Using AI Systems to Optimize the Clinical Outcome of Stroke Patients

ConditionsAcute Stroke Intervention Acute Stroke Acute Ischemic Stroke

Trial statusRecruiting

Trial phaseNot applicable

Trial typeInterventional

Biological sexAll

Age65-80

SponsorChinese University of Hong Kong

About this trial

This project addresses the imminent challenge of providing adequate motor rehabilitation to a growing number of stroke survivors amidst the ageing population, decreasing age of stroke, and shortage of physical/occupational therapists in Hong Kong through AI and precision rehabilitation. To reduce the socioeconomic burden from the stroke survivors' loss of independence and their care (\>HK$15 billion/year), the efficacy of rehabilitation and efficiency of its delivery must be improved. These goals can be achieved by prescribing them with individually tailored rehabilitations predicted to yield maximal functional return. Defining a predictive model for such personalization remains challenging given the immense heterogeneity of stroke. The investigators aim to build an explainable AI system that predicts a subject's recovery potential and the treatment option that may realize this potential based on multi-modal pre-rehab assessments. Data from clinical, neuroimaging, neurophysiological, and multi-omic evaluations will be collected from stroke survivors (N≥400) before they undergo upper limb rehab with usual care, neuromuscular stimulation, robotic training, or acupuncture. Machine learning-extracted data features will be used to train decision-tree and neural-network AI algorithms for robust predictions. As soon as the model is validated, the investigators will deploy it to implement a personalized rehab program in the community. Our model's ability to predict the optimal intervention from a wide spectrum of input modalities distinguishes ours from previous less-than-accurate models. Our interdisciplinary team of 13 PIs with expertise in neurology, PT/OT, acupuncture, electrical/biomed. engineering, robotics, neuroscience, neuroimaging, multi-omics, data science, and clinical trial management will put us in a world-unique position to execute this project successfully and generate opportunities of interdisciplinary education. In the long run, our prediction system will accelerate marketization of new rehab strategies by facilitating their clinical-trial evaluations in more targeted subjects, thereby leading Hong Kong to be a future global hub of innovative rehabilitation.

Eligibility criteria

This trial does not accept healthy volunteers

Qualifiers

Age 65-80

1-6 months after onset of a first-time unilateral stroke rostral to midbrain

Moderate-to-severe motor impairment of one upper limb (Fugl-Meyer Assessment for Upper Extremity of 10-50 out of 66);

Able to provide written informed consent;

Disqualifiers

Unconscious or bed-bound;

Uncontrollable diabetes;

Anticipated non-adherence to treatment schedule;

On cardiac pacemaker;

Trial design

Design model

Parallel

Treatments tested in this trial

Acupuncture
Other intervention
Patients will receive 12 weeks of acupuncture with 3 half-hour sessions weekly. Acupoints will include (1) a basic formula of 8 Bo's abdomen acupoints (paretic side, 0.5cm depth vertically) \[66\]; (2) 12 conventional acupoints (bilateral, opposite to paretic side first, 1-4 cm depth vertically) \[67\]; and (3) 3 scalp acupoints (opposite to paretic side, 0.5-cm depth at 15-30 degrees) \[68\]At most 3 additional supplementary acupoints will be included, depending on the clinician's professional judgement. Sterile needles will be used after skin disinfection. Manual rotating manipulation \[69\] will be performed every 10 minutes on the conventional acupoints to achieve De-qi sensation \[70\]. Abdomen and scalp acupoints do not require De-qi. Subjects will be monitored throughout the course of therapy with face-to-face assessments by blinded and trained clinicians after 0, 4, 8, 12 weeks of intervention. Clinical scales will also be recorded at 0 (A0), 8 (A½), and 12 weeks (A1).
control group
Other intervention
Subjects in this group will receive the typical post-stroke care offered to stroke survivors of Hong Kong. This care emphasizes restoring function and independence through a comprehensive approach. Physical therapy focuses on improving mobility with exercises and training; occupational therapy helps patients relearn essential daily living skills, facilitating a smooth transition back to everyday life. Speech therapy is integral for addressing communication challenges, and psychological support is provided to help patients manage the emotional impact of stroke. All subjects will receive the above care for 8 weeks (≥3 hourly sessions per week). Clinical scales will be recorded at 0 (A0), 4 (A½), and 8 weeks (A1).
Robotic Training
Device
For this treatment group, CMC (corticomuscular coherence)-EMG-triggered control will assist wrist extension with hand open and wrist flexion with hand close alternately by mechanical pneumatic support \[60\]. During wrist-hand extension, the pneumatic fingers will assist a hand-open motion with constant inflation till the inner pressure reaches 90kPa; during the wrist-hand flexion, the pneumatic fingers will deflate constantly to assist a hand-close motion. To trigger ENMS assistance, two criteria must be met: (1) the average EMGs of target muscles exceed a pre-defined threshold, and (2) a significant CMC peak value with peak frequency in the beta band is captured. Here, the target muscle groups will be the extensor digitorum and extensor carpi ulnaris (ED-ECU) and flexor digitorum and flexor carpi radialis (FD-FCR), and the EMG and CMC will be evaluated during sustained contraction of these muscles over a 3-sec window.
Neuromuscular Electrical Stimulation Group
Device
For this treatment group, constant NMES (70V, 40Hz, 0-300µs square wave bursts \[63\]) will be delivered to the ED-ECU and FD-FCR muscles to assist in wrist-hand extension and flexion, respectively. The pulse width of NMES will be individually adjusted to achieve the maximum muscle contraction with the minimum stimulation intensity. The control strategy of CMC-EMG-triggered NMES will be the same as that used in the robot.

Treatment groups

400 Participants

are divided into 4 treatment groups

Group A: Rehabilitation Control GroupActive comparator 1 intervention

Group B: Acupuncture GroupActive comparator 1 intervention

Group C: Robotic Training GroupActive comparator 1 intervention

Group D: NMES GroupActive comparator 1 intervention

Trial outcomes

Primary outcomes

Surface EMG Recordings

Surface EMG electrodes may be conveniently used to record myoelectric activities of multiple muscles during voluntary movement. This data can reveal the muscle coordination patterns utilized by the motor system for control. We have previously shown that muscle synergies, building blocks of muscle coordination decomposed from EMG using machine learning, may serve as markers of post-stroke functional assessment and prognosis. Here, we will record EMG from 16 affected-side muscles during 8-10 activities of daily living.

Time frame