Low-Latency 4k Medical Endoscopy

Introduction

With the development of technology, medical imaging, one of the major achievements of science and technology, has been used in various applications in the field of non-invasive diagnosis and treatment. One such application is endoscopy, which became more widespread in the 1990s when it became possible to transmit images to a monitor using a charge-coupled device. To help doctors better identify and locate lesions, manufacturers have continued to increase the resolution of endoscopes, with the resolution of human medical endoscopes gradually evolving from 1080p to today’s 4K.

Furthermore, the strategic incorporation of fluorescence and 3D technologies has revolutionized medical diagnostics and procedures. The Ministry of Industry and Information Technology, in its comprehensive “Medical Equipment Industry Development Plan (2021-2025)”, has meticulously outlined the breakthrough directions for medical equipment development, specifically focusing on advancing medical endoscopes and other diagnostic imaging equipment. This strategic planning ensures that policymakers are well-informed and actively involved in the advancements.

Challenge

The maximum image latency allowed for endoscopic applications in a clinical session is between 50 and 150 milliseconds. However, for surgical procedures, endoscopes need to respond in real-time or near real-time while performing functions such as image correction, color noise restoration, edge enhancement, scaling, etc.

Additionally, the terminal for this application should be as small as possible, with the display having 4K definition, 3D, fluorescence and support for SDI/HDMI interfaces. 4K requires only one HD camera, while fluorescence and 3D each require an additional HD camera, which poses a challenge regarding core board resources, data transfer and processing rates, and algorithmic efficiency.

Solution

The endoscopic system with Enclustra FPGA SoC as the core technology can do real-time 4K video streaming. 4K image sensors are responsible for capturing the image information, while the signal processing of the image is done by Enclustra’s Mercury+ XU8 FPGA (SoC) module.

The captured video stream is fed into the Mercury+ XU8 FPGA (SoC) module for image pre-processing, and the processed high-definition images are then presented to the surgeon on a high-definition monitor via a display interface via the image management and storage unit.

The Mercury+ XU8 core boards , which have long occupied the AMD homepage, are available in XCZU4CG, XCZU5EV, and XCZU7EV models. Users can choose a higher-end model when more resources are needed, making upgrading simple and convenient.

In addition to the Enclustra Mercury+ XU8 FPGA (SoC), other SoCs can be considered for this application, such as the Mercury+ XU5 and Mercury+ XU9 .

The Enclustra Mercury+ XU8 FPGA (SoC) module enables a high degree of hardware system integration, significantly reducing development time. At the same time, by supporting a variety of peripheral interfaces, future feature updates and expansions can be realized more quickly and easily.

Enclustra’s extensive product portfolio allows users to choose from a wide range of core board modules in the AMD Kintex-7, Zynq-7000, and Zynq Ultrascale+ MPSoC series. Enclustra’s core board modules are pin-compatible with most of the other core boards in its family (Mars, Mercury, Andromeda), which means that users can also plan a clear upgrade path, minimize the amount of work involved in upgrading, and even change the core board model on an ad-hoc basis during the course of project development.

Enclustra’s FPGA core board modules have a minimum expected lifecycle of 10 years or more, and the hardware is designed with a focus on future-proof availability and performance, with all products available for long-term delivery.

For more information on how the Enclustra Mercury+ XU8 FPGA SoC module can enhance your medical imaging applications, read the full article here .

#Enclustra #MercuryXU8 #MedicalImaging #Endoscopy #4KVideo #FPGA #EmbeddedComputing #Innovation #MedTech