Scoliosis is the three-dimensional (3D) deformity of the spine and trunk. The majority or 80% of the scoliosis cases are idiopathic which means that the cause is unknown. The progression of scoliosis is much more frequently seen in females and severely scoliotic patients face a higher risk of different health problems in their adult life Bracing is the most commonly used non-invasive treatment for patients with smaller spinal curves which are usually between 21 and 40 degrees at Risser stage 0, 1 or 2. However, scoliosis braces are traditionally constructed based on the experience of orthotists. Since 3D spinal deformities are complex, the design and evaluation of braces are challenging with no consensus on an optimal brace design. AI and ML methods therefore constitute a new approach to address the difficulties in designing braces.
The designs of flexible brace include: an artificial hinge bone is used to stabilize the corrective components. Corrective straps are attached to the artificial hinge to exert corrective forces onto the scoliotic spine. Semi-rigid silicon pads with high conformability to the body are used to enhance the corrective forces. This study is a multi-disciplinary collaborative project with experts in computer science, orthopaedics, materials science, garment technology, engineering and biomechanics. A biomechanical computational model will be developed to recommend modifications to the flexible brace design. The effectiveness of the flexible braces will be evaluated through low-dose X-ray radiography at pre-, post-expert and post-ML designed treatment. A polynomial scoring system will be created to assess the therapeutic similarity of the ML-designed flexible brace to the expert-designed flexible brace based on an estimate of the corrective forces applied through the braces. Reinforcement learning with scores based on estimated pressure force will guide AI training towards superior therapeutic designs.
Study Type
INTERVENTIONAL
Allocation
NA
Purpose
TREATMENT
Masking
NONE
Enrollment
100
Participants will be invited to undergo a fitting session of a flexible brace. After the fitting session, participants will join a two-hours wear trial of the flexible brace. Participants are required to wear the brace for two-hours and undergo assessments before and after the wear trial. The measurement outcomes of the assessment include 1) X-ray scan, 2) 3D body scan, 3) garment pressure, and 4) questionnaire.
The Hong Kong Polytechnic University
Hong Kong, China
RECRUITINGIn-flexible brace Cobb angle measurements
1\. Number of Participants Maintaining Cobb Angle Stability (\<5° Increase) After Wearing the Brace for 2 Hours, as Measured by Spine X-ray (AP View).
Time frame: 2 hours after wearing the brace
Idea pressure level measurements
1\. The amount of pressure forces (kPa) exerted by the brace at minimum Cobb angles will be aggregated to ideal pressure level measurements (kPa/Cobb angles). Pressure forces will be measured by the Novel Pliance-xf-16 analyzer, while the Cobb angles will be measured by spine MRI.
Time frame: 2 hours after wearing the brace
Visual analogue scale (VAS) Scores
1. Change in Visual Analogue Scale (VAS) Scores for Pain Before and After Brace Usage. 2. Change in Visual Analogue Scale (VAS) Scores for Quality of Life Before and After Brace Usage. VAS is a straight horizontal line measuring 100 mm. The ends are defined as the extreme limits of pain, oriented from the left at 0 mm to the right at 100 mm. Typically, higher scores indicate a worse outcome, i.e., more pain.
Time frame: 2 hours after wearing the brace
This platform is for informational purposes only and does not constitute medical advice. Always consult a qualified healthcare professional.