Eyeing on the enormous potential of big data and artificial intelligence, the Chimei Medical Center established the “Medical Big Data AI Computing Center” (AI Center) in 2019 to analyze various risk factors on the individual basis for precise medical predictions as well as to develop AI models for solving difficulties in clinical situations.
Also started in 2019, Chimei has become Taiwan’s first medical center to employ AI technologies in ER operations for the clinic professionals to have better judgments on patient status.
As risks could exist in medical practices, it’s imperative to reduce them to the minimum via precise, individualized predictions based on actual data. To fully utilize AI in such information and applications in a structural manner, the Chimei Medical Center founded the AI Center in 2019 for training and development of AI models.
“At Chimei, we are not doing AI just for more research papers,” said Dr. Chi-ching Wang, the major facilitator behind the new AI Center. “We want it to empower our clinical practices for the benefit of our colleagues and patients alike.”
How does AI Model change your workflow?
Turn problems into models with cross-department collaborations
“To initiate an AI project in the medical center requires three roles: clinical professionals, analysts and IT engineers,” explained Mr. Chung-feng Liu, Supervisor of the AI Center. The clinical professionals bring problems forward from the ones they encounter in the practices, and the AI Center would help them to turn problems into solutions.
The experts from the AI Center would visit every department to introduce how AI solutions were developed from data acquired, and encourage the colleagues to come up with possible applications.
Efficient resource utilization for quick, useful results
To the IT Department, waves after waves of difficulties are a part of the daily development works; and even the most outstanding staff members are still limited by the same 24 hours in a day.
They spent most of their time maintaining the online medical systems rather than studying unfamiliar tools such as Docker or Kubernetes environments, optimizing GPU resources or debugging models in response to error messages.
Moreover, it takes days to build a model, and even more time in communications between the AI Center and IT Department itself.
Fast Model building in an hour with PrimeHub
The aforementioned problems are where the PrimeHub comes in. It helps image creation, GPU environment configuration, parameter fine-tuning and version control to save time spent on long waiting and painstaking two-way communications.
Now the AI Center can build models in no time, while the IT Department easily manages the systems — all without the effort to fine-tune the environment or spend days on model deployments. All thanks to the PrimeHub’s blazing-fast ability to make completed models available online in merely an hour.
The PrimeHub helps us to perform system management tasks easily and efficiently, and it even allows us to deploy AI models online in an hour.
IT Department, Chimei Medical Center
90% prediction accuracy reported by international journals
AI Prediction System for ER Chest Pain Cases, the brainchild of collaboration of the AI Center and ER, has the amazingly high accuracy of 90% reported on SCI, a venerable international medical journal.
Commonly seen on ER patients, chest pain might be the result of cardiovascular diseases. To offer a quick, effective solution for high-accuracy diagnostics, the AI Center introduced Supervised Machine Learning techniques that analyzed papers and experiences from senior doctors to elect 14 characteristics for constructing AI prediction models. The model is capable of making medical predictions on an individual basis for better treatment quality.
Today, the ER Prediction Systems not only diagnose chest pains, but also influenza of seniors, Acute Pancreatitis, Hyperglycemia, Dengue Fever and Pneumonia; and the systems are being deployed in various departments such as Anesthesiology and ICU as well.
The AI Center needs to build and deploy models quickly, and the PrimeHub solved our urgency by eliminating most cost spent on communications. It’s priceless.