A Method for Efficient De-identification of DICOM Metadata and Burned-in Pixel Text.

De-identification of DICOM images is an essential component of medical image research. While many established methods exist for the safe removal of protected health information (PHI) in DICOM metadata, approaches for the removal of PHI "burned-in" to image pixel data are typically manual, and automated high-throughput approaches are not well validated. Emerging optical character recognition (OCR) models can potentially detect and remove PHI-bearing text from medical images but are very time-consuming to run on the high volume of images found in typical research studies. We present a data processing method that performs metadata de-identification for all images combined with a targeted approach to only apply OCR to images with a high likelihood of burned-in text. The method was validated on a dataset of 415,182 images across ten modalities representative of the de-identification requests submitted at our institution over a 20-year span. Of the 12,578 images in this dataset with burned-in text of any kind, only 10 passed undetected with the method. OCR was only required for 6050 images (1.5% of the dataset).

Improving eSource Site Start-Up Practices.

Background: eSource software that copies patient electronic health record data into a clinical trial electronic case report form holds promise for increasing data quality while reducing data collection, monitoring and source document verification costs. Integrating eSource into multicenter clinical trial start-up procedures could facilitate the use of eSource technologies in clinical trials. Methods: We conducted a qualitative integrative analysis to identify eSource site start-up key steps, challenges that might occur in executing those steps, and potential solutions to those challenges. We then conducted a value analysis to determine the challenges and solutions with the greatest impacts for eSource implementation teams. Results: There were 16 workshop participants: 10 pharmaceutical sponsor, 3 academic site, and 1 eSource vendor representatives. Participants identified 36 Site Start-Up Key Steps, 11 Site Start-Up Challenges, and 14 Site Start-Up Solutions for eSource-enabled studies. Participants also identified 77 potential impacts of the Challenges upon the Site Start-Up Key Steps and 70 ways in which the Solutions might impact Site Start-Up Challenges. The most important Challenges were: (1) not being able to identify a site eSource champion and (2) not agreeing on an eSource approach. The most important Solutions were: (1) vendors accepting electronic data in the FHIR standard, (2) creating standard content for eSource-related legal documents, and (3) creating a common eSource site readiness checklist. Conclusions: Site start-up for eSource-enabled multi-center clinical trials is a complex socio-technical problem. This study's Start-Up Solutions provide a basic infrastructure for scalable eSource implementation.

Retooling institutional support infrastructure for clinical research.

Clinical research activities at academic medical centers are challenging to oversee. Without effective research administration, a continually evolving set of regulatory and institutional requirements can divert investigator and study team attention away from a focus on scientific gain, study conduct, and patient safety. However, even when the need for research administration is recognized, there can be struggles over what form it should take. Central research administration may be viewed negatively, with individual groups preferring to maintain autonomy over processes. Conversely, a proliferation of individualized approaches across an institution can create inefficiencies or invite risk. This article describes experiences establishing a unified research support office at the Duke University School of Medicine based on a framework of customer support. The Duke Office of Clinical Research was formed in 2012 with a vision that research administration at academic medical centers should help clinical investigators navigate the complex research environment and operationalize research ideas. The office provides an array of services that have received high satisfaction ratings. The authors describe the ongoing culture change necessary for success of the unified research support office. Lessons learned from implementation of the Duke Office of Clinical Research may serve as a model for other institutions undergoing a similar transition.

Research management team (RMT): a model for research support services at Duke University.

Collecting and managing data for clinical and translational research presents significant challenges for clinical and translational researchers, many of whom lack needed access to data management expertise, methods, and tools. At many institutions, funding constraints result in differential levels of research informatics support among investigators. In addition, the lack of widely shared models and ontologies for clinical research informatics and health information technology hampers the accurate assessment of investigators' needs and complicates the efficient allocation of crucial resources for research projects, ultimately affecting the quality and reliability of research. In this paper, we present a model for providing flexible, cost-efficient institutional support for clinical and translational research data management and informatics, the research management team, and describe our initial experiences with deploying this model at our institution.