Hardware Development Trends

This might look like a hologram, but it is actually a real MRI scan reconstructed as an interactive 3D model. The Illumetry IO display uses stereoscopic glasses and head tracking to show a slightly different image to each eye. As the viewer moves, the perspective changes with them, making the anatomy appear to extend out of the screen. A stylus can then be used to move and explore the model without wearing a bulky VR headset. The colours are added during segmentation to distinguish anatomical structures such as bones or muscles.  The technology exists, but this particular demonstration should be viewed primarily as a new interface for medical visualisation, instead of a clinically proven replacement for radiologists scrolling through MRI slices.

China just bent the rules of electronics — literally. Facinating? Chinese and global researchers are advancing Metal-Polymer Conductors (MPCs) — circuits made from liquid metals like gallium–indium embedded in elastic polymers — that defy traditional rigid wiring by remaining conductive even when stretched up to 500% or more. Why this is a big deal: 🔹 High Stretchability: Certain liquid-metal conductors maintain electrical conductivity even when stretched 5× their original length. 🔹 Durability: Printable metal-polymer conductors can withstand over 10,000 cycles of stretching with minimal resistance change (<3%). 🔹 Conductivity: Hybrid conductors based on indium alloys can achieve extremely high conductivity (~2.98 × 10⁶ S/m) with minimal resistance change under extreme strain. 🔹 Fine Feature Sizes: Advanced techniques can pattern circuits as small as 5 micrometers, rivaling conventional PCBs. Market Insight: The global market for wearable and flexible devices is expected to surge into the hundreds of billions of dollars, with advanced stretchable materials at the core of the next wave of innovation. (Wearable tech projected >US$150B by 2026 in soft electronics growth — wearable industry data) Where AI Fits In: AI is not just hype — it’s accelerating how we design and discover materials like MPCs. AI/ML models help predict material properties — like conductivity and mechanical resilience — before physical prototypes are made. Computational simulations can evaluate thousands of polymer + metal combinations far faster than physical testing alone. AI-assisted optimization reduces lab iterations, cutting time and cost in early-stage development. In other words: AI + materials science = faster discovery of smarter, stretchable electronics. Potential Applications: Soft robotics that mimic human motion Wearables that feel like fabric Artificial skin with embedded sensing Health monitoring devices that conform to the body On-skin motion recognition and bioelectronics. The era of electronics you can twist, stretch, and wear is here — and AI is helping make it a reality. #FlexibleElectronics #MaterialsScience #AIinInnovation #SoftRobotics #WearableTech #DeepTech #FutureOfElectronics #Innovation

Today, Science Robotics has published our work on the first drone performing fully #neuromorphic vision and control for autonomous flight! 🥳 Deep neural networks have led to amazing progress in Artificial Intelligence and promise to be a game-changer as well for autonomous robots 🤖. A major challenge is that the computing hardware for running deep neural networks can still be quite heavy and power consuming. This is particularly problematic for small robots like lightweight drones, for which most deep nets are currently out of reach. A new type of neuromorphic hardware draws inspiration from the efficiency of animal eyes 👁 and brains 🧠. Neuromorphic cameras do not record images at a fixed frame rate, but instead have the pixels track the brightness over time, sending a signal only when the brightness changes. These signals can now be sent to a neuromorphic processor, in which the neurons communicate with each other via binary spikes, simplifying calculations. The resulting asynchronous, sparse sensing and processing promises to be both quick and energy efficient! 🔋 In our article, we investigated how a spiking neural network (#SNN) can be trained and deployed on a neuromorphic processor for perceiving and controlling drone flight 🚁. Specifically, we split the network in two. First, we trained an SNN to transform the signals from a downward looking neuromorphic camera to estimates of the drone’s own motion. This network was trained on data coming from our drone itself, with self-supervised learning. Second, we used an artificial evolution 🦠🐒🚶♂️ to train another SNN for controlling a simulated drone. This network transformed the simulated drone’s motion into motor commands such as the drone’s orientation. We then merged the two SNNs 👩🏻🤝👩🏻 and deployed the resulting network on Intel Labs’ neuromorphic research chip "Loihi". The merged network immediately worked on the drone, successfully bridging the reality gap. Moreover, the results highlight the promises of neuromorphic sensing and processing: The network ran 10-64x faster 🏎💨 than a comparable network on a traditional embedded GPU and used 3x less energy. I want to first congratulate all co-authors at TU Delft | Aerospace Engineering: Federico Paredes Vallés, Jesse Hagenaars, Julien Dupeyroux, Stein Stroobants, and Yingfu Xu 🎉 Moreover, I would like to thank the Intel Labs' Neuromorphic Computing Lab and the Intel Neuromorphic Research Community (#INRC) for their support with Loihi (among others Mike Davies and Yulia Sandamirskaya). Finally, I would like to thank NWO (Dutch Research Council), the Air Force Office of Scientific Research (AFOSR) and Office of Naval Research Global (ONR Global) for funding this project. All relevant links can be found below. Delft University of Technology, Science Magazine #neuromorphic #spiking #SNN #spikingneuralnetworks #drones #AI #robotics #robot #opticalflow #control #realitygap

The first Apple Vision Pro-assisted cataract eye surgery was just completed successfully. An ophthalmologist in San Diego, Dr. Tommy Korn, used Apple Vision Pro during a live cataract procedure as part of a clinical study at Sharp HealthCare. Instead of constantly turning between microscopes, monitors, and different screens during surgery, he used the headset to view critical patient data, surgical imaging, and real-time visual overlays directly in front of him. The setup also included a ZEISS digital surgical microscope and ClearSphere software to create a more immersive operating environment. The goal of the study is simple. See whether spatial computing can improve surgical precision, depth perception, workflow, and surgeon ergonomics while reducing the physical strain doctors have dealt with for decades. As Dr. Korn said, surgeons have been sitting in front of microscopes since 1946, and he doesn’t want to go back to the old way. And honestly… this makes a lot of sense. $3,500 sounds expensive for consumers. For hospitals and surgical centers, that cost is almost nothing if it helps doctors operate more efficiently and potentially improves patient outcomes. We spent years using VR headsets for gaming, entertainment, and watching courtside NBA games. Now we’re starting to see what happens when the same technology enters operating rooms. This is where things get really interesting. Healthcare might be one of the biggest winners of the next wave of consumer tech. Follow Endrit Restelica for more tech stuff.

💡 𝗖𝗵𝗼𝗼𝘀𝗶𝗻𝗴 𝘁𝗵𝗲 𝗿𝗶𝗴𝗵𝘁 𝗺𝗲𝗺𝗼𝗿𝘆 𝗶𝘀 𝗼𝗻𝗲 𝗼𝗳 𝘁𝗵𝗲 𝗺𝗼𝘀𝘁 𝗶𝗺𝗽𝗼𝗿𝘁𝗮𝗻𝘁 𝗮𝗿𝗰𝗵𝗶𝘁𝗲𝗰𝘁𝘂𝗿𝗮𝗹 𝗱𝗲𝗰𝗶𝘀𝗶𝗼𝗻𝘀 𝗶𝗻 𝗲𝗺𝗯𝗲𝗱𝗱𝗲𝗱 𝘀𝘆𝘀𝘁𝗲𝗺𝘀. Speed, cost, power consumption, endurance, and persistence all depend on which memory you select. 👉 This diagram summarizes the complete memory landscape used in modern embedded systems. Let’s break it down. ⚡ 𝐕𝐨𝐥𝐚𝐭𝐢𝐥𝐞 𝐌𝐞𝐦𝐨𝐫𝐲 (𝐃𝐚𝐭𝐚 𝐥𝐨𝐬𝐭 𝐰𝐡𝐞𝐧 𝐩𝐨𝐰𝐞𝐫 𝐢𝐬 𝐨𝐟𝐟) Used for runtime execution and temporary data. 🧩 𝐂𝐏𝐔 𝐈𝐧𝐭𝐞𝐫𝐧𝐚𝐥 𝐌𝐞𝐦𝐨𝐫𝐲 Integrated directly inside the microcontroller. 𝐑𝐞𝐠𝐢𝐬𝐭𝐞𝐫 𝐌𝐞𝐦𝐨𝐫𝐲  • Fastest memory in the system  • Used directly by CPU instructions  • Stores operands and execution state 𝐂𝐚𝐜𝐡𝐞 𝐌𝐞𝐦𝐨𝐫𝐲  • Stores frequently accessed data  • Reduces external memory latency  • Major performance accelerator 🧠 𝐑𝐀𝐌 — 𝐑𝐮𝐧𝐭𝐢𝐦𝐞 𝐃𝐚𝐭𝐚 𝐒𝐭𝐨𝐫𝐚𝐠𝐞 𝐒𝐑𝐀𝐌 (Static RAM)  • Extremely fast access  • No refresh required  • High silicon cost 👉 Used for CPU cache, buffers, stacks 𝐃𝐑𝐀𝐌 (Dynamic RAM)  • Higher density than SRAM  • Requires periodic refresh 👉 External system memory 𝐒𝐃𝐑𝐀𝐌  • Clock-synchronized DRAM  • Higher bandwidth operation 👉 Main memory for MPU/SoC systems 🔋 𝐍𝐨𝐧-𝐕𝐨𝐥𝐚𝐭𝐢𝐥𝐞 𝐌𝐞𝐦𝐨𝐫𝐲 (𝐃𝐚𝐭𝐚 𝐫𝐞𝐭𝐚𝐢𝐧𝐞𝐝 𝐰𝐢𝐭𝐡𝐨𝐮𝐭 𝐩𝐨𝐰𝐞𝐫) 📘 𝐑𝐎𝐌 𝐅𝐚𝐦𝐢𝐥𝐲 𝐌𝐚𝐬𝐤 𝐑𝐎𝐌  • Programmed during manufacturing  • Not reprogrammable 👉 Mass production devices 𝐏𝐑𝐎𝐌  • Programmable once 👉 One-time configuration 𝐄𝐏𝐑𝐎𝐌  • UV erasable  • Reusable but slow erase process 𝐄𝐄𝐏𝐑𝐎𝐌  • Electrically erasable  • Byte-level write access  • Limited write cycles 👉 Configuration storage ⚙️ 𝐅𝐥𝐚𝐬𝐡 𝐌𝐞𝐦𝐨𝐫𝐲 𝐍𝐎𝐑 𝐅𝐥𝐚𝐬𝐡  • Random read access  • Execute-In-Place (XIP) possible 👉 Firmware storage 𝐍𝐀𝐍𝐃 𝐅𝐥𝐚𝐬𝐡  • Block-based access  • Very high density 👉 Data storage & file systems 𝐞𝐌𝐌𝐂  • NAND Flash + embedded controller  • Wear leveling & management included 👉 Mass embedded storage 💾 𝐍𝐕𝐑𝐀𝐌  • Combines RAM behavior + Flash persistence.  • Byte-level updates  • Fast writes  • Retains data after reset 👉 Ideal for: Retain critical system state across power loss There is no universal best memory. Good embedded architecture balances: Speed, Power consumption, Cost, Endurance, Persistence 💬 Which memory type do you use most in your projects — SRAM, SDRAM, or Flash?

New chip matches human brain speed for the first time... Researchers (see citation below) introduced the world’s first chip capable of matching the operational speed of the human brain. Fabricated using a standard 40-nanometer process, the sub-10-millisecond neural dynamical system utilizes phase-change memristors to execute key mathematical calculations directly within memory. Occupying just 0.28 square millimeters, the chip achieves up to a 478-fold speedup over enterprise GPUs in real-time cortical surface reconstruction while drastically lowering energy demands. This hardware milestone paves the way for real-time brain-computer interfaces, intraoperative surgical navigation, and full-scale digital brain twins. Key Facts 🧠 Sub-10-Millisecond Brain-Speed Simulation: The chip processes continuous neural dynamics at operational speeds matching the native millisecond temporal scale of the human brain. 🧠 Massive GPU Acceleration: In 3D cortical surface reconstruction tasks (mapping brain white and gray matter folds), the phase-change memristor chip achieved up to a 478.18× speedup compared to an enterprise-grade NVIDIA A100 GPU. 🧠 Superior Energy and Latency Metrics: Compared to state-of-the-art Application-Specific Integrated Circuits (ASICs), the neuromorphic architecture operates 3.82× to 36.27× faster while consuming 11.75× to 24.73× less energy. 🧠 Overcoming the Memory Wall: By utilizing in-memory computing with 9 pipeline stages running at 50 MHz, the system eliminates traditional data-shuttling overheads between memory and CPU/GPU processors. 🧠 High-Fidelity Anatomical Reconstruction: The system generated smooth, closed, and topologically accurate 3D manifold cortical meshes, scoring exceptionally high on Average Symmetric Surface Distance (ASSD) and Hausdorff Distance metrics for neuroimaging accuracy. Why it matters Fast and accurate brain modeling is important for technologies that must respond in real time, including brain–computer interfaces, surgical navigation, and medical imaging. Existing hardware often requires too much time and power for these demanding calculations. By performing key operations directly in memory, the new chip reduces data movement and brings high-quality brain modeling closer to real-time use. Based on article circulated by @neuorosciencenews Research: “A sub–10-millisecond neural dynamical system based on phase-change memristors” by Lei Cai, Yaoyu Tao, Chenchen Xie, Longhao Yan, Shiqian Li, Ruihong Shen, Zelun Pan, Xile Wang, Bowen Wang, Daijing Shi, Yihang Zhu, Teng Zhang, Yixin Zhu, Xi Li, Zhitang Song, Ru Huang, Yuchao Yang. Science DOI:10.1126/science.aee6277 sharing for interest: Raoul Groening James Moore Monet Stuckey Ankit Aggarwal Aditi Adlakha Malaika Humpy Anthony Gray Divyabh Mishra Mads Frank Tom Chapman Marcin Wawryszczuk, PhD, MBA, PMP, PSM Anna Lazar Denis Astapchenia Christian Hammerschmidt Nur Alam Jorge F. Guedes, Ph.D. Leeds University Business School Antonino Sgalambro

7 wearable and sensor developments you should know about this month: 🔘identifyHer’s new wearable,Peri, tracks physiological patterns linked to perimenopause, giving users real-time insight into hot flashes, sleep shifts and mood changes in a space where clinical-grade tools have been scarce. 🔘 Lampsy Health has developed a smart lamp that uses AI to detect epileptic seizures with over 99% accuracy, reducing false alarms and moving seizure monitoring into everyday environments. 🔘 Withings BeamO has received FDA clearance for its handheld device that combines a thermometer, ECG and digital stethoscope, enabling basic heart, lung and temperature checks at home in about a minute. 🔘 Lumia™ has launched Lumia 2, smart earrings with in-ear sensors that track blood flow to the head to capture signals around energy, focus, mental clarity, temperature and sleep. 🔘 Kohler Health has introduced Dekoda, a sensor-enabled toilet accessory that uses AI to analyse urine and stool, flagging hydration issues, gut-health changes and potential blood. 🔘 Apple is reportedly developing Apple Health+, an AI health-coaching layer for the Health app. Samsung and Google have moved early here, but Apple’s entry usually signals a category shift toward mainstream adoption. 🔘 Researchers at UC San Diego have built a battery-free electronic sticker that turns any drinking cup into a sensor by analysing vitamin C from fingertip sweat, a modest biomarker, but an interesting signal for near-frictionless diagnostics 👇links to news in comments #digitalhealth #ai #wearables

When it comes to low-power AC/DC converters, managing efficiency while maintaining performance in standby mode is a challenge. Many power supplies spend much of their time in a "standby" state, where energy efficiency and quick recovery from low-load conditions become critical. The problem becomes even more pronounced in applications like mobile devices, where minimizing standby power is a top priority. While there are ways to reduce power consumption, the trade-offs—especially in terms of transient load response and the complexity of the control scheme—can be significant. One solution to this challenge is the use of Primary Side Regulation (PSR) in flyback converters. PSR offers several advantages, such as reducing component count, improving reliability, and cutting down on size and cost. By leveraging magnetic feedback from an auxiliary transformer winding, PSR enables precise voltage regulation while minimizing standby power dissipation. However, achieving the ideal performance with PSR isn't without its challenges. Accurate voltage sampling, managing leakage inductance, and addressing transient responses require careful attention to detail. Still, the potential benefits in terms of reduced standby power and improved efficiency are clear. This white paper from Texas Instruments gets into the complexities of designing low-power AC/DC converters with PSR, covering everything from voltage regulation errors to the impact of switch-node capacitance on overall efficiency. It provides actionable insights on how to navigate these design considerations for optimal performance.

Is AI's Growth Sustainable? How to Make Generative Applications Greener. The rise of generative AI tools like ChatGPT and others has been remarkable, but their environmental impact is often overlooked. The data center industry, housing these systems, accounts for up to 3% of global greenhouse gas emissions, with energy consumption doubling every two years. Hyperscale cloud providers like Amazon AWS, Google Cloud, and Microsoft Azure play a significant role in powering these models, leading to major carbon footprints. Understanding the carbon footprint lifecycle of AI models is crucial. Large generative models consume extensive energy during training, and fine-tuning can be a more energy-efficient option. Inference sessions, though less energy-intensive, involve many more sessions, contributing to ongoing energy consumption. Efforts to reduce energy usage include employing less computationally expensive approaches like TinyML and using large models only when significantly valuable. To make AI greener, companies can use existing models from providers instead of creating new ones. Fine-tuning existing models on specific content domains consumes less energy and provides more value. Utilizing energy sources from carbon-friendly regions and monitoring carbon emissions can significantly reduce AI's environmental impact. Reusing models and resources, incorporating AI activity into carbon monitoring, and encouraging green AI practices are crucial steps in promoting sustainability. 1. Prioritize Fine-Tuning: Instead of training new generative models from scratch, focus on fine-tuning existing models for specific content domains. Fine-tuning consumes less energy and provides more value to businesses. 2. Explore Energy-Conserving Methods: Adopt energy-conserving computational approaches like TinyML for processing data. TinyML allows running ML models on low-powered edge devices, significantly reducing energy consumption. 3. Re-use and Open Source Models: Opt for reusing open-source models instead of creating new ones. Recycling tech can lower the carbon impact of AI practices and reduce the need for energy-intensive model development. 4. Monitor Carbon Emissions: Include AI activity in carbon monitoring practices to understand the carbon footprint of AI-related operations. Share footprint numbers to make informed decisions about AI partnerships. 5. Choose Green Energy Sources: Select cloud providers and data centers that prioritize environmentally friendly power resources. Running AI models in regions with carbon-free energy sources can significantly reduce operational emissions. Have you already considered the impact of using compute-heavy applications on our planet? Are you tracking the impact of compute in your sustainability report? #genai #aivalue #sustainableai #sustainability