China Medical University Hospital’s AI systems saved the lives of 30% of patients


April 25, 2024

One of the platforms is a triage system that can identify heart attacks in nearly 40 seconds.

With over 190,000 outpatient visits and 7,600 inpatient admissions monthly, China Medical University Hospital (CMUH) has developed three AI systems to speed up medical response for patients and reduce staff workload.

Dr. Der-Yang Cho, CMUH superintendent, first introduced the IAMS or intelligent anti-microbial systems that can detect bacteria in just one hour.

It is being used by the hospital more than 10,000 times every month. The AI system notifies clinicians of drug resistance from 24 hours to 72 hours before the sepsis develops.

The second AI system that Dr. Cho introduced was the AI-assisted STEMI Triage from the ambulance to the emergency room (ER). The system is designed to analyse data collected and sent back from an ambulance. After this, it can revert results in about 37 seconds.

It not only cuts time for examinations upon the patient’s arrival but also fast-tracks the entire ER procedure, saving up to 12 minutes for each patient.

Meanwhile, Dr. Cho has also set up the gHi system, which is a clinical documentation tool that uses speech recognition as well as generative AI.

It is known as the first bilingual tool, which uses English and Mandarin, creates transcripts of patient visits, and narratives for nursing, and compiles reports for medical exams.

Overall, the platform can free up 75% of the time on administrative tasks.

“The platform covers the English spoken language and it can give us [documents on] medical assessments, the medical diagnosis and medical training,” Dr. Cho told Healthcare Asia in an interview at the recent Healthcare+ Expo in Taiwan.

More AI tools

Another platform that has helped the hospital is the iHi platform, which assists in managing the datasets from three million citizens spanning two decades. These records include 60,000 cancer registry entries.

The repository also integrates clinical and NHI records, medical images, genomic data, and living factors.

This data can create digital twins which will facilitate the correlations between genes and diseases, which focuses on identifying genes linked with Alzheimer’s and vascular dementia for personalised prevention and treatment.

At present, CMUH has three new solutions deployed in the hospital, with one cleared by the Food and Drug Administration and another tool expected to receive clearance by the end of the year.

CMUH also garnered the HIMSS’s Davies Award Excellence distinction, which honours healthcare institutions that provide thoughtful application of health information and technology to improve clinical care delivery, patient outcomes and population health worldwide.

– Hemanta Poudel, Associate Editor, Health Sansar Publication