DICOM for Clinical Research: PACS-Integrated Electronic Data Capture in Multi-Center Trials

Abstract

Providing surrogate endpoints in clinical trials, medical imaging has become increasingly important in human-centered research. Nowadays, electronic data capture systems (EDCS) are used but binary image data is integrated insufficiently. There exists no structured way, neither to manage digital imaging and communications in medicine (DICOM) data in EDCS nor to interconnect EDCS with picture archiving and communication systems (PACS). Manual detours in the trial workflow yield errors, delays, and costs. In this paper, requirements for a DICOM-based system interconnection of EDCS and research PACS are analysed. Several workflow architectures are compared. Optimized for multicenter trials, we propose an entirely web-based solution integrating EDCS, PACS, and DICOM viewer, which has been implemented using the open source projects OpenClinica, DCM4CHEE, and Weasis, respectively. The EDCS forms the primary access point. EDCS to PACS interchange is integrated seamlessly on the data and the context levels. DICOM data is viewed directly from the electronic case report form (eCRF), while PACS-based management is hidden from the user. Data privacy is ensured by automatic de-identification and re-labelling with study identifiers. Our concept is evaluated on a variety of 13 DICOM modalities and transfer syntaxes. We have implemented the system in an ongoing investigator-initiated trial (IIT), where five centers have recruited 24 patients so far, performing decentralized computed tomography (CT) screening. Using our system, the chief radiologist is reading DICOM data directly from the eCRF. Errors and workflow processing time are reduced. Furthermore, an imaging database is built that may support future research.

A Survey of DICOM Viewer Software to Integrate Clinical Research and Medical Imaging

“…In general, we identified the open source tools MIPAV and Weasis as superior. As we have shown in our previous work, a good choice of components simplifies the build-up of a suitable system architecture [5]. Of course, new interfaces such as DICOM Store over the Web by Representations State Transfer (STOW-RS) [22] arise and have to be considered to find an optimal viewer candidate in the future.…”

“…Here, it is advantageous to integrate the DICOM viewer directly into the data-storing systems. Decentral viewing of image data within the eCRF optimally supports the workflow of the study personnel [5]. Since eCRFs are, today, usually offered via web, only web-or platform-independent viewers can be integrated.…”

“…The demand on workflow-optimized and integrated systems for imagingbased clinical trials increases [4]. For instance, powerful functionality for rendering of three-dimensional (3D) volume data such as computer tomography (CT) may be required within the electronic case report form (eCRF) to decide on inclusion or exclusion of subjects in clinical trials [5] or in rare disease registries [6]. Almost 370 free DICOM software projects are currently listed in the BI Do Imaging^database 1 , and finding an optimal viewer turns challenging, particularly for research applications.…”

“…Tight coupling of systems allows easy access to shared data. Integrating data, context, and functionality of components improve the workflow of medical personal, particularly in clinical trials [5]. Various approaches have been published, where DICOM viewer tools have been integrated into system architectures [11][12][13].…”

Design of an Electronic Upload and Reporting System aimed at Corelab Tasks and Responsibilities in Multi:center Clinical Trials

“…In the search for alternatives that better suit an ICL's workflow, we found some solutions that focused more on electronic data capture (EDC) in clinical trials [2] or on reuse of clinical data for research [3,4] rather than on integration with an ICL's MIS. In addition, analysis on dedicated echo workstations and QA-operations regarding the images are beyond the scope of these solutions.…”