Curating Retrospective Multimodal and Longitudinal Data for Community Cohorts at Risk for Lung Cancer

Abstract

AbstractLarge community cohorts are useful for lung cancer research, allowing for the development and validation of predictive models. A robust methodology for (1) identifying lung cancer and pulmonary nodules from electronic health record (EHRs) as well as (2) associating longitudinal data with these conditions is needed to optimally curate cohorts at scale from clinical data. Both objectives present the challenge of labeling noisy multimodal data while minimizing assumptions about the data structure specific to any institution. In this study, we leveraged (1) SNOMED concepts to develop ICD-based decision rules for building a cohort that captured lung cancer and pulmonary nodules and (2) clinical knowledge to define time windows for collecting longitudinal imaging and clinical concepts. We curated three cohorts with clinical concepts and repeated imaging for subjects with pulmonary nodules from our Vanderbilt University Medical Center. Our approach achieved an estimated sensitivity 0.930 (95% CI: [0.879, 0.969]), specificity of 0.996 (95% CI: [0.989, 1.00]), positive predictive value of 0.979 (95% CI: [0.959, 1.000]), and negative predictive value of 0.987 (95% CI: [0.976, 0.994]). for distinguishing lung cancer from subjects with SPNs. This work represents a strategy for high-throughput curation of multi-modal longitudinal cohorts at risk for lung cancer from routinely collected EHRs.