Advancing liver transplant medicine through artificial intelligence

Growing up with a father involved in mechanical engineering research, Dr. Mamatha Bhat sensed that discovery would shape her own future. Today, she channels that early inspiration into her work as a hepatologist and clinician-scientist at University Health Network (UHN), while mentoring and training the next generation of researchers.

During her residency in internal medicine and hepatology at McGill University, Dr. Bhat was struck by the profound impact of liver transplantation on patients’ lives. Drawn to both the science and complexity of liver physiology, and inspired by exceptional mentors, she began to work toward a career in hepatology research.

“Having a career where my clinical experience would inspire and inform the questions that I investigate really appealed to me,” she says. “Sometimes, there are also discussions I have with patients that will then trigger a research idea.”

Dr. Bhat was later recruited to join UHN’s world-class Multi Organ Transplant Program and the University of Toronto (U of T) as a clinician-scientist.

“I saw U of T as having highly interdisciplinary opportunities and resources to support research, and as an internationally recognized platform, it attracted me to pursue my research there,” she says.

Following her clinical training, she pursued additional education in medical biophysics and wet lab research, equipping herself to collaborate across disciplines and tackle complex questions.

“The questions that arise in my mind as a clinician inspire the discussions I have with people in other disciplines who can’t fully understand what the clinical environment is like or what patients really care about,” she says.

Her interest in artificial intelligence (AI) began in 2017 while completing her PhD in medical biophysics and was sparked by publications in the Journal of the American Medical Association (JAMA) on skin cancer and diabetic retinopathy recognition.

“I thought it would be interesting to see whether patterns in liver patient data could similarly be recognized because I felt like I was doing that intuitively in the clinic, trying to bring together different pieces of data and personalize the care of patients,” she says.

As co-lead of the Transplant AI Initiative at the Ajmera Transplant Centre, Dr. Bhat now oversees the development and deployment of AI tools designed to address pressing clinical questions in transplantation. She is also leading work to optimize transplant candidacy assessment, leveraging AI to prioritize patients accurately and evaluate their condition at the time of transplant and post-transplant.

“I think there are a lot of opportunities to model the complexity of our transplant patients’ care to optimize their outcomes,” Dr. Bhat says.

In her role as Partnership and Engagement Lead at the Temerty Centre for Artificial Intelligence Research and Education in Medicine (T-CAIREM), Dr. Bhat fosters connections across the healthcare AI community – locally, nationally and internationally. Through this work, she contributed to a position paper published in The Lancet Digital Health, developed in partnership with the Alliance of Centers of Artificial Intelligence in Medicine (ACAIM) and focused on the responsible deployment of multimodal AI tools in healthcare. T-CAIREM is also spearheading memorandums of understanding with institutions in several countries to advance global collaboration.

Dr. Bhat is currently trialing a multi-agentic AI transplant selection committee to support liver transplant candidacy decisions. The tool integrates multiple AI agents, each representing a different specialty, to collaboratively evaluate patients. Trained on more than 8,000 cases, it identifies appropriate candidates, flags absolute contraindications and provides a rationale.

“Our goal is to assess how well the decision of the multi-agentic committee aligns with the human committee,” she says.

Following a pilot at UHN where the model achieved 92% accuracy using 8,400 U.S. cases, an international trial was launched. The tool is now deployed in clinical settings at the Mayo Clinic, the University of California, San Francisco (UCSF), the National University Health System (NUHS) in Singapore, and transplant institutes in India, Saudi Arabia and Australia.

Since December 2021, Dr. Bhat has also served as Director of the Clinician-Scientist Training Program (CSTP) within the Department of Medicine, guiding aspiring clinician-scientists toward research careers.

“Through this program, successful trainees are able to receive support as Eliot Phillipson scholars, which is critical to supporting their research training,” says Dr. Bhat.

The program offers networking events and workshops to help trainees understand how clinical research is conducted across the department and how to position themselves for faculty roles. Dr. Bhat also mentors trainees directly, connecting them with scientists across U of T and its affiliated hospitals to help them navigate the academic landscape.

Over the years, several CSTP trainees have gone on to join the department as faculty.

“There are a lot of internationally recognized people enthusiastic to build bridges across disciplines at U of T,” says Dr. Bhat. “It is important to learn new skillsets and reach out to people with expertise different from your own, so that you can create truly impactful knowledge that will benefit health outcomes. Being a clinician-scientist is a challenging and highly rewarding journey with potential for widespread impact on patients’ lives worldwide, and I would encourage trainees to explore the wonderful opportunities to create new knowledge across U of T!”