Dynamic Gesture Recognition Using a Smart Glove in Hand-Assisted Laparoscopic Surgery

Abstract

This paper presents a system developed for the assistance with a collaborative robot in 12 hand-assisted laparoscopic surgery (HALS). The system includes a sensing glove with 13 piezoresistive sensors which capture continuously the flexion degree of the surgeon's fingers.14 These data are analyzed using an algorithm that detects and recognize the selected movements. 15This information is sent as commands to the collaborative robot throughout the surgical operation. 16The bending patterns, speed and execution times of the movements are modelled in a pre-phase in 17 which it will extract all the necessary information for later detection during the motion execution. 18The results obtained with 10 different volunteers show a high degree of accuracy and a low false 19 discovery rate. 20

Smart Glove and Hand Gesture-Based Control Interface for Multi-Rotor Aerial Vehicles in a Multi-Subject Environment

“…The proposed smart glove converts hand gestures to a set of pre-defined commands [26], which maps the fingers' dynamic movements to a set of pre-defined surgery robot actions. We use the flex sensors' output values to determine the main subject command to control a multi-rotor aerial vehicle.…”

One for All: An End-to-End Compact Solution for Hand Gesture Recognition

“…The HGR methods can be divided into two broad categories: sensor and vision-based techniques. Sensor based techniques [5], [6] used gloves and other electronic devices to measure the joint angles position of the fingers, position of the hands to extract the features of hands. Although, glove-based techniques have sufficient cues to identify hand gestures, but gloves with wires and sensors are too expensive and makes people uncomfortable to wear.…”

Hybrid-Flexible Bimodal Sensing Wearable Glove System for Complex Hand Gesture Recognition

“…Hence, the total number of existing recognizable unique gestures is limited in practice. [9][10][11]18,24,36 Contrary to adding more sensors, carefully designed bimodal sensors integrated with the appropriate machine learning (ML) models are capable of extracting unique embedded information on a deeper level that enables recognition beyond static gestures. 10,11,15,18,24,36−38 Here, we developed a hybrid-flexible wearable system that detects and recognizes hand gestures using an in-house low power code division multiple access (CDMA)-based multichannel interfacing circuit in real-time.…”