Expert artificial intelligence (AI)-based natural language processing (NLP) algorithms may help clinicians and researchers systematically identify childhood asthma and its subgroups with distinctive characteristics from electronic health records (EHRs), according to the results of a cross-sectional analysis published in BMJ Open Respiratory Research.
The lack of effective, consistent, reproducible, and efficient asthma ascertainment methods can result in inconsistent cohorts and study results for asthma clinical trials or other studies. Thus, researchers assessed whether application of expert AI-based NLP algorithms for 2 existing asthma criteria to EHRs of a pediatric population could systematically identify childhood asthma and its subgroups with distinctive characteristics.
The researchers categorized patients into 4 groups (both criteria positive: NLP-Predetermined Asthma Criteria [PAC]+/NLP-Asthma Predictive Index [API]+); PAC positive only: NLP-PAC+ only; API positive only: NLP-API+ only; and both criteria negative: NLP-PAC−/NLP-API−) and characterized them. Of the 8196 patients, 1614 (20%) were NLP-PAC+/NLP-API+, 954 (12%) were NLP-PAC+ only, 105 (1%) were NLP-API+ only, and 5523 (67%) were NLP-PAC−/NLP-API−.
Children with asthma who were classified as NLP-PAC+/NLP-API+ showed earlier asthma onset, more of a Th2-high profile, poorer lung function, more frequent asthma exacerbations, and a higher risk for asthma-associated comorbidities compared with other groups. These results were consistent with those based on an unsupervised cluster analysis and laboratory and pulmonary function test data from a random sample of 300 children.
“In conclusion, an expert AI-based NLP algorithms for two existing asthma criteria systematically identified childhood asthma on a large scale and its subgroup with distinctive characteristics minimising methodological heterogeneity in defining asthma and maximising our abilities to detect true biological heterogeneity among asthmatic patients,” stated the study authors.
Seol HY, Rolfes MC, Chung W, et al. Expert artificial intelligence-based natural language processing characterises childhood asthma. BMJ Open Resp Res. 2020;7:e000524. doi:10.1136/bmjresp-2019-000524 http://dx.doi.org/10.1136/bmjresp-2019-000524