JMIR Publications

ABSTRACT

Background:

Reliable and interpretable automatic extraction of clinical phenotypes from large electronic medical records databases remains a challenge, especially in a language other than English.

Objective:

We aimed to provide an automated end-to-end extraction of cohorts of similar patients from electronic health records for systemic diseases.

Methods:

Our multistep algorithm includes a named-entity recognition step, a multilabel classification using Medical Subject Headings ontology and the computation of patient similarity. A selection of cohorts of similar patients on a priori annotated phenotypes was performed. Six phenotypes were selected for their clinical significance: P1-osteoporosis, P2-nephritis in systemic erythematosus lupus, P3-interstitial lung disease in systemic sclerosis, P4-lung infection, P5-obstetric antiphospholipid syndrome, and P6-Takayasu stroke. We used a training set of 151 clinical notes and an independent validation set of 256 clinical notes, with annotated phenotypes, both extracted from the Assistance Publique-Hôpitaux de Paris data warehouse. We evaluated the precision of the 3 patients closest to the index patient for each phenotype with the precision-at-3, and the recall and average precision.

Results:

For P1-P4, the precision-at-3 ranged from 0.85 [0.75, 0.95] to 0.99 [0.98, 1], the recall ranged from 0.53[0.50, 0.55] to 0.83[0.81, 0.84], and the average precision ranged from 0.58 [0.54, 0.62] to 0.88 [0.85, 0.90], respectively. P5-P6 phenotypes could not be analysed due to a limited number of phenotypes.

Conclusions:

Using a method close to clinical reasoning, we built a scalable and interpretable end-to-end algorithm to extract cohorts of similar patients.