ONC announces winners of Patient Matching Algorithm Challenge

Ability to match patients is critical to health IT interoperability, says National Coordinator for HIT Don Rucker.


The Office of the National Coordinator for Health IT has announced the winners of its Patient Matching Algorithm Challenge, a competition to help solve the problem of misidentifying patients which impedes interoperability, provider efficiency and puts patient safety at risk.

“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, MD, national coordinator for health information technology.

ONC awarded six cash prizes totaling $75,000. The agency selected the winners, which used widely different methods, from more than 140 competing teams and nearly 7,000 submissions.

Also See: CHIME announces finalists for $1M National Patient ID Challenge

Participants in the challenge were provided with a dataset by ONC and were scored against a master key. The major prize category included three prizes for the highest “F-Score”—a combination of best precision and recall. In addition, “best in category” prizes were awarded for “best precision” (least mismatched patients), “best recall” (least missed matches) and “best first F-Score run.”



Rucker-Don-CROP.jpgFirst place ($25,000) for Best “F-score” was awarded to Vynca, which used a stacked model that combined the predictions of eight different models. According to ONC, they reported that they manually reviewed fewer than 0.01 percent of the records.

Second place ($20,000) for Best “F-score” went to PICSURE, used an algorithm based on the Fellegi-Sunter (1969) method for probabilistic record matching and performed a significant amount of manual review.

Third Place ($15,000) for Best “F-score” was awarded to Information Softworks. “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,” states ONC’s announcement.

In addition, Best First Run ($5,000) went to Information Softworks, while Best Recall ($5,000) went to PICSURE and Best Precision ($5,000) was awarded to Ocuvera.

The goal of the challenge was to spur the development of innovative new algorithms, benchmark current performance, and help industry coalesce around common metrics for success. For those interested in additional analysis and algorithm development, ONC is making available the dataset and scoring platform used in the Patient Matching Algorithm Challenge. They can be accessed here.

More for you

Loading data for hdm_tax_topic #better-outcomes...