Identification of pediatric respiratory diseases using a fine-grained diagnosis system.

Autor: Yu G; Department of IT Center, The Children's Hospital, Zhejiang University School of Medicine, China; National Clinical Research Center for Child Health, China., Yu Z; Deepwise AI Lab, Beijing, China., Shi Y; Department of Computer Science, School of EE&CS, Peking University, Beijing, China., Wang Y; Department of Pulmonology, The Children's Hospital, Zhejiang University School of Medicine, China; National Clinical Research Center for Child Health, China., Liu X; Deepwise AI Lab, Beijing, China., Li Z; Department of IT Center, The Children's Hospital, Zhejiang University School of Medicine, China; National Clinical Research Center for Child Health, China., Zhao Y; Department of IT Center, The Children's Hospital, Zhejiang University School of Medicine, China; National Clinical Research Center for Child Health, China., Sun F; Deepwise AI Lab, Beijing, China., Yu Y; Department of Computer Science, The University of Hong Kong, Hong Kong. Electronic address: yizhouy@acm.org., Shu Q; National Clinical Research Center for Child Health, China. Electronic address: shuqiang@zju.edu.cn.
Jazyk: angličtina
Zdroj: Journal of biomedical informatics [J Biomed Inform] 2021 May; Vol. 117, pp. 103754. Date of Electronic Publication: 2021 Apr 06.
DOI: 10.1016/j.jbi.2021.103754
Abstrakt: Respiratory diseases, including asthma, bronchitis, pneumonia, and upper respiratory tract infection (RTI), are among the most common diseases in clinics. The similarities among the symptoms of these diseases precludes prompt diagnosis upon the patients' arrival. In pediatrics, the patients' limited ability in expressing their situation makes precise diagnosis even harder. This becomes worse in primary hospitals, where the lack of medical imaging devices and the doctors' limited experience further increase the difficulty of distinguishing among similar diseases. In this paper, a pediatric fine-grained diagnosis-assistant system is proposed to provide prompt and precise diagnosis using solely clinical notes upon admission, which would assist clinicians without changing the diagnostic process. The proposed system consists of two stages: a test result structuralization stage and a disease identification stage. The first stage structuralizes test results by extracting relevant numerical values from clinical notes, and the disease identification stage provides a diagnosis based on text-form clinical notes and the structured data obtained from the first stage. A novel deep learning algorithm was developed for the disease identification stage, where techniques including adaptive feature infusion and multi-modal attentive fusion were introduced to fuse structured and text data together. Clinical notes from over 12000 patients with respiratory diseases were used to train a deep learning model, and clinical notes from a non-overlapping set of about 1800 patients were used to evaluate the performance of the trained model. The average precisions (AP) for pneumonia, RTI, bronchitis and asthma are 0.878, 0.857, 0.714, and 0.825, respectively, achieving a mean AP (mAP) of 0.819. These results demonstrate that our proposed fine-grained diagnosis-assistant system provides precise identification of the diseases.
(Copyright © 2021 Elsevier Inc. All rights reserved.)
Databáze: MEDLINE
Externí odkaz: