Global healthcare AI platform Ubie and the American Kidney Fund (AKF) will collaborate to fine-tune Ubie’s AI-enabled symptom checker for kidney disease detection with the aim of accelerating a patient’s time to treatment.
Ubie’s AI is trained on clinical research papers, and then its physician review panel analyzes the prediction pathways and symptoms associated with a disease.
At this point, the AI can predict 1,100 diseases in the ICD 11, the World Health Organization’s International Classification of Diseases.
“We have a unique clinical usage of our symptom checker. We are deployed in 1,700 clinics and hospitals in Japan, and the physician usage of this product goes back to training our AI, which then is used for our patient-facing symptom checker,” Sanjeev Menon, head of partnerships for Ubie, told MobiHealthNews.
Through the partnership, AKF will provide 25 to 30 patients who know the cause of their kidney disease to assist Ubie in refining its symptom checker specifically pertaining to diagnosing kidney disease.
“These are patients that we know have ‘traditional kidney disease’ (so diabetes or high blood pressure, which are the two main causes) or some rare kidney disease, but they know the cause of it and that’s important because, with Ubie’s algorithm and the way they do it, patients ask questions,” Mike Spigler, VP of patient support and education at AKF, told MobiHealthNews.
Menon said Ubie will have the already-diagnosed patients run through its symptom checker to obtain aggregated data.
The company will then analyze the sessions each patient goes through and the answers they receive to determine which responses match the patient’s diagnosis. Ubie will then nail down where gaps in the AI exist to refine its algorithm for precision.
“You want to make sure it’s accurate and that the questions being asked are driving people to the right place, but it’s also the way in which the AI presents the questions and how people respond,” Spigler said.
Menon said the user element in AI is often overlooked, and Ubie is looking to fix that.
“What we’re trying to do with our symptom checker is get the patient voice and incorporate the patient voice into it to make sure we understand how users use the platform–phrase things in a way that is most commonly phrased by the user, not a clinician, and, therefore, it will lead to better output,” Menon said.
“Better output in AI, in our case, means better care guidance and better care decisions made by the patient.”
After patients understand their potential for kidney disease through Ubie, the digital health company says it will guide them to the American Kidney Fund to obtain more information, accelerating their care journey.
“The journey to diagnosis, especially for some of the less common diseases, can take years,” Menon said. “If we can help people have more informed conversations and find the right doctors off the bat, then we’re going to shorten that time to the correct diagnosis.”
The HIMSS AI in Healthcare Forum is scheduled to take place September 5-6 in Boston. Learn more and register.