Info

Clinical Imaging

Objective: Maintain an online archive, de-identification and upload clients and viewers for clinical images to support the storage and exchange of images between researchers in multicenter studies and enable re-use of imaging data.

Clinical Imaging

Objective: Maintain an online archive, de-identification and upload clients and viewers for clinical images to support the storage and exchange of images between researchers in multicenter studies and enable re-use of imaging data.

Multicenter image review studies require a distribution infrastructure to be able to conduct external review of the images. The TraIT Bio-Medical Imaging Archive (BMIA) provides safe and sustainable storage of imaging data collected during research. The images can be analyzed directly online via image analysis software or downloaded to a locally preferred viewer.

See TraIT Tools (XNAT and CTP) for Tools and Services, Security and Privacy, Implementation and other features.

 

Rationale

Virtually every clinical study requires and/or produces images (e.g. PET, CT, MRI, Ultrasound). Clinical researchers working on the same project in different hospitals benefit from a simple and easy-to-use way to share such images. Currently, due to the large size of these files, storing images on a DVD and sending it by regular mail is often the most efficient way. Needless to say there is room for improvement here.

In addition, clinical images retain their value and can be of high interest to researchers working on other clinical studies, but also to technology developers working on data processing tools, visualization technologies or image analysis pipelines. Within TraIT Clinical Imaging, a central image archive, called the Biomedical Image Archive (BMIA) has been implemented. Participating hospitals and research centers can store clinical images - after local de-identification - in the BMIA and authorized users can retrieve images using a 'webshop' view and download functionality or can directly access images for more advanced image analysis applications.

 

Target groups

- Clinical researchers

- Technology developers involved in image analysis, visualization technologies etc.

 

Approach

Before BMIA, every hospital had to operate its own image archive, usually referred to as a PACS (Picture Archive Communication System). Also, hospitals are bound to strict rules in terms of de-identification when exporting images to external systems. Therefore, TraIT supplies and supports hospitals with a local de-identification and upload client called CTP (Clinical Trial Processor). This application removes and encrypts all the information linked to a patient's identity from the image itself and/or the image metadata. Once the images are anonymized and have passed a manual quality check by study administrators, they are made available on the BMIA. The accessibility of images can vary. Depending on the study requirements, some images will be available in the public domain, whereas others will be only accessible to authenticated and authorized researchers.

 

 

Status

- BMIA comprised of 4 tools, NBIA, CTP, keosys and XNAT, but NBIA and keosys are being decomissioned. Interested studies will be migrated to XNAT.

- In 2019 BMIA supports more than 100 image collections containing more than 20 TB of imaging data.

- the XNAT application and can handle more advanced image analysis application than NBIA did.

- CTP clients are now running in all university medical centers in the Netherlands and in many other hospitals in the Netherlands and abroad

 

Work Package Leader: André Dekker (MAASTRO Clinic)

andre.dekker <at> maastro.nl