AI in Rehabilitation and Physical Therapy

Excited to share some recent work from the CMU MetaMobility Lab! This was presented at ICORR Consortium during RehabWeek and it's the first in a series of projects we've been working on this past year. We explored how computer vision (CV) can be leveraged to personalize exoskeleton control. Traditionally, control strategies rely on analytical models or deep learning to interpret user motion or environmental context. But what if CV could further enhance this process? We believe it can! Here, we showed that kinematics extracted from CV can serve as a new ground truth to fine-tune the exoskeleton deep learning-based kinematics estimator. This adaptation only requires video data from 1~2 gait cycles, captured using a single RGB camera. The adapted model achieved: 1. 10% higher accuracy than the pre-trained model 2. 20% higher accuracy than a model trained from scratch using just the same short video snippet While this is a proof of concept, it opens up exciting possibilities: using just a smartphone to capture personalized motion data and fine-tune AI models that modulate exoskeleton assistance. We're excited about the potential of this direction! This work was led by my PhD student Changseob Song along with Bogdan Ivanyuk-Skulskiy Adrian Krieger Kaitao Luo Paper Link: https://lnkd.in/eJA_bj84 #WearableRobotics #Exoskeleton #ComputerVision #DeepLearning #PersonalizedMobility #MetaMobilityLab

Researchers have developed a noninvasive 🧠-spine interface that can detect when a person thinks about moving and use that signal to stimulate the spinal cord. By using EEG caps, they trained a decoder to distinguish between actual and imagined leg movements in volunteers without spinal cord injuries. The system was able to predict movement intention, even without physical motion, by focusing on neural activity alone. Overall, this approach could enable rehabilitation therapies where spinal stimulation is triggered by brain signals in patients with paralysis and restoring voluntary movement using noninvasive techniques. Learn more: https://lnkd.in/gt8jzxvz One love #brain #spine #invisible #disability

AI and the Future of Knee Care The knee, a crucial joint for our daily movement and weight-bearing, is benefiting immensely from advancements in AI. From diagnostics to rehabilitation, AI is redefining how we approach knee conditions. Here’s a closer look at AI’s impact on knee care: - Diagnostics and Imaging:  - Enhanced Imaging: AI algorithms now identify subtle structural changes in the knee, improving early diagnosis.  - Automated Assessments: Conditions like osteoarthritis are diagnosed more accurately with AI's ability to measure joint spaces and spot abnormalities. - Predictive Analytics:  - Risk Prediction: AI helps forecast the likelihood of knee problems, allowing for proactive measures and early intervention. - Surgical Planning and Assistance:  - Advanced Planning: AI generates precise 3D knee models for accurate surgical planning and employs virtual reality for practice.  - Robotic Surgery: AI-driven robots support precise surgical techniques, ensuring accurate cuts and minimal invasiveness. - Rehabilitation and Recovery:  - Custom Rehabilitation: AI creates personalised exercise regimens based on individual needs and recovery progress.  - Remote Monitoring: AI-powered tools enable remote physiotherapy and track patient progress, making timely treatment adjustments possible. AI is paving the way for more effective and personalised knee care solutions. How is AI improving your healthcare experience?  #AI #techinhealth #healthCare #Health

Significant break through to help patients of Spinal Cord Injuries integrating disciplines: neuroscience, bio medical engineering and Artificial intelligence. The ARC nerve stimulation therapy system from startup Onward Medical passed another developmental milestone as the company announced the first successful installation of its brainwave-driven implantable electrode array to restore function and feeling to a patient’s hands and arms. The news comes just five months after the researchers implanted a similar system in a different patient to help them regain a more natural walking gait. The ARC system used differs depending on how what issue it's being applied to. The ARC-EX is an external, non-invasive stimulator array that sits on the patient’s neck and helps regulate their bladder control and blood pressure, as well as improving limb function and control. Onward’s lower limb study from May employed the IM along with a BCI controller from CEA-Clinatec to create a “digital bridge” spanning the gap in the patient’s spinal column.   The study published Wednesday instead utilized the ARC-IM, an implantable version of the company’s stimulator array which is installed near the spinal cord and is controlled through wearable components and a smartwatch. Onward had previously used the IM system to enable paralyzed patients to stand and walk short distances without assistance, for which it was awarded an FDA Breakthrough Device Designation in 2020. Medical professionals led by by neurosurgeon Dr. Jocelyne Bloch, implanted the ARC-IM and the Clinatec BCI into a 46-year-old patient suffering from a C4 spinal injury, in mid-August. The BCI’s hair-thin leads pick up electrical signals in the patient’s brain, convert those analog signals into digital ones that machines can understand, and then transmits them to a nearby computing device where a machine learning AI interprets the patient’s electrical signals and issues commands to the implanted stimulator array. The patient thinks about what they want to do and these two devices work to translate that intent into computer-controlled movement.