Feature selection for gesture recognition in Internet-of-Things for healthcare

TL;DR

This paper introduces a novel feature selection algorithm for gesture recognition in IoT healthcare, enhancing classification robustness and interpretability while reducing data traffic and preserving physiological relevance.

Contribution

It presents a new feature selection method combining consensus clustering and nested cross-validation for robust, meaningful gesture classification in IoT healthcare applications.

Findings

Improved classification accuracy for grasping tasks

Reduced data traffic in IoT communication

Selected features retain physiological significance

Abstract

Internet of Things is rapidly spreading across several fields, including healthcare, posing relevant questions related to communication capabilities, energy efficiency and sensors unobtrusiveness. Particularly, in the context of recognition of gestures, e.g., grasping of different objects, brain and muscular activity could be simultaneously recorded via EEG and EMG, respectively, and analyzed to identify the gesture that is being accomplished, and the quality of its performance. This paper proposes a new algorithm that aims (i) to robustly extract the most relevant features to classify different grasping tasks, and (ii) to retain the natural meaning of the selected features. This, in turn, gives the opportunity to simplify the recording setup to minimize the data traffic over the communication network, including Internet, and provide physiologically significant features for medical interpretation. The algorithm robustness is ensured both by consensus clustering as a feature selection strategy, and by nested cross-validation scheme to evaluate its classification performance.