Proposed Topic (Most preferred): :
Research and Innovations (new projects / technology / innovations / service models)
Authors (including presenting author) :
Chan WKV, Cheung H, Wong TY, Lee MH, Kwok KW, Lo SS, Sze MY
Affiliation :
Occupational Therapy Department, Pamela Youde Nethersole Eastern Hospital
Introduction :
Patients with different extents of cognitive impairment would receive trainings at occupational therapy department based on different level of care. Patients with Mild Neurocognitive Disorder (Mild NCD) & Mild Cognitive Impairment (MCI) usually demonstrated limitation in their capacity for functional mental tasks. Clinical observations and evaluations revealed that rehabilitation outcomes of these patients (HK-MoCA between 2nd- 16th percentiles) were proportionally related to the number of rehabilitation sessions, patient’s motivation, and timely adjustment to level of challenges in the prescribed treatments. Yet due to the imbalance between high patient demands and relatively low therapists’ supplies, patient may not always be able to see therapist timely for program/ training adjustments.
Literature Review & Program Development:
Research revealed that advancements of rehabilitation robots can enhance elderly’s memory, attention & concentration, and social functioning by reducing their negative psychophysiological feelings (Yuan et. al., Frontiers in Robotics and AI; 11 May 2021). Referencing the above, a pilot AI cognitive program incorporated prompting, cueing and automation in level adjustment was developed to evaluate whether application of AI in cognitive training/intervention can enhance training effects in older adults, ultimately improve their overall daily functioning.
Objectives :
To evaluate the effectiveness of integrating AI to enhance delivery of cognitive trainings prescribed by occupational therapists through high repeatability and automation
To provide diversion therapy services through sensory enrichment, positive social engagement and entertainment
To facilitate better compliance of home-based cognitive trainings
Methodology :
A convenience sample of 20 patients with HK-MoCA score between 2nd-16th percentiles meeting the following criteria were recruited:
Inclusion: able to read simple instructions in Chinese & Medically Stable
Exclusion: medically unstable & illiterate
Patients will be assigned to Group A (AI program) or Group B (conventional training) in alternate order, i.e., A-B-A-B...etc. Patients will undergo treatment programs twice per week for 4 weeks working on domains of Reality Orientation, Attention & Concentration, Memory & Learning, and Executive Function at each session which last for 30 minutes.
Result & Outcome :
SPSS 25 revealed no significant differences in baseline HK-MoCA overall scores and most cognitive sub-domain scores between the two groups except attention ( p=0.01). Significant improvements were noted in HK-MoCA overall scores (p=0.02), cognitive sub-domain of executive function (p=0.05), attention (p=0.02), and orientation (p=0.01) post treatment.
Conclusions & Way Forward:
Results suggested our hypothesis that integrating AI in occupational therapy cognitive training programs can augment training intensity and enhance therapeutic outcomes. Hence the program application will extend beyond OPD.
