they also reported extremely limited manual review.

The dataset and scoring platform used in the challenge will remain available for students, researchers, or anyone else interested in additional analysis and algorithm development, and can be accessed via the Patient Matching Algorithm Challenge exit disclaimer icon website.

Prizes
First Place Winner $25,000.00
Second Place Winner $20,000.00
Third Place Winner $15,000.00
Best in category: Precision $5,000.00
Best in Category: Best recall $5,000.00
Best in Category: First F-Score run $5,000.00
Solutions
No solutions have been posted for this challenge yet.
Rules
Submit Solution
Submissions for this competition are being accepted on a third-party site. Please visit the external site for instructions on submitting: https://www.patientmatchingchallenge.com/
Challenge Followers
Public Profile: 1
Private Profile: 14