Significantly improve reporting efficiency for Pneumothorax patients

Pneumothorax is a collapsed lung, which occurs when air leaks into the space between a person’s lung and chest wall. Pneumothorax may rapidly progress to respiratory insufficiency, cardiovascular collapse, and death if, unrecognized and untreated timely. It is a critical condition that requires communication and immediate action. In order to prevent significant morbidity or patient death, early detection is crucial.

Shin Kong Wu Ho-Su Memorial Hospital built an AI application to screen daily chest x-ray images for potential pneumothorax. The AI model was integrated into the hospital’s IT environment with Deeploy for automation workflow. Once the AI identifies the presence of possible pneumothorax, an SMS notification will automatically sent out to an on-site clinical personnel.

The pneumothorax AI model has demonstrated remarkable performance, with a testing accuracy of 98% and a clinical application accuracy of 99%.An analysis comparing the results from one year before and one year after the intervention of the AI model showed a significant improvement in reporting efficiency. 1. Overall reporting rate increased from 49.5% to 91.3% (1.8 times).

2. The detection rate within one hour increased from 29.7% to 85.3% (2.9 times). 3. The median detection time shortened from 7 hours to 12 minutes (35 times). All three key performance indicators were significantly optimized.

The project leader at Shin Kong Hospital emphasized, "As AI technology rapidly advances, close collaboration with information technology becomes increasingly crucial. This project leverages the no-code DeepQ platform, enabling the medical team to easily develop and deploy AI models, thereby reducing technical barriers and accelerating the application and promotion of AI technology in healthcare.