Standardizing the Crowdsourcing of Healthcare Data Using Modular Ontologies

Abstract

Crowdsourcing data is an essential part of information collection in healthcare. Patient data serves as the foundation for creating healthcare policy, creating new pharmaceuticals, and determining treatment. In this paper, we propose a novel conceptual method of standardizing and classifying the crowdsourcing of healthcare data using modular ontologies, authoritative medical ontologies (AMOs) and other sources. A modular ontology can be constructed to guide data collection for specific aspects of an illness. We will examine this conceptual approach for patients of depression. This will be done by finding association rules in a pre-existing National Institutes of Mental Health (NIMH)'s study on Sequenced Treatment Alternatives to Relieve Depression (STAR*D) patient dataset, and standardized medical terminology found in the Medical Dictionary for Regulatory Activities Terminology (MedDRA) ontology. We will also use classification knowledge from Diagnostic and Statistical Manual of Mental Disorders (DSM-5). Our conceptual method will ensure that newly crowdsourced data can be used to confirm and improve upon the accuracy of previously known symptom associations.