of The National University Health System recently showcased the capabilities of its new AI platform that can predict hospital bed availability up to two weeks in advance.
The company’s ENDEAVOR AI platform is a next-generation EMR (NGEMR) system for computing multiple AI insights.
It hosts several AI tools, one of which can predict the estimated length of stay for each patient admitted to three public hospitals under NUHS. The AI tool does this by reading patient history and doctor notes in real time, running up to 30 times per hour. It can also provide clinical insight into factors that contribute to long-term patient stays.
According to the press release, the AI’s accuracy had been validated using NGEMR data for the past six months. It is said to be able to predict hospital bed conditions up to two weeks in advance to optimize bed capacity and patient placement.
why it matters
The predictive tool is NUHS’ solution to rising bed occupancy and increased bed wait times in EDs. This allows doctors to anticipate problems and intervene early. For example, a patient who has been in the hospital for more than her two weeks can be flagged so the medical team can change management or make plans for early transfer of the patient to a regional hospital for rehabilitation. can do.
With the ability to read notes, vital signs, and other test reports, AI tools can also predict the risk of deterioration in hospitalized patients.
NUHS plans to further develop AI’s ability to recommend care plans.” [can] Alter the trajectory of a patient’s disease course. “
Meanwhile, ENDEAVOR AI also has the ability to automatically alert managers of rising ED wait times, enabling early activation of human resources. Depending on how your resources are deployed, you can reduce latency from 30 minutes to several hours.
the bigger trend
NUHS also revealed plans to evaluate and later deploy an imaging AI model to enhance radiographic assessment of scoliosis. About 7,000 x-rays are taken each year in the scoliosis screening program in Singapore. Physicians still measure the curvature of the spine manually, which is time consuming and error prone. Also, it may take some time for the results to be communicated.
A new AI model has been designed to automatically measure the degree of scoliosis, improving the physician’s ability to interpret the scans. Early trials show that AI models can reduce reporting time with significant accuracy, improve clinician productivity, and help communicate results and expert referrals and treatment early. I was.
on record
“We are leveraging AI to improve healthcare practices and outcomes, empowering clinicians to make faster, more accurate diagnoses and more precise treatments.Today’s healthcare organizations aggregate vast amounts of data. However, most of this data has been analyzed retrospectively.With the technology of ENDEAVOR AI, we have been working with NUHS Associate Professor and CTO Ngiam Kee Yuan.
