Upgrades in Medical Endoscopy via Integrated, Flexible Expansion Solutions
9/3/2025
Industry Trend
AI is transforming medical diagnostics and treatment efficiency. According to the national library of medicine, “a 1% increase in adenoma detection rate (ADR) can reduce interval colorectal cancer risk by up to 3%.” This highlights the vital role of AI in endoscopic procedures, where early and accurate detection significantly improves outcomes. At the same time, global regulatory bodies like the FDA are introducing stricter requirements for AI-enabled medical devices. These regulations emphasize the need for improved accuracy, sensitivity, and specificity to enhance overall application safety, robustness, and effectiveness. As a result, it is essential for computing platform to support AI upgradeability to remain compliant and effective throughout their lifecycle.
Challenge
To accelerate AI development at the edge, discrete graphics card with vertical-specific software stacks plays an important role for solution iterations. However, these typically require full-sized motherboards with standard PCIe expansion cards, resulting in larger system footprints that are unsuitable for compact or mobile medical devices. On the other hand, smaller form-factor boards often lack the high-speed interfaces needed to support external AI accelerators, limiting their ability to handle advanced AI workloads in space-constrained medical applications. Moreover, medical devices are frequently required to perform multiple tasks simultaneously, such as image capture, real-time analysis, data post-processing, AI inference, and multi-screen display, leading to complex system architectures with dense wiring. This complexity increases deployment difficulty and raises the risk of operational errors in clinical settings.
Solution
Advantech’s MIO-5379 single board computer features USB4 technology, enabling high-speed transmission over USB, displayport, and PCI Express Gen.3 x4—delivering bandwidths of up to 40 Gb/s through a single cable. This integrated, high-performance interface helps developers reduce overall system size while maintaining robust AI capabilities. When paired with the USB4 interface on the MIOe-UMXM extension module, the MIO-5379 supports external GPUs and AI accelerator cards within a compact footprint, eliminating the need for a full system redesign. This flexible, cable-based expansion approach enhances configuration versatility, making it ideal for compact, AI-powered medical endoscopy devices.
Simplifying Clinical System Design with USB Type-C
To reduce wiring complexity in limited clinical environments, the client sought a streamlined connectivity solution without compromising performance. Advantech responded by integrating its MIOe-TCPD extension module—featuring USB Type-C—into the MIO-5379 single board computer. This enabled display signal transmission via alternate mode or USB communication for touchscreens, and up to 100W Power Delivery—all through a single standardized cable—greatly simplifying system design, lowering installation costs and space, enhancing reliability to meet the demands of modern clinical settings.
Result
The client ultimately selected the MIO-5379, paired with the MIOe-UMXM and MIOe-TCPD extension modules. This combination delivers a compact yet powerful solution—supporting external GPU expansion and leveraging Type-C PD to streamline medical deployment. The result is a next-generation mobile endoscopy system that is AI-upgradable, space-efficient, and ready to meet both clinical demands and evolving regulatory standards.
-25/MIO-5379_03-20250120135305.png)
-25/MIOe-TCPD_04-拷貝-(5)20250505164422.png)
-25/420231208083115.jpg)
