Hallym University Kangnam Sacred Heart Hospital

Hallym University Chuncheon Sacred Heart Hospital developed an AI model that predicts the prevalence risk of aspiration pneumonia caused by dysphagia for the first in Korea.

Aspiration pneumonia from dysphagia is a disease that occurs when the food, which is supposed to pass through the esophagus, enters the lung through airways and causes inflammation. It is one of the main causes of death for inpatients, and when it occurs to patients with poor immune function, it can lead to serious in-hospitals infections. In particular, it is fatal to intubated patients in the ICUs and elderly patients with weak immune systems.

Chuncheon hospital developed this AI model by extracting data of 6,500 patients who were hospitalized only by aspiration pneumonia in the recent 10 years and by applying them into machine learning algorithms. The AI model learned more than 20 types of data including level of consciousness, vomit, oxygenation, dehydration, and use of medicine drugs.

“In the past, we had to cope with the risk of aspiration pneumonia based only on clinical conditions such as repeated cerebral infarction, dementia, and mental deterioration, but now the AI model enables us to identify and cope with the risk of onset in real-time,” said Professor Jong Hee Sohn of Neurology.

Hallym University Medical Center plans to introduce this AI model to all hospitals and use it for inpatient care.

By Jonghun Im, Int’l Cooperation Team, HUMC (imjonghun@hallym.or.kr)