AI smartphones may help monitor chronic wounds from home

Individuals increasingly can check on their health conditions at home. Devices that measure blood pressure, glucose and blood oxygen levels, for example, enable patients’ providers to make decisions for their care.

In the near future, patients may also be able to monitor their chronic wounds from the comfort of their homes through mobile artificial intelligence (AI) technology being developed at UB.

Wenyao Xu, professor and associate chair of the Department of Computer Science and Engineering, is leading a team that recently was awarded $3 million from the National Institutes of Health for the four-year research project, “Mobile Health Technologies for Assessing Blood Perfusion in Chronic Wounds.”

Approximately 8.2 million Americans suffer from chronic wounds that stem from a multitude of conditions, including diabetes and major surgery. This technology — accessible through a smartphone — will help streamline treatment, Xu says.

“Patients with these conditions frequently visit their health care providers or specialty wound care clinics and, in and some cases, need to be hospitalized,” Xu explains. “This can be a financial burden and a time burden to both the patient and the health care system.”

While the new technology will benefit all patients, it will particularly help individuals who have barriers to transportation or who live in rural areas where wound care services are limited or unavailable, he says.

“The approach proposed in this project is innovative,” he says, “because it will explore and validate new mobile computing and data-driven techniques for wound care and blood perfusion assessment based on evidence-based truth.”

The goal is to empower patients by giving them free, usable technology they can download that will provide exact measurements on how the wound is healing — and keep them out of the hospital.

“Chronic wounds often result from poor blood flow to the affected area, which can lead to a lack of oxygen and nutrients necessary for the healing process,” Xu says. “The long-term objective of this project is to improve wound care and impact wound treatment services by developing low-cost technologies that aid with chronic wound treatment and care.”

Other UB researchers on the project are Jun Xia, professor in the Department of Biomedical Engineering; Linda M. Harris, professor of surgery and chief of UB’s Vascular Surgery Division; Saptarshi Chakraborty, assistant professor in the Department of Biostatistics; Praveen Arany, associate professor in the Department of Oral Biology; and Heamchand Subryan, director of interaction design at the Center for Inclusive Design and Environmental Access (IDEA Center). They are also working with a research team at NEC Laboratories America, based in Princeton, N.J.

“We are engaging with an industry partner early on to pave the way for mass production of this technology,” Xu notes. 

The research began about five years ago, stemming from a conversation Xu had with Xia on how to advance AI and mobile optics together — mobile AI — to improve health care, Xu explains.

“We then connected with Dr. Harris, a professor of surgery at UB who has a specialty in vascular surgery,” Xu says. “This is an underexplored area for advanced mobile AI technology.”

Currently, they are collaborating with vascular clinics in the UBMD Physicians’ Group, where the team recruits patients in the technology evaluation and clinical study.

“The AI technology analyzes the mobile images and predicts if the wound is on the right track for healing,” Xu says. “If it’s not, the patients can make an appointment to see the doctor.”

The work with the clinics is the first in a two-stage process and will focus on fine-tuning the prototype and determining what needs to be improved over the next two years. It will involve more than 200 participants in Western New York.

“We will build upon this work to enhance a commodity smartphone with a low-cost multispectral imaging sensor. An app will be available to download on users’ phones to obtain perfusion-related measures,” Xu says. “We are developing machine learning algorithms that extract the biomarkers of healing and rigorously assess the tools for usability.” 

In early 2026, the team will implement the second stage: a participatory design, feedback and usability study that will include patients and human-computer interaction experts to assess and improve the technology’s usability.

The hope is that eventually anyone who is suffering from a chronic wound can access this technology, Xu says.

“This AI technology has the potential to advance the current standard of wound care for recovery after discharge from clinical therapies,” he says. “These contributions are significant because they can extend the health self-management of our society through proactive health care and real-time intervention and reduce subjective bias and financial burden for patients, their caregivers and society.”