Empowering Real-Time Eye Health Diagnostics with ASUS IoT PE4000G Edge AI Computers
Background: Why poor global eye health is a growing concern
Vision problems are becoming more prevalent in modern society, as changing habits such as increasing screen time contribute to eye strain, digital fatigue, and a rise in conditions like myopia and dry eye syndrome worldwide.
Myopia, for instance, more commonly referred to as short-sightedness and characterized by an inability to see objects far away clearly, will affect more than half the world's population1 by 2050 based on PubMED Central study. Uncorrected, it can disrupt reading and learning in children, significantly impeding educational outcomes. For adults, it can severely limit daily activities and impact economic, emotional, and social well-being.
Consequently, ophthalmologists are keen to adopt new diagnostic technologies to help them more accurately assess the condition of people's eyes and prevent vision problems from emerging. One increasingly popular technique is to use camera-based eye examination systems equipped with real-time, efficient, and stable computing capabilities to perform complex image processing and visual analysis tasks.
Ophthalmologists use this technology to capture high-resolution images of the back of the eye. This information is fed into a powerful processor using artificial intelligence to automatically examine these images and look for problems or signs that would make specific types of eye surgery risky. Instead of a trained professional having to spend time manually studying each image, the AI can immediately highlight abnormalities, improving the speed of diagnoses and patient outcomes.
The emergence of these advanced AI-inspired solutions is driving rising revenues in vision-based health markets. Indeed, the market size of ophthalmic devices has expanded rapidly in recent years. According to Global Information, Inc., it will grow from $46.68 billion in 2024 to $51.75 billion in 2025 at a compound annual growth rate of 10.9% and could grow to $70.64 billion in 2029.
"There is clearly a tremendous amount of innovation and commercialization around vision-based technologies, and that bodes well for better eye health worldwide," says Phoebe Peng, product manager with ASUS IoT. "Many of these new technologies are enabled by high-performance edge computing, AI integration and camera compatibility. ASUS IoT works closely with ophthalmic device suppliers through an end-to-end approach that gets innovation to market quickly and efficiently."
Solution: Powerful Edge computing for real-time inferencing
So let's look in greater detail at how ASUS IoT underpins eye health advances. At the heart of any camera-based eye diagnostic system are advanced computing capabilities that can perform complex image processing and visual analysis tasks in real-time at the edge. ASUS IoT has developed the PE4000G Edge AI Industrial Computer for such applications. It is a highly expandable and stable platform tailored to support AI-driven intelligence in ophthalmology equipment.
In more detail, the PE4000G is powered by 14th/13th/12th Gen Intel® Core™ processors. AI application models for ophthalmology equipment require CPU support for AI pre-processing and planning. The ASUS PE4000G supports latest Intel® Core™ processors, which can meet the high-performance computing needs of all-femtosecond applications in different service scenarios. The ASUS PE4000G is designed to support GPU cards with power consumption up to 220W, offering flexibility for a wide range of high-performance GPUs. For example, it is compatible with models such as the NVIDIA RTX™ Ada series, enabling enhanced image processing, synthesis performance and real-time AI inference at the edge through deep learning and CNN modules.
In laser eye treatments, deep learning AI systems help medical teams make better decisions. These systems can analyse detailed maps of the eye’s surface and detect eye conditions like keratoconus with very high accuracy - up to 99.3%2. They’re also excellent at correctly identifying people who do have the condition (99.4% sensitivity) and those who don’t (98.7% specificity). This performance is far better than older methods, which aren’t as accurate. Real-time image synthesis and AI inference during procedures enable enhanced diagnostic precision and surgical planning, allowing for more personalized treatment approaches that account for individual irregularities and ultimately improve patient outcomes through data-driven clinical decision making.
In terms of I/O configuration, the PE4000G features a modular design, providing powerful expansion capabilities and supporting up to 15 sufficient USB ports. It facilitates the integration of multi-modal image input devices, touch screens, and storage modules for medical customers, enabling engineering teams to complete medical system integration and deployment quickly. In addition, all USB ports are on the same side, such design allows easy cable management. Also, PE4000G is DC-in design, so 8-48V wide voltage support makes it easier for equipment vendors to integrate this edge AI computer into the current system. Moreover, the 2xhot swappable SSD facilitates system maintenance.
In terms of hardware reliability, the product features industrial-grade heat dissipation mechanisms that can withstand the environmental requirements of high temperatures and prolonged operation in clinical settings. Equipment racks often contains several diagnostic devices, including a Biometer, Corneal Tomographer, and Wavefront Aberrometer, which typically result in higher temperatures throughout the system. The PE4000G’s -20~60° wide temperature range and innovative thermal design are crucial for meeting the requirements of medical equipment integration applications, ensuring long-term stable operation of the equipment, and reducing maintenance risks. The PE4000G uses passive cooling for the Intel® Core™ Processors (Series 2) mainboard. Its thermal module features an aluminium block near the CPU and Platform Controller Hub (PCH) to efficiently dissipate heat. The module also incorporates a fin design that maximizes the contact surface area, enhancing thermal efficiency.
The PE4000G is also designed to meet the stringent requirements of the MIL-STD-810H standard for shock and vibration resistance. In a hospital setting, eye diagnostic equipment tends to be moved from one clinic room to another with live status. Therefore, anti-shock design is crucial for meeting mobile demand. Furthermore, PE4000G introduces an innovative cable locker mechanism that prevents cable fall-off during intense shock and vibration. Additionally, there is an anti-vibration damper on the wall mount for added protection.
ASUS IoT has also considered flexibility and future-proofing. The PE4000G is part of a family of Edge AI computer solutions that specialize in PCIe GPU computing, with other variants including the PE5101D, PE6000G, and PE8000G. This choice means customers have options whatever their AI application.
Overall, the technical capabilities of the PE4000G provide an ability to process high-resolution images in real-time, maintain stable operation in demanding clinical environments, and integrate seamlessly with hospital systems. These attributes are critical for ophthalmic systems, as they directly impact diagnostic accuracy, speed, and patient outcomes.
ASUS IoT PE4000G Edge AI Computer in Ophthalmic Laser System
Outcome: How ASUS IoT delivers end-to-end advantages
To support ophthalmic system suppliers in bringing innovation to market, ASUS IoT offers an end-to-end, one-stop-shop approach to product development, including customization, deployment and ongoing support. "We offer a unique service to ophthalmic device makers and the broader healthcare sector," says Phoebe Peng. "As a company. ASUS IoT has extensive domain knowledge with dedicated teams servicing the tier 1 medical suppliers. We can supply consumer, commercial and industrial grade solutions, and cutting-edge technology in AI with GPU and CPU. This makes for a compelling offering for device developers and suppliers across the healthcare market."
ASUS IoT also operates globally, with localized technical support and supply chain stability. This 'global but local' approach helps equipment manufacturers accelerate product introduction into medical environments - regardless of location. "It also means we can help our customers meet medical equipment regulations and testing requirements of different regions, helping equipment manufacturers shorten development and time-to-market," adds Phoebe Peng.
Looking to the future, the PE4000G is likely to be expanded for applications in telemedicine, smart ophthalmology clinics, vision screening vehicle systems, and even real-time image analysis and visual assistance in operating rooms. Its highly modular design allows medical equipment manufacturers and system integrators to flexibly combine modules according to different needs to create truly customized AI solutions, improving diagnostic efficiency and clinical effectiveness.
"Development of the PE4000G and the broader family of products shows how committed we are to servicing healthcare customers with the latest cutting-edge technologies," adds Phoebe Peng. "ASUS IoT is the partner of choice for Edge AI computers for medical applications, and we look forward to helping drive advances in eye health and indeed in many other related areas."
Citation
1. https://pmc.ncbi.nlm.nih.gov/articles/PMC8759558/
2. https://www.nature.com/articles/s41598-023-46903-5