Key Points
- AI exoskeleton reduces therapist requirement from four to two per patient session
- System has accumulated over 40,000 patient training records across 22 countries
- Technology deployed in one-third of Taiwan’s medical-centre-level hospitals
Taiwanese medical device manufacturer Onyx Healthcare has partnered with Intel and robotics firm FREE Bionics to deploy an artificial intelligence-powered exoskeleton that helps patients with lower-limb injuries regain the ability to walk, the companies announced at Computex Taiwan.
The FREE Walk system — a wearable robotic frame worn by patients during rehabilitation — uses Intel’s Core Ultra processors and OpenVINO software to guide repetitive limb movements that retrain the brain’s neural pathways. The technology has been deployed in approximately one-third of medical-centre-level hospitals in Taiwan and has accumulated more than 40,000 patient training records across 22 countries, according to Onyx Healthcare.
EVENT
Infosec Reimagined
Infosec Reimagined 2026 is the premier information security summit where top leaders—CISOs, CROs, CIOs, CTOs and risk executives—converge to redefine cyber resilience.
EVENT
Digital Senate
Digital Senate is a premier conference uniting government leaders, technologists and innovators to share ideas, success stories and strategies on digital governance, public sector transformation, cybersecurity and emerging technologies in India.
EVENT
CIO Prism
CIO Prism unites forward-thinking technology leaders to exchange transformative insights, shape digital strategies, and foster innovation, empowering enterprises to excel in an era of rapid technological change.
How the AI exoskeleton works
Patients with lower-limb injuries typically require intensive clinical rehabilitation involving hours of expert-supervised physiotherapy. Traditional methods demand significant manual effort — a single session may require up to four therapists to assist one patient through movements designed to promote neuroplasticity, the brain’s ability to form new neural connections and relearn movement patterns.
The FREE Walk exoskeleton addresses these limitations by automating the guidance process. The device runs AI algorithms on Intel’s Core Ultra processor, which combines a central processing unit, graphics processing unit and neural processing unit — a specialised chip designed to accelerate AI calculations — to handle different computational tasks simultaneously.
“Over the last 16 years, we’ve been able to leverage Intel’s technology to transform the landscape of medical device care, which really sets the stage for widespread AI adoption in this field,” said John Chuang, president, Onyx Healthcare.
By the numbers
- 40,000+
- patient training records accumulated globally
- 50%
- reduction in therapists required per session
- 22
- countries where the system is deployed
The system integrates multiple sensors to monitor patient movement in real time. Pressure sensors track weight distribution, while electromyography sensors — devices that detect electrical signals generated when muscles contract — measure muscle activity. Inertial sensors monitor body position and orientation. Together, these components analyse the patient’s bio-electric muscle signals to predict and guide the next movement.
Edge processing reduces response delays
Onyx Healthcare uses Intel’s OpenVINO software toolkit to optimise how AI models run across the processor’s different computing units. This approach keeps time-sensitive processing on the device itself rather than sending data to remote servers, reducing latency — the delay between a patient’s movement and the system’s response — to levels that enable real-time interaction.
Advertisement
“Intel’s highly efficient, low-power NPU and OpenVINO software stack really transforms smart rehabilitation from a conceptual showcase on paper into a medical AI solution that is now being used across the world,” Chuang said.
The platform processes training data locally during sessions while accumulating information for cloud-based review and product improvement. This architecture balances immediate responsiveness with longer-term learning, according to Chuang.
Staffing efficiency improves by 50 per cent
The AI-powered exoskeleton reduces the number of therapists required per session from four to two, improving staffing efficiency by approximately 50 per cent, the company said. This addresses a significant constraint in rehabilitation medicine, where specialist physiotherapists are in limited supply and patient sessions can last several hours.
The technology has expanded beyond Taiwan to medical facilities in 22 countries, though Onyx Healthcare did not specify which markets or the total number of deployed units.
“We’re excited to continue our collaboration with Intel to bring more advanced AI technologies into reliable, safe, and long-term deployable medical-grade solutions that operate effectively in real clinical environments,” Chuang said.
Onyx Healthcare, founded in 2010, specialises in medical-grade computing hardware and has worked with Intel for over 16 years on healthcare technology applications.
Your Questions, Answered
How does the AI exoskeleton help patients walk again?
The FREE Walk exoskeleton uses Intel processors and sensors to guide repetitive limb movements during rehabilitation. This retrains the brain’s neural pathways through neuroplasticity, helping patients with lower-limb injuries gradually regain mobility.
How much does the AI exoskeleton reduce staffing requirements?
Traditional lower-limb rehabilitation requires up to four therapists per patient session. The AI-powered exoskeleton reduces this to two therapists, improving staffing efficiency by approximately 50 per cent.
Where is the FREE Walk exoskeleton currently deployed?
The system is used in approximately one-third of medical-centre-level hospitals in Taiwan and has been deployed across 22 other countries, accumulating over 40,000 patient training records.
What Intel technology powers the rehabilitation exoskeleton?
The exoskeleton runs on Intel Core Ultra processors combining CPU, GPU and NPU capabilities, along with Intel’s OpenVINO software toolkit that optimises AI processing for real-time response during patient movement.