Rehabilitation Engineering

Rehabilitation engineering applies engineering principles and technologies to restore, reinforce or compensate for lost physical functions. It encompasses the design of assistive devices, therapeutic robotics, wearable sensors and human–machine interfaces to support patients with neurological, musculoskeletal or traumatic injuries. By integrating biomechanics, materials science, control theory and clinical feedback, rehabilitation engineers develop systems that deliver intensive, task‐specific practice, augment residual motor capacity and enable objective monitoring of progress. Key areas include robotic exoskeletons and end‐effector devices for gait and upper‐limb training, adaptive prosthetic sockets and neuroprosthetic interfaces that translate neural or muscular signals into movement. Advances in low‐profile actuators, soft robotics and embedded sensing have led to lighter, more compliant devices that can be worn during daily activities, bridging the gap between clinic and home. Ultimately, rehabilitation engineering aims to enhance independence, accelerate recovery and improve quality of life across diverse patient populations.

Research from Nature Portfolio

Continuous neural control of a bionic limb restores biomimetic gait after amputation: Recent work has shown that surgically linked agonist–antagonist muscle constructs in a bionic ankle–foot prosthesis can augment residual muscle afferents by nearly 20 per cent of intact levels. In a cohort of below‐knee amputees, this interface enabled volitional and reflexive control, yielding near‐natural walking speeds on slopes and stairs and restoring adaptive gait modulation across varied terrains.

Shaping high-performance wearable robots for human motor and sensory reconstruction and enhancement: A perspective argues for multi-modal fusion of environmental, physiological and device-derived data through human-in-the-loop control. By embedding neuromuscular interfaces in flexible electronics and employing biomechatronic chips for real-time signal processing, next-generation exoskeletons can dynamically adapt assistance to intent, improving embodiment and facilitating motor recovery and sensory feedback.

Personalising exoskeleton assistance while walking in the real world: A portable ankle exoskeleton was optimised outdoors using wearable sensors and a data-driven method that derived assistance parameters within one hour of naturalistic walking. Real-world optimisation increased self-selected speed by 9 ± 4 per cent and reduced energy cost by 17 ± 5 per cent, matching the benefits of laboratory tuning four times faster and demonstrating feasibility of field-based personalisation.

Research from all publishers

A data-driven design framework for structural optimisation to enhance wearing adaptability of prosthetic hands: A multi-index optimisation method combined pressure and motion data with clinical pain thresholds from forearm amputees. By adjusting regional socket contours via additive manufacturing, adapted sockets yielded 19 per cent reduced muscle effort and up to 99 per cent lower fatigue rates, highlighting the role of empirical sensor data in personalised socket geometry.

Adaptive gait training of a lower limb rehabilitation robot based on human–robot interaction force measurement: An active–passive hybrid exoskeleton equipped with multi-axis force sensors measured interaction torques and estimated intended joint moments. A dynamic model used sensed forces to modulate gait trajectories, aligning assistance with user effort. Clinical tests revealed enhanced participation, safety and the ability to tailor support to residual strength, promoting volitional engagement.

Artificial intelligence-driven virtual rehabilitation for people living in the community: A scoping review identified AI methods—fuzzy rule-based systems, template matching and deep neural networks—applied to home-based virtual therapy for post-treatment stroke survivors. AI analysed multi-modal sensor data to assess exercise quality and deliver automated feedback. Early studies suggest AI-augmented programmes can improve motor and cognitive outcomes, though standardisation and larger trials are needed.

Rehabilitation Engineering publication trend

The graph below shows the total number of articles in rehabilitation engineering across all publications each year (not limited to Nature Index journals).

Technical terms

Exoskeleton: A wearable, articulated frame that supports or augments limb movement through powered joints aligned with human anatomy.

Exosuit: A soft, garment-like wearable robot employing textile-based actuators and cables to assist movement with minimal rigid structure.

Assistance-as-needed: A control strategy that provides robotic support only when a user cannot complete a movement independently, reducing slacking and promoting active engagement.

Human-in-the-loop control: A real-time adaptive framework that continuously adjusts device assistance based on user inputs or physiological feedback.

Neuromuscular interface: A system that records neural or muscular signals and translates them into commands for assistive or therapeutic devices.

References

1. Continuous neural control of a bionic limb restores biomimetic gait after amputation. Nature Medicine (2024).
2. Shaping high-performance wearable robots for human motor and sensory reconstruction and enhancement. Nature Communications (2024).
3. Personalizing exoskeleton assistance while walking in the real world. Nature (2022).
4. A Data-Driven Design Framework for Structural Optimization to Enhance Wearing Adaptability of Prosthetic Hands. IEEE Transactions on Neural Systems and Rehabilitation Engineering (2024).
5. Adaptive Gait Training of a Lower Limb Rehabilitation Robot Based on Human–Robot Interaction Force Measurement. Cyborg and Bionic Systems (2024).
6. Artificial intelligence-driven virtual rehabilitation for people living in the community: A scoping review. npj Digital Medicine (2024).