Sheba Will Revolutionize Mental Health Care by Developing “Liv” Generative AI Platform

Sheba’s ARC Innovation Center, in collaboration with KPMG, has leveraged Microsoft Cloud and Microsoft Azure OpenAI Service towards innovating mental health care. Current research, led by Sheba, aims to develop a GenAI-based platform designed to assist mental health professionals in their workflow.

The research investigated how an interactive conversational platform that engages users in a natural language may alleviate the bottlenecks in the mental health care system, thus substantially shortening the time to treatment.

Sheba believes that by creating such a platform it will mark a significant advancement in the accessibility and efficiency of mental health care. The statistical results of the initial research were analyzed by the Gertner Institute for Epidemiology and Health Policy Research.

Motivated by the need to address the growing global mental health challenges exacerbated by COVID and other global crises, a research platform named “Liv” was developed. The platform captures intake sessions through both voice and written conversations, providing a personalized experience. The research platform uses DSM-5, a manual of mental disorders published by the American Psychiatric Association, and adheres to responsible AI principles, mental health protocols, professional guidelines, and recent literature in the field to guide the conversation.

The intial study findings demonstrated that the research platform can assist the practitioner with intake process for Post-Traumatic Stress Disorder (PTSD), depression, anxiety and more. The research platform generates a user summary that is sent to the practitioners to workflow, while always keeping a human in the loop.

Another significant finding from the study indicates that using the research platform, “Liv”, can shorten the cycle of providing mental health services to users, allowing practitioners to assist up to five times as many users when compared to current practices, helping more users secure much needed mental health services.

“The high global demand for mental health services necessitates solutions that can bridge the gap created by the lack of sufficient, relevant workforce resources,” said Prof. Eyal Zimlichman, Deputy Director-General and Chief Innovation & Transformation Officer at Sheba Medical Center. “We are exploring how Artificial intelligence, and specifically Generative AI, can assist clinicians in treating more users, promoting effective therapy and improving accessibility to high quality care. Through our collaboration with KPMG and leveraging Microsoft Azure OpenAI Service, we are investigating the potential that Gen AI can provide to what we believe will be a transformative change in healthcare. We anticipate that our research will result in a fully operational platform within a few months, once we have thoroughly examined all aspects of its use and established the necessary guard rails.”

“In an era where the human touch is irreplaceably valuable, the research serves as a beacon of hope,” said Iris Shtein, Director of Strategic Partnerships and the initiative’s lead at Sheba. “Our aim is to enhance the capabilities of our health professionals, ensuring that every user feels heard and cared for in a way that was never before possible.”

The initial study, led by Prof. Mark Weiser, Head of the Psychiatric Division at Sheba Medical Center and Dr. Asaf Caspi, Director of Sheba’s Psychiatry Department, in collaboration with M.S.R – The Israeli Center for Medical Simulation compared the research platform results to those of practicing psychiatrists who were provided with the same data and user information. The study found that the research platform results were highly accurate.

In addition, feedback from study participants highlighted the research platform’s effective engagement, marking a notable achievement in creating a positive experience for users. This contribution suggests the research platform’s role in enriching mental health interactions, demonstrating its potential to positively affect user experiences and user engagement.