Telehealth Data-derived Visual Analytics for Health Informatics Applications in Coordinated Care of Patients with Multiple Comorbidities

We describe telehealth data-derived visual analytics (VA) approaches aiming to deliver better healthcare outcomes to our telehealth service users with multiple comorbidities through enhanced real-time and real-world clinical decisions. Our telehealth center provides telehealth services and integrates telehealth datasets with electronic medical record (EMR)-derived information. In addition to the continuous vital sign data acquired from years of telemonitoring, the telehealth datasets also contain large amounts of unstructured service data including intervention notes and lifestyle-related information. These data are analyzed by our natural language processing (NLP) program for extracting coded data and subsequent machine learning. We develop a Gantt chart-based approach to interactively visualizing these complex analytics results in order to perform insightful examinations of users’ health status. The interactive Gantt chart presentations provide clear and quick comprehension of users’ disease status and progression, medication management, and treatment outcomes. Our VA approach potentially can optimize clinical decision making, facilitate efficient care coordination, ensure care continuum, and provide cognitive supports for reducing workload burdens. To our knowledge, this is the first case of using Gantt chart model for analyzing telehealth datasets and the integrated health information of both providers- and users-generated sources.