Assessment of change in drug-induced parkinsonism in daily-life gait using vision-based machine learning
6 affiliations
- 1Department of Engineering Systems and Services, Delft University of Technology, Delft, the Netherlands
- 2Institute for Health Systems Science, Delft University of Technology, Delft, the Netherlands
- 3KITE Research Institute, University Health Network, Toronto, Ontario, Canada
- 4Vector Institute, Toronto, Ontario, Canada
- 5Department of Computer Science, University of Toronto, Toronto, Ontario, Canada
- 6Institute of Biomedical Engineering, University of Toronto, Toronto, Ontario, Canada
Abstract
Drug-induced parkinsonism is a potentially reversible movement disorder that may be subtle and difficult to monitor in residential dementia care. Vision-based machine learning analysis of gait may provide unobtrusive longitudinal monitoring around medication changes. We conducted a two-center cohort study of 61 residents with dementia receiving psychotropic medication. Hallway gait sequences were recorded using ceiling-mounted cameras and converted to three-dimensional body pose. A TabPFN classifier estimated UPDRS-III item 29 gait severity for each walk. We evaluated associations between acute antipsychotic (AP) and antidepressant (AD) dose reduction/cessation (≥50% dose reduction or cessation, 0–21 days after change) and daily mean predicted gait severity using linear mixed-effects models in ever-exposed participants, with complementary equal-weight paired analyses among participants observed in both medication states. In mixed-effects models, acute AP dose reduction/cessation was associated with lower predicted gait severity (β=−0.127, p = 0.022), whereas AD dose reduction/cessation was not (β=−0.060, p = 0.210). Equal-weight paired analyses found no statistically significant participant-average within-person change among paired participants (AP mean change = 0.031, 95% CI [−0.114, 0.176], p = 0.653; AD mean change =−0.021, 95% CI [−0.098, 0.056], p = 0.575), consistent with heterogeneous individual responses. Vision-based monitoring captured within-participant gait variation around psychotropic dose-reduction periods, but findings are associative. Larger prospective studies with documented reasons for medication changes are needed before using such monitoring to guide medication decisions. Author summary: Drug-induced parkinsonism is a movement disorder that can occur when older adults take medications used for psychiatric or behavioral symptoms. It can be difficult to detect because symptoms may be subtle and develop gradually, while clinical assessments are often infrequent. We studied whether a camera-based monitoring system installed in residential-care hallways could help track walking changes after medication dose reductions. The system turns short walking videos into body-pose measurements and uses these measurements to estimate gait severity. In 61 residents with dementia, we found that antipsychotic dose reduction was linked with lower estimated parkinsonism gait severity in a participant-day mixed-effects analysis. However, when we gave each paired participant one equal-weight contrast among participants observed both on medication and after a dose reduction/cessation event, we did not find a clear average improvement, and individual responses varied substantially. Antidepressant dose reduction showed no consistent pattern. Our findings suggest that unobtrusive gait monitoring may help identify people whose walking changes around medication adjustments, but it should not be used alone to guide medication decisions.
Related
- Antipsychotic-Induced Parkinsonism With Concurrent Benztropine-Associated Anticholinergic Effects in a Community Emergency Department Setting: A Case Report
- Evaluating Laryngeal Stability in Drug-Induced Parkinsonism Using Hilbert–Huang Transform and Chaos Dynamics: A Multi-Vowel Nonlinear Acoustic Framework for Differential Diagnosis
- Predicting subsequent-year physical fitness assessment failure in university students using longitudinal data and explainable machine learning
- A longitudinal study of rotenone-induced Parkinsonism in mice
- Application of spectroscopic and multispectral imaging technologies on the assessment of ready-to-eat pineapple quality: A performance evaluation study of machine learning models generated from two commercial data analytics tools
BibTeX
@article{filtjens2026assessment,
title = {{Assessment of change in drug-induced parkinsonism in daily-life gait using vision-based machine learning}},
author = {Benjamin Filtjens and Vida Adeli and Soroush Mehraban and Andrea Sabo and Babak Taati and Andrea Iaboni},
journal = {PLOS Digital Health},
year = {2026},
month = oct,
doi = {10.1371/journal.pdig.0001780},
url = {https://journals.plos.org/digitalhealth/article?id=10.1371/journal.pdig.0001780},
note = {CC BY},
}
RIS
TY - JOUR
TI - Assessment of change in drug-induced parkinsonism in daily-life gait using vision-based machine learning
AU - Benjamin Filtjens
AU - Vida Adeli
AU - Soroush Mehraban
AU - Andrea Sabo
AU - Babak Taati
AU - Andrea Iaboni
PY - 2026
DA - 2026/10/01
JO - PLOS Digital Health
DO - 10.1371/journal.pdig.0001780
UR - https://journals.plos.org/digitalhealth/article?id=10.1371/journal.pdig.0001780
AB - Drug-induced parkinsonism is a potentially reversible movement disorder that may be subtle and difficult to monitor in residential dementia care. Vision-based machine learning analysis of gait may provide unobtrusive longitudinal monitoring around medication changes. We conducted a two-center cohort study of 61 residents with dementia receiving psychotropic medication. Hallway gait sequences were recorded using ceiling-mounted cameras and converted to three-dimensional body pose. A TabPFN classifier estimated UPDRS-III item 29 gait severity for each walk. We evaluated associations between acute antipsychotic (AP) and antidepressant (AD) dose reduction/cessation (≥50% dose reduction or cessation, 0–21 days after change) and daily mean predicted gait severity using linear mixed-effects models in ever-exposed participants, with complementary equal-weight paired analyses among participants observed in both medication states. In mixed-effects models, acute AP dose reduction/cessation was associated with lower predicted gait severity (β=−0.127, p = 0.022), whereas AD dose reduction/cessation was not (β=−0.060, p = 0.210). Equal-weight paired analyses found no statistically significant participant-average within-person change among paired participants (AP mean change = 0.031, 95% CI [−0.114, 0.176], p = 0.653; AD mean change =−0.021, 95% CI [−0.098, 0.056], p = 0.575), consistent with heterogeneous individual responses. Vision-based monitoring captured within-participant gait variation around psychotropic dose-reduction periods, but findings are associative. Larger prospective studies with documented reasons for medication changes are needed before using such monitoring to guide medication decisions. Author summary: Drug-induced parkinsonism is a movement disorder that can occur when older adults take medications used for psychiatric or behavioral symptoms. It can be difficult to detect because symptoms may be subtle and develop gradually, while clinical assessments are often infrequent. We studied whether a camera-based monitoring system installed in residential-care hallways could help track walking changes after medication dose reductions. The system turns short walking videos into body-pose measurements and uses these measurements to estimate gait severity. In 61 residents with dementia, we found that antipsychotic dose reduction was linked with lower estimated parkinsonism gait severity in a participant-day mixed-effects analysis. However, when we gave each paired participant one equal-weight contrast among participants observed both on medication and after a dose reduction/cessation event, we did not find a clear average improvement, and individual responses varied substantially. Antidepressant dose reduction showed no consistent pattern. Our findings suggest that unobtrusive gait monitoring may help identify people whose walking changes around medication adjustments, but it should not be used alone to guide medication decisions.
SN - 2767-3170
ER -