Abstract 2379: A Comprehensive Imaging Informatics System with Decision-Support Tools for the Interdisciplinary Comprehensive Arm Rehabilitation Trial
BACKGROUND: Stroke is one of the major causes of death and disability in America. The Interdisciplinary Comprehensive Arm Rehabilitation Evaluation (ICARE) project aims to compare a theoetically-defensible, evidence-based arm therapy, Accelerated Skill Acquisition Program, with traditional therapies. Imaging of the brain provides evidence of the location and severity of the stroke lesion. Researchers can analyze selected images acquired prior to enrollment and perform knowledge discovery of correlation between the size and location of the lesion and specific rehabilitation outcomes. However, large-scale trials such as ICARE encounter challenges for integrating imaging data from multiple healthcare institutions across the country. The ICARE project has 7 clinical sites distributed across the country, with more than 160 patients randomized thus far, tens of gigabytes of imaging data are involved. Therefore, we propose to develop a web-based system with tools that can support imaging and informatics related data within such a randomized controlled trial and allow for tele-consultation and collaboration. EVALUATION: The system includes a database and file storage system, a Health Insurance Portability and Accountability Act (HIPAA)-compliant anonymizer, a web-based Graphical User Interface (GUI) and decision-support tools such as lesion quantifying tools and image analysis. The GUI allows users to upload, search and review the images, patient information and annotations. The HIPAA-compliant anonymizer will anonymize any private information according to HIPAA. The lesion quantifying tool will help clinicians to measure, quantify, and characterize the size of lesion and evaluate the rehabilitation progress. The system will be used within 7 clinical sites and evaluated by researchers in the ICARE trial. DISCUSSION: The system allows for tele-consultation and is facilitated through the development of a thin-client image viewing application. Clinicians and researchers across the country can manage the imaging data, make and store annotations, measure the lesion size, and collaborate to identify the imaging biomarkers that would ultimately enhance the existing ICARE clinical trial database. Moreover, the anonymizer tool can be used not only for imaging data in this rehabilitation trial, but also for other data from multi-site clinical trials since most studies are required to be anonymized. CONCLUSION: A comprehensive imaging informatics system for a large-scale controlled stroke rehabilitation trial is developed. Physicians and clinical researchers are able to collect, organize and analyze stroke cases efficiently and effectively across multiple sites.