Using MediaPipe to track upper-limb reaching movements after stroke: a proof-of-principle study
Bibliographic record
Abstract
Emerging work supports the use of artificial intelligence-based markerless motion capture systems to complement standardized clinical measures when assessing post-stroke motor recovery. MediaPipe Pose Landmarker is an open-sourced, machine learning tool, requiring only one camera, which can be used to track upper limb movements and quantify kinematics. Here we aimed to test the use-case of MediaPipe Pose Landmarker in tracking upper limb movements after stroke in a 2-dimensional cartesian coordinate space. Participants (N = 7, > 2 months after stroke, upper extremity portion of the Fugl-Meyer Assessment (FMA-UE) range of 40-66) engaged in five sessions of a previously established, semi-immersive, gamified reaching task, involving movements of the hand and arm. Movements at four time points (first and last block of session 1 and 5) were captured by a video camera, with videos processed through the MediaPipe Pose Landmarker pipeline to extract coordinates of effectors of interest and subsequently analyze kinematic outcomes (related to movements of the hand, shoulder, and trunk). Kinematics of the hand (mean palm speed, palm bivariate variable error; BVE, where a low BVE reflects greater consistency), shoulder (BVE), and trunk (BVE) were extracted for each individual, separately across time points. Exploratory analyses indicate increased mean palm speed and palm BVE across time points. Further, analyses suggest that shoulder and trunk movements may contribute to improvements in hand-related outcomes for some individuals. Overall, our findings provide support for the use of MediaPipe Pose Landmarker in tracking upper limb movements in individuals with motor impairment after stroke.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.003 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.004 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".