From image-guided surgery to vision-equipped service robots, real-time video is enabling new levels of precision and treatment while driving fundamental changes in how healthcare services are delivered. However, as medical imaging applications multiply and systems become more complex, designers are increasingly challenged to improve system usability and drive down costs.

Fig. 1 – Images from an X-ray detector and lamp head camera are converted to GigE Vision and multicast to an operating room dashboard and computing platforms used for image processing, storage, and monitoring in a control room.
Even as imaging systems handle increasingly sophisticated analyses, they must be intuitive and easy to use for staff in operating rooms, nursing stations, and healthcare clinics. Budget pressures mean performance enhancements must be achieved without sacrificing investments in existing imaging and processing equipment. Finally, systems must be simple and cost-effective to maintain and scale.

An important first step towards meeting these challenges is choosing the right video interface—the hardware and software used to format imaging data, send it over a cable or wirelessly, and receive it at a computer or display. Although the video interface is a small part of the overall system, it has a large impact on the usability, cost, and future scalability of the final product.

This article describes the video interfaces used today in medical imaging systems for digital radiography and compares them to two interfaces—GigE Vision® and USB3™ Vision—initially developed for and deployed in industrial machine vision applications. It discusses the cost and performance benefits of GigE Vision and USB3 Vision in medical imaging, with specific examples of how they can be used in networked operating rooms and telepresence applications.

Machine Vision and Medical Imaging

Fig. 2 – USB3 Vision video interfaces reduce component count, costs, and system complexity while extending the operating life of battery-powered service robots.
A bin-picking factory robot and an image-guided surgery system may seem worlds apart, but video interface standards developed for industrial applications are playing an important role in the continuing evolution of medical imaging.

Like their high-speed industrial counterparts, medical imaging applications require video interfaces that can reliably transfer high-resolution imaging data in real time between cameras or image sensors and computers or displays with low, consistent latency (or delay). Legacy medical imaging systems typically rely on point-to-point interfaces based on custom equipment to meet these requirements. However, developing proprietary interfaces is expensive, time-consuming, takes valuable resources away from developing core functionality, adds maintenance costs, and poses scalability challenges in multi-vendor environments.

As a result, many medical designers have leveraged video standards from other markets, including the TIA/EIA-644 low-voltage differential signaling (LVDS) standard commonly found in telecom and consumer applications, the HDMI/DVI television standard, and the Camera Link® machine vision standard. While all three of these standards support high-performance, real-time video delivery, they require a dedicated connection between each camera and end-point and a PCiE frame grabber to capture data. As medical imaging systems grow in sophistication, these limitations drive up cost and system complexity.

For example, in applications where images are displayed across multiple screens, such as image-guided surgery, the cabling required for umbilical connections becomes costly and difficult to manage and scale. Moreover, the PCIe frame grabber needed at each endpoint limits the types of computers that can be used, drives up component costs, and increases complexity. Endusers are also “locked in” to the frame grabber vendor for support, relying on them to write drivers for specific operating systems and processing architectures. In addition, expensive switching equipment is required to support realtime video networking.

The machine vision industry faced similar issues, and in response released the GigE Vision standard for high-speed video delivery over commercial GigE equipment in 2004. GigE Vision enables the transmission of full-resolution, uncompressed video with low, consistent latency and device commands between an imaging source and an existing port on a computing platform over any Ethernet connection, including over a GigE, 10 GigE, or 802.11 wireless. GigE Vision additionally supports transmission of DICOM-compatible metadata and compressed images (JPEG, JPEG 2000, and H.264). Today, GigE Vision is the most widely used video interface standard in industrial applications and is steadily gaining a foothold in medical imaging systems.

GigE Vision is a natural choice for video transmission in a wide range of medical imaging modalities. The standard allow medical imaging system designers to fully support required point-to-point connections, while gaining the flexibility of video networking, the ability to interwork with a range of computing platforms, and the benefits of lightweight, low-cost cabling or wireless connectivity. In addition, as the standard has gained widespread adoption, a broad selection of off-the-shelf, GigE Visioncompliant products from global vendors helps address interoperability concerns.

GigE Vision in the Networked Operating Room

The wide range of commercially available GigE Vision-compliant video and networking products, including external frame grabbers and embedded hardware, make it relatively straightforward for designers to create a real-time networked operating room without sacrificing investments in existing equipment.

In the example shown in Figure 1, images from an digital X-ray detector in a C-arm are sent over an existing Camera Link or LVDS interface to an external frame grabber, where the images are converted to a GigE Vision-compliant video stream. Alternatively, embedded video interface hardware allows designers to directly integrate GigE Vision-compliant video connectivity into flat panel detectors (FPDs) that can fit into existing systems as replacement for film-based panels.

A second external frame grabber converts images from a camera mounted in the lamp head into the same compliant GigE format. These image feeds are aggregated at an off-the-shelf Ethernet switch and multicast to processing, analysis, display, and recording equipment. Per-frame metadata, such as a precise time-stamp of image acquisition and sensor settings, is transmitted with the images over the Ethernet link for easy integration with DICOM-compliant software and hardware.

The long reach of Ethernet, 100 meters point-to-point over ordinary Cat 5/6 cabling, means processing and analysis equipment can be located outside the sterile environment. This reduces the costs of sterilizing equipment, lowers the risks of patient infection, and allows data to be easily shared across multiple departments. The more flexible cabling also enables a wider degree of movement for robotic applications. With GigE Vision delivered over an 802.11 wireless link, portable X-ray panels can be better positioned to help reduce complexity in the operating room.

At the computer, the video streams in through the Ethernet port, allowing the use of lower cost computing platforms. The video processor highlights areas of interest, including pre-op images, and overlays vital signs information. The composite image is then multicast over the Ethernet network to various displays. In the operating room, an external frame grabber converts the GigE Vision image stream to HDMI/DVI signals for viewing on a high-definition dashboard monitor used by the surgeon to track real-time patient data from different imaging devices and systems.

Expanding the Operating Theatre

One of the key advantages of GigE-based distributed network architectures is the ability to locate intelligent nodes at locations where data collection and control occurs and create video distribution groups with a single server multicasting data to several clients.

Integrating previously isolated image sources and patient data onto a common network and aggregating the information on a single dashboard can increase the situational awareness of operating room staff. In the operating room, for example, the single screen dashboard displays real-time patient data from different imaging devices and systems. The surgeon or operating team members can easily switch between imaging sources, such as white light and fluoroscopic cameras and pre-operative and real-time images, without configuring hardware or software. The image destination can also be easily changed, with real-time video from the operating theatre transmitted to OR scheduling staff, a conference room, multiple departments, or shared with remote specialists.

At the transport layer, the imaging device sends only one copy of the data to a network switch. The Ethernet switch replicates the data for distribution to display panels and processing platforms as required. This ensures video distribution doesn’t impact server performance. Leveraging Ethernet’s inherent multicast capabilities, display and processing functions can be distributed from a single device to multiple devices to help ensure reliability. This also helps GigE Vision cameras to multicast high-quality, uncompressed video to multiple displays and processing nodes simultaneously with the lowest possible latency.

Interfaces based on GigE Vision also speed the design and boost the performance of advanced applications. In full-motion video applications, for example fluoroscopy that uses multiple moving X-ray sources to obtain real-time images of a patient, legacy umbilical interfaces are uneconomical and cumbersome. With 10 GigE interfaces, which support ten times the bandwidth of GigE, multiple image sources can be transmitted simultaneously over a switched Ethernet network to a processor for 3D image generation.

USB 3.0 in Medical Imaging

Building on the concepts developed for GigE Vision, the machine vision industry standardized the transport of high-speed imaging and video data over a USB 3.0 cable with the release of USB3 Vision in February 2013.

With USB3 Vision, video and data (including metadata for DICOM compliance) is transmitted from cameras and sensors to existing ports on a computer, laptops, or tablet over flexible USB 3.0 cabling. The USB 3.0 bus delivers throughput approaching 3.0 Gb/s over short distances, and is ideal for applications such as transmitting images from a microscope camera directly to a port on a laptop or tablet with plug-and-play ease.

For robot manufacturers, an off-the-shelf USB3 Vision interface shortens time-to-market and allows R&D resources to be focused on system design and data analysis. In the example shown in Figure 2, images from Sony block cameras used for inspection and navigation in a service robot are converted to USB3 Vision-compliant video streams. The video can then be transmitted over high-bandwidth, flexible, lower cost USB cables directly to ports on an integrated single-board computing platform. By eliminating PCIe (Peripheral Component Interconnect Express) frame grabbers within the robot, designers can reduce system complexity, component count, and costs. In addition, decreasing the weight and power consumption of the robot extends battery life, which translates into more patient visits between charges.

The Right Design Choice

Starting with the first clinical use of Xrays almost 120 years ago, medical imaging has played an ever-increasing role in healthcare. Today, almost all aspects of care, from initial examination, to surgery, and nursing, rely on real-time video to identify issues, make accurate diagnoses, and provide treatments.

While the video interface is a small part of an overall medical vision system, choosing the right interface will deliver significant design advantages for manufacturers, cost-savings for healthcare providers, and performance benefits to help improve patient comfort and care. Designing or upgrading medical imaging systems based on off-the-shelf GigE Vision and USB3 Vision interfaces allows manufacturers to shorten time-to-market, reduce risk, and lower system cost and complexity, while delivering interoperability and performance benefits to enhance the value of their solutions.

This article was written by Rudi Rincker, Vice President of Business Development, Pleora Technologies, Kanata, Ontario, Canada. For more information, Click Here .