Dandelion Health Launches Pilot to Evaluate Healthcare Algorithm Bias & Equity

Dandelion Health, Inc., a startup specialising in healthcare data platforms, artificial intelligence, and precision analytics, has announced the introduction of a new public service aimed at rigorously evaluating the performance, fairness, and equity of health algorithms.

The pilot program will specifically target algorithms that predict cardiology outcomes using electrocardiogram waveforms, examining their performance and bias across various racial, ethnic, and geographic subgroups. Dandelion has developed this program with a patient privacy-first approach, making it accessible to algorithm developers at no cost.

Leveraging Dandelion's extensive collection of de-identified clinical data spanning from 2016 to the present, the pilot program aims to accelerate the development of responsible AI solutions to significantly enhance health outcomes while upholding patient privacy.

Validating algorithm performance against the desired populations has long been hindered by the challenge of securing high-quality clinical datasets encompassing diverse groups while prioritising patient privacy. Dandelion Health's pilot program is the first to eliminate this barrier. Through its consortium comprising three major healthcare systems—Sharp HealthCare (San Diego, California), Sanford Health (Sioux Falls, South Dakota), and Texas Health Resources (Arlington, Texas)—Dandelion is aggregating and de-identifying clinical data from approximately 10 million patients across the United States, representing various geographies, ethnicities, races, care settings, and more. The resulting raw de-identified clinical data, combined with Dandelion's profound expertise in creating top-notch clinical datasets, uniquely positions the company to validate algorithms using data that has not been de-identified or available at this scale before.

As AI tools continue to grow, develop, and integrate into healthcare, prominent industry experts, including the U.S. Food and Drug Administration and the Coalition for Health AI, have emphasised the need for rigorous methods to measure and validate the safety and effectiveness of AI algorithms. Dandelion's pilot program offers the first widely accessible solution to address this need.

The names of participating researchers and companies will remain confidential, and results will only be shared with the algorithm developers unless they grant permission otherwise.

Starting from July 15, 2023, Dandelion will accept electrocardiogram waveform algorithms designed to be representative of the U.S. population for validation from developers worldwide. The initial validation period will span three months. Following the program's launch, Dandelion may expand the validation initiative to include additional clinical data modalities, such as clinical notes and radiology imaging.

Click here to read the original news story.