TY  - JOUR
AU  - Okiyama, Sho
AU  - Fukuda, Memori
AU  - Sode, Masashi
AU  - Takahashi, Wataru
AU  - Ikeda, Masahiro
AU  - Kato, Hiroaki
AU  - Tsugawa, Yusuke
AU  - Iwagami, Masao
PY  - 2022
DA  - 2022/12/23
TI  - Examining the Use of an Artificial Intelligence Model to Diagnose Influenza: Development and Validation Study
JO  - J Med Internet Res
SP  - e38751
VL  - 24
IS  - 12
KW  - influenza
KW  - physical examination
KW  - pharynx
KW  - deep learning
KW  - diagnostic prediction
AB  - Background: The global burden of influenza is substantial. It is a major disease that causes annual epidemics and occasionally, pandemics. Given that influenza primarily infects the upper respiratory system, it may be possible to diagnose influenza infection by applying deep learning to pharyngeal images. Objective: We aimed to develop a deep learning model to diagnose influenza infection using pharyngeal images and clinical information. Methods: We recruited patients who visited clinics and hospitals because of influenza-like symptoms. In the training stage, we developed a diagnostic prediction artificial intelligence (AI) model based on deep learning to predict polymerase chain reaction (PCR)–confirmed influenza from pharyngeal images and clinical information. In the validation stage, we assessed the diagnostic performance of the AI model. In additional analysis, we compared the diagnostic performance of the AI model with that of 3 physicians and interpreted the AI model using importance heat maps. Results: We enrolled a total of 7831 patients at 64 hospitals between November 1, 2019, and January 21, 2020, in the training stage and 659 patients (including 196 patients with PCR-confirmed influenza) at 11 hospitals between January 25, 2020, and March 13, 2020, in the validation stage. The area under the receiver operating characteristic curve for the AI model was 0.90 (95% CI 0.87-0.93), and its sensitivity and specificity were 76% (70%-82%) and 88% (85%-91%), respectively, outperforming 3 physicians. In the importance heat maps, the AI model often focused on follicles on the posterior pharyngeal wall. Conclusions: We developed the first AI model that can accurately diagnose influenza from pharyngeal images, which has the potential to help physicians to make a timely diagnosis. 
SN  - 1438-8871
UR  - https://www.jmir.org/2022/12/e38751
UR  - https://doi.org/10.2196/38751
UR  - http://www.ncbi.nlm.nih.gov/pubmed/36374004
DO  - 10.2196/38751
ID  - info:doi/10.2196/38751
ER  -