Engineers, Clinicians Partner to Advance Human Health with AI
Engineers are collaborating with clinicians to develop AI-based solutions in health at the University of California, Davis. These technologies aim to provide a reliable method for early detection of Alzheimer’s disease, expand care and support options for survivors of sexual assault and improve outcomes for patients expecting corneal transplants.
The projects are made possible by AI and Health Seed Funding Awards from the Center for Information Technology Research in the Interest of Society and the Banatao Institute, or CITRIS. The awards are focused on strengthening interdisciplinary connections between schools and colleges across UC Davis and UC Davis Health and establishing the university as a leader in AI-enabled healthcare technology research.
“An engineer can build a model that scores well on a benchmark and still fails in a clinic, because we don’t know which errors matter to a patient,” said Sam King, a professor of computer science and director of CITRIS at UC Davis. “Clinicians know, but they can’t build these systems alone. The technology is improving faster than those partnerships are forming, and these AI and Health Seed Funding Awards exist to close that gap by putting engineers and clinicians on the same problem before the design decisions are made.”
Biomarkers for Early Alzheimer’s Detection
Alzheimer’s disease, the leading cause of dementia, is a progressive condition ultimately leading to one’s inability to care for oneself. The earlier the disease is detected, the better the chance the individual has of prolonging their level of mental function, and the longer they and their families have to prepare for the terminal condition.
However, the early signs of Alzheimer’s are subtle and difficult to detect with current tools, leading to many diagnoses coming late into the disease’s progression. To address this issue, Chen-Nee Chuah, Child Family Professor in Engineering in the Department of Electrical and Computer Engineering, is developing an AI model that can pick up early signs of Alzheimer’s disease from a patient’s speech.
Chuah is partnering with Michelle Cohn, a postdoctoral scholar in linguistics, and Alyssa Weakley, an assistant professor of neurology, to examine conversational interviews with older adults from the UC Davis Alzheimer’s Disease Research Center Corpus. The center’s conversational data is annotated with the patient’s cognitive status, such as mild cognitive impairment or late-stage Alzheimer’s. The team will use the conversations to train an AI model to identify biomarkers of cognitive impairment by analyzing how a patient sounds and their manner of speech.
So far, the team has completed the necessary datasets they need to train the large language model on acoustic and linguistic features of cognitive impairment in older adults. The research team has also received a boost from the National Institutes of Health to develop an AI-enabled remote monitoring system to detect health issues in older adults, which will directly support their efforts to develop an early detection tool for Alzheimer’s disease. The project will culminate in a real-world pilot deployment within a local senior living community.
A Chatbot for Facilitating Care
Medical care for survivors of sexual assault often relies on in-person visits. Yet symptoms of trauma and stigma around sexual assault can make it difficult for survivors to feel comfortable receiving or continuing medical care — not to mention the often-complicated referral systems or a potential lack of local resources that make it challenging to begin care in the first place.
Josh McCoy, an associate professor of computer science, is leading a project to develop a trauma-informed, AI-powered chatbot capable of connecting survivors with the appropriate social and health support services. The private chatbot aims to both expand access to and simplify the process of seeking care for survivors.
McCoy’s co-lead is Jessica Draughon Moret, an associate professor at the UC Davis Betty Irene Moore School of Nursing. Together, they are working to ensure the chatbot is informed by human-centered design and built on expert consultation with survivors.
The team has a prototype for the platform, called “CARE-COMPASS: Community Assistance and Resource Engagement,” and is currently in the refinement and user feedback phase of the project. The goal is to seek additional funding from the National Institutes of Health for further development and broader use.
AI-Assisted Tool for Donor Assessments
The cornea is the eye’s first line of defense. The dome-shaped barrier prevents dust, germs and other particles from reaching deep into the eye while letting light through. Damage to the cornea can lead to impaired vision and is among the top five causes of blindness worldwide.
For patients with damage to their cornea, corneal transplantation is one of the most successful interventions for restoring sight. But the success of the surgery relies on high-quality donor tissue, which can be tough and time-consuming for clinicians to evaluate.
Hamed Pirsiavash, a professor of computer science, is co-leading a project with Mark Mannis, a distinguished professor of ophthalmology and vision science, to develop an AI-powered evaluation system for donated corneal tissue. They are joined by Min Zhao, a professor of ophthalmology and vision science, Ana Sandoval Castellanos, a postdoctoral fellow in the Laboratory for Cell Nanoengineering, and several medical students.
The team has been working with Sierra Donor Services, a non-profit donor procurement organization serving the Northern California and Northern Nevada regions, to source a curated dataset of microscopy images of corneal tissue for the AI system.
Through self-supervised learning on the dataset, the system uses computer vision and machine learning to distinguish between healthy and abnormal corneal tissue. The goal is to reduce workload for clinicians from hours to minutes when evaluating tissue and minimize variability in the quality of donated tissue. The project’s next step is to test the AI model with larger sample sizes.
The Value of Partnerships
The AI and Health Seed Funding Awards dovetail with the UC Davis AI Center in Engineering’s AI and Health research area, which accelerates discovery by facilitating conversations between engineers and clinicians.
The center, known as AICE for short, was influential in establishing the seed funding awards, co-hosting a matchmaking mixer for engineers and clinicians in November 2025. CITRIS, the Healthy Aging in a Digital World initiative and the Clinical and Translational Science Center were partners for the mixer.
The collaborative event resulted in six awarded seed grants, including the three co-led by engineers. Through AICE’s partnership, the AI and Health Seed Funding awards are helping realize CITRIS Health’s mission to encourage innovations in technology that improve health outcomes and increase access to medicine.