Machine learning reveals sleeve pressure patterns to screen arm posture
What the Sleeve Feels: Explainable Machine Learning for Textile Pressure-Based Postural Screening
Artificial Intelligence
Summary
Posture can be monitored cheaply using smart clothes that sense pressure on the body. The authors studied a knitted sleeve that senses pressure on the forearm to classify arm positions into three categories. They designed easy-to-understand features from the pressure data and showed their method works well for people not seen before. Their approach also explains what parts of the pressure data influence the decision, and handles small changes when putting the sleeve on again.
What this means in practice
- •For wearable device designers: Design wearable smart sleeves that reliably detect arm posture using interpretable machine learning from pressure patterns.
- •For ergonomics teams: Screen workers’ forearm postures in real time to identify potentially harmful positions using pressure-sensing sleeves and explainable AI methods.
Authors
Limon Bin Hossain, Md Sadib Rahman Ananta
Abstract
Pressure-sensing smart textiles convert body-surface contact into a dense, image-like signal closely tied to posture and movement, making them a promising low-cost route to wearable posture screening. Realizing that promise, however, requires more than classification accuracy: a deployable system must generalize to wearers unseen during training, expose the physical evidence behind its decisions, and tolerate the small donning offsets that occur whenever a garment is removed and re-worn. This paper addresses these three requirements jointly using a knitted piezoresistive sleeve worn on the forearm as a testbed. We regroup fine-grained everyday activities into three coarser screening categories (neutral, potentially undesirable, and functional or transitional), engineer 29 interpretable pressure-distribution features spanning global intensity, spatial center of pressure, quadrant asymmetry, distribution complexity, and short-horizon temporal change, and evaluate under a strict subject-wise split. A tuned XGBoost classifier reaches 0.818 accuracy, 0.788 balanced accuracy, and 0.801 macro F1 on unseen test subjects, with tight frame-level bootstrap 95% intervals of about plus-minus 0.01 and a subject-to-subject standard deviation near 0.06 under leave-one-subject-out cross-validation. A simple 2D-CNN baseline trained on raw frames achieves broadly similar performance, showing that hand-engineered features are not left behind by a learned spatial representation on this task. SHAP-based explanation, a feature-group ablation, per-activity error analysis inside the pooled undesirable class, class-mapping sensitivity, and a simulated donning-rotation stress test together locate what the model relies on, where it degrades, and why, directly targeting the generalization, interpretability, and robustness gaps that determine whether such a system is deployable.