What does this research mean for the field?

Integrating serious game-based cognitive assessments with wearable sensor-derived upper-limb motor measures provides complementary information that can effectively distinguish between mild and moderate dementia severity. Novelty: ClaimNovelty.INCREMENTAL. Consensus alignment: ConsensusAlignment.NEUTRAL.

What question did this study set out to answer?

To examine whether combining serious game-based cognitive tasks with wearable sensor-derived upper-limb motor assessments can accurately distinguish mild from moderate dementia in community-dwelling older adults.

September 8, 2026BMC GeriatricsOpen Access

Multimodal cognitive and upper-limb motor markers for distinguishing mild and moderate dementia in older adults

View Full Paper

Ask AI

Bookmark

Authors

TMTso‐Yen MaoChaoyang University of Technology HLHau-Tung Lai

Discussion

Member takes

Implication

Cross-sectional study reveals that combined cognitive and upper-limb motor markers distinguish mild from moderate dementia in older adults, suggesting scalable tools for community-based screening.

Key Points

To examine whether combining serious game-based cognitive tasks with wearable sensor-derived upper-limb motor assessments can accurately distinguish mild from moderate dementia in community-dwelling older adults.
Assessed 103 community-dwelling older adults in Taiwan using four game-based cognitive tasks (memory, attention, reaction speed, logical reasoning) and wearable sensor-derived upper-limb motor metrics in a cross-sectional design.
Evaluated group differences using Welch’s t-tests with false discovery rate (FDR) correction, built Random Forest classifiers using SMOTE class balancing with repeated stratified five-fold cross-validation, and conducted fuzzy-set qualitative comparative analysis (fsQCA).
Participants with mild dementia demonstrated significantly higher performance than those with moderate dementia across three cognitive tasks and multiple upper-limb motor features following FDR correction.
Random Forest models trained on motor-only and multimodal feature sets achieved equivalent discriminatory performance with an AUROC of 0.700.
fsQCA identified a specific cognitive-motor profile consisting of high TAKA composite performance and a high left shoulder abduction index associated with mild dementia status (consistency = 0.92).

Cite This Study

Mao et al. (2026) studied this question.

synapsesocial.com/papers/6a9fd82058e84d0ff5b47613 https://doi.org/10.1186/s12877-026-08249-1

View Full Paper

Ask AI

Bookmark

Implication

Cross-sectional study reveals that combined cognitive and upper-limb motor markers distinguish mild from moderate dementia in older adults, suggesting scalable tools for community-based screening.

Multimodal cognitive and upper-limb motor markers for distinguishing mild and moderate dementia in older adults

Discussion

Member takes

Implication

Key Points

Cite This Study

Multimodal cognitive and upper-limb motor markers for distinguishing mild and moderate dementia in older adults

Discussion

Member takes

Implication

Key Points

Cite This Study

Also Consider

Also Consider