Tech

Jefferson Health uses AI to help diagnose patients

The technology will help physicians save time and reduce administrative burden.
article cover

Ada

· less than 3 min read

Navigate the healthcare industry

Healthcare Brew covers pharmaceutical developments, health startups, the latest tech, and how it impacts hospitals and providers to keep administrators and providers informed.

The future is now. Some patients are turning to artificial intelligence (AI) instead of physicians for a diagnosis.

Berlin-based digital health company Ada Health is bringing its AI-based symptom assessment technology to Jefferson Health, greater Philadelphia’s largest health system, which has 18 hospitals across Pennsylvania and New Jersey, according to an April 13 release.

The technology will help physicians save time and costs by determining if a patient’s condition is manageable at home or if they need to come into the clinic for care, “helping to reduce unnecessary interactions with the system,” Claire Novorol, co-founder and chief medical officer at Ada Health, told Healthcare Brew.

“Patients live increasingly digital lives, and so integrating digital care navigation platforms lowers the barrier to entry for care while improving operational efficiency,” Edmund Pribitkin, chief physician executive and EVP at Jefferson Health, said in a statement.

Jefferson Health patients will have access to Ada’s symptom assessment and care navigation platform in both English and Spanish through the health system’s website, patient portal, and apps. The technology navigates through 10,000 symptoms and risk factors, as well as “3,600 conditions linked to 31,000 ICD-10 codes,” per the release. If the platform determines that a patient needs to be seen by a physician, their results are integrated into their electronic health record so it’s ready for their physician to access.

The cost for adapting this technology depends on the size of the health system, the number of patients, and how deeply integrated the technology is into the system, according to Novorol.

Implementing new AI-supported diagnostic decision support tools can be “expensive and complex” as “clinical operations differ substantially among healthcare systems, medical specialties, patient populations, and geographic areas,” according to a 2022 US Government Accountability Office report.

When only provided with basic health and symptom-related medical history, Ada’s technology was able to correctly identify the diagnosis 54% of the time, compared to 32% for physicians, according to a 2022 study of 33 German physicians. The physicians and the AI technology were tasked with identifying the presence of inflammatory rheumatic disease and suggesting diagnoses for 20 real-world patient examples.

Navigate the healthcare industry

Healthcare Brew covers pharmaceutical developments, health startups, the latest tech, and how it impacts hospitals and providers to keep administrators and providers informed.