Gemini Enhances Care in Japanese Hospitals

Before joining Ubie, a health tech startup in Japan that leverages AI to revolutionize patient experiences and streamline healthcare operations, I worked as a neurologist. The sheer volume of clinical duties and the endless paperwork, often handwritten, left me feeling overwhelmed. I constantly worried about how this affected the quality of care and attention I could provide to my patients. This concern drove me to join Ubie.

At Ubie, our mission is to enable healthcare professionals to focus on what truly matters: spending quality time with their patients. We aim to make administrative tasks, such as inputting medical records and writing referrals, more efficient.

Currently, we're developing products that incorporate Google's Gemini models, which we fine-tune on Google Cloud's Vertex AI platform, for use in real-world clinical environments. Our collaboration with Google allows us to customize these sophisticated models to grasp the unique aspects of Japanese healthcare. This customization enables the AI to assist doctors and nurses with voice recognition and summarization, significantly easing the burden of creating numerous discharge summaries, referral letters, and informed consent documents.

Our efforts are already yielding results in medium- to large-sized hospitals in rural Japan:

• At Keiju General Hospital, our AI-powered discharge summary tools have cut the time nurses spend on these tasks by 42.5% and reduced their psychological burden by 27.2%. Notably, the impact was most pronounced for longer hospital stays, where the cognitive load on nurses is higher.
• At Yokokura Hospital in Kyushu, the implementation of voice transcription and summarization features, powered by Gemini, has improved the efficiency of documenting patient explanations by 33%.
• During a trial at Kyushu University Hospital, one of Japan's largest hospitals, the use of our technology for summarizing and standardizing referral letters resulted in a 54% efficiency increase for doctors preparing admission summaries.

These improvements allow healthcare professionals to dedicate more time to their patients, directly enhancing the quality of care—something that motivated me to become a doctor in the first place.

Efficiency in healthcare is crucial in Japan, where an aging population and a diminishing workforce pose significant challenges. Additionally, finding administrative staff, particularly in rural areas, is difficult, adding to the operational strain on hospitals. AI has the potential to be a game-changer, alleviating some of these pressures and fundamentally changing how we work and care for our patients.

[ttpp][yyxx]