Back to the list
Congress: ECR25
Poster Number: C-18800
Type: Poster: EPOS Radiologist (educational)
Authorblock: T. Kimpe, E. De Brauwer, L. De Paepe, O. De Wel, A. Xthona; Kortrijk/BE
Disclosures:
Tom Kimpe: Employee: Barco NV
Elie De Brauwer: Employee: Barco NV
Lode De Paepe: Employee: Barco NV
Ofelie De Wel: Employee: Barco NV
Albert Xthona: Employee: Barco NV
Keywords: Breast, Computer applications, eHealth, Image manipulation / Reconstruction, PACS, Teleradiology, Computer Applications-Teleradiology, Image compression, Teleradiology, Image verification
Learning objectives

Radiology viewing solutions increasingly are making use of streaming technology, where images are rendered on a server and then sent over a network to a client to visualize these images. In such a configuration, applications such as the EMR or PACS viewer are (mostly) running on the server side, while on the client side the primary functionality is to decode the image stream, visualize images on the client display, and to feedback user interaction such as mouse and keyboard to the server.

Viewing of these streamed images on the client side can take place in a viewing application, a browser environment, or a virtual desktop environment (VDI, eg. Citrix, VMWare).

Streaming technology has many advantages. An important benefit is increased security, as images and PHI remain on the server and are never stored on the client. Moreover, in the case of VDI the entire environment (including operating system, drivers, libraries, applications, etc.) is managed on the server side, hereby making it much easier for IT departments to enforce configuration control and to ensure (security) patches are timely performed. A second benefit is that the client PC can be a mini PC or regular laptop or desktop computer. As applications are executed on the server side, high-end compute specifications are not needed on the client side.

But streaming technology also faces important challenges. The connection between the server and the client has limited bandwidth, will suffer from packet loss and add latency and jitter. To overcome limited network bandwidth, lossy or lossless compression is typically used and this can introduce artifacts. Network characteristics and server load (eg. number of concurrent users on the server) will fluctuate over time, and this can result into unpredictable and varying quality of medical images.

But also the configuration of the VDI system (eg. Citrix, VMWare) itself has a strong influence on the quality that can be achieved. Important settings include for example image codec settings (lossless, lossy, quality levels); number of allowed concurrent users; server side and client side display settings (resolution, color, framerate settings).

The learning objective is to educate on the quality implications of using streaming technology.

GALLERY