Receive a weekly summary and discussion of the top papers of the week by leading researchers in the field.

In Journal of medical Internet research ; h5-index 88.0

BACKGROUND : As the coronavirus disease (COVID-19) epidemic worsens, the burden of quarantine stations (Q stations) outside of emergency rooms (ERs) at every hospital increases daily. To prepare for the screening workload inside Q stations, all staff with medical licenses are required to support the working shift. Therefore, the need to simplify the workflow and decision-making process for physicians and surgeons from all subspecialist fields is necessary.

OBJECTIVE : To demonstrate how the NCKUH AI trilogy of smart Q station diversion, AI-assisted image interpretation, and a built-in clinical decision-making algorithm improves medical care and reduces quarantine processing time.

METHODS : This observational study on the emerging COVID-19 pandemic included constitutively 643 patients. The artificial intelligence (AI) trilogy, i.e., 1) smart Q station diversion, 2) AI-assisted image interpretation, and 3) a built-in clinical decision-making algorithm on a tablet computer, was applied to shorten the quarantine survey and reduce processing time during the COVID-19 pandemic.

RESULTS : The use of the AI trilogy facilitated the processing of suspected cases, with or without symptoms, travel, occupation, and contact or clustering histories, which were performed with a tablet computer device. A separate AI-mode function that could quickly recognize pulmonary infiltrates on chest X-rays was merged into the smart clinical assisting system (SCAS), and this model was subsequently trained with COVID-19 pneumonia cases from the GitHub open source dataset. The detection rates were 93.2% and 45.5% in posteroanterior and anteroposterior chest X-rays, respectively. The SCAS algorithm was continuously adjusted based on the frequently updated Taiwan Center for Disease Control public safety guidelines for faster clinical decision making. Our ex vivo study demonstrated the efficiency of 75% alcohol disinfection on the tablet computer surface for a 20-μL positive SARS-CoV-2 virus solution. The positive rate of a real-time polymerase chain reaction was 100% and became 75% and 0% after one and two disinfection procedures (n=4), respectively. To further analyze the effect of the AI application in the Q station, we subdivided the Q station into with or without AI groups. Compared with the conventional ER track (n=281), the survey time at the clinical Q station (n=1520) was significantly shortened [median survey time (95% confidence interval; CI) at the ER: 153 (108.5-205) min vs. at the clinical Q station: 35 (24-56) min; p<0.0001]. Furthermore, the use of the AI application in the Q station reduced the survey time in the Q station [median survey time (95% CI) without AI: 100.5 (40.3-152.5) min vs. with AI in the Q station: 34 (24-53) min; p<0.0001].

CONCLUSIONS : The AI trilogy improves medical care workflow safely by shortening the quarantine survey and reducing processing time, especially during an emerging epidemic infectious disease.

CLINICALTRIAL :

Liu Ping-Yen, Tsai Yi-Shan, Chen Po-Lin, Tsai Huey-Pin, Hsu Ling-Wei, Wang Chi-Shiang, Lee Nan-Yao, Huang Mu-Shiang, Wu Yun-Chiao, Ko Wen-Chien, Yang Yi-Ching, Chiang Jung-Hsien, Shen Meng-Ru

2020-Sep-16