Sibling researchers Dr. Vishnu Vardhan and Harsh Vardhini were inspired by their mother’s struggle with diabetes and postpartum hemorrhaging to start their own women’s health research center.
The research center, Vijayalakshmi Innovation Center in Women’s Health Analytics and Research, is the result of a partnership between the Pitt School of Medicine and Texas start-up Vizzhy Inc.
The center aims to merge AI with patient data to better predict women’s health by creating a “digital health twin” for patients. This AI simulation would show the effects of certain variables such as heavy drinking on the long-term health of the virtual patient.
Associate Professor Vanathi Gopalakrishnan has been working with the School of Medicine to shift its curriculum from treatment to prevention. Gopalakrishnan said she sees a communication gap between doctors and patients. If doctors have a stronger understanding of the factors and predictability of illness, they can inform their patients better, according to Gopalakrishnan. This gap is where AI jumps in.
“A doctor is supposed to prevent problems,” Gopalakrishnan said. “The question is, does the doctor actually have a medical education to teach other people how to prevent problems?”.
“AI Agents” — algorithms that stay up to date on information and work autonomously on a specific task, as opposed to traditional AI which needs to be updated with information manually — have gained popularity in recent years. A prime example is ChatGPT, which updates the information it repeats based on interactions with humans.
Gopalakrishnan said she hopes to give these tools and data to medical professionals to better predict and prevent chronic illness, especially in women who are also underrepresented minorities.
“How do you keep [women] mentally and physically well?” Gopalakrishnan said. “We have all this data about diseases and about health. What if we could create a digital representation of all of this?”
By creating accurate “female digital health wins,” healthcare can have a better grasp of what impact these major life events have on females. Gopalakrishnan said patients will become more informed about their health with these case simulations. These simulations will predict the outcomes of various treatments for women’s health issues.
According to a study by Pew Research Center, “Six in ten Americans say they would feel uncomfortable with their health care provider relying on AI.” Gopalakrishnan said she wants to include humans in the AI analysis so as to relieve patient doubts about the accuracy of AI.
“I always put a human in the loop. So machines and humans working together, they can create efficient systems that are fool-proof so that you do not give away private information,” Gopalakrishnan said.
She cites the work of Dr. Gregory Cooper as laying foundational research for VIHAR. Dr. Cooper is the Vice Chair of Research for the Department of Biomedical Informatics.
“My task now would be how to train the next generation of people,” Gopalakrishnan said. “As youth and as people who are going to be the next generation of the planet, I am looking at VIHAR as a help to [students]. We will have a bunch of like-minded people who have a very vast experience.”
Gopalakrishnan asked for the input of students for the direction for VIHAR. Submissions can be sent to vanathi@pitt.edu.