Asthma Exacerbation Prediction and Risk Factor Analysis Based on a Time-Sensitive, Attentive Neural Network: Retrospective Cohort Study

• DOI: 10.2196/16981
• Publication Date: 2020-07-31T14:00:47Z
• Authors: Yang Xiang; Hangyu Ji; Yujia Zhou; Fang Li; Jingcheng Du; Laila Rasmy; Stephen Wu; W Jim Zheng; Hua Xu; Degui Zhi; Yaoyun Zhang; Cui Tao
• Keywords: Asthma Exacerbations

Background Asthma exacerbation is an acute or subacute episode of progressive worsening asthma symptoms and can have a significant impact on patients’ quality life. However, efficient methods that help identify personalized risk factors make early predictions are lacking. Objective This study aims to use advanced deep learning models better predict the exacerbations explore potential involved in asthma. Methods We proposed novel time-sensitive, attentive neural network using clinical variables from large electronic health records. The were collected Cerner Health Facts database between 1992 2015, including 31,433 adult patients with Interpretations both patient cohort levels investigated based model parameters. Results obtained area under curve value 0.7003 through five-fold cross-validation, which outperformed baseline methods. results also demonstrated addition elapsed time embeddings considerably improved prediction performance. Further analysis observed diverse distributions contributing across as well some possible cohort-level factors, could be found supporting evidence peer-reviewed literature such respiratory diseases esophageal reflux. Conclusions performed than previous for exacerbation. believe scores analyses clinicians assess individual’s level disease progression afford opportunity adjust treatment, prevent exacerbation, improve outcomes.