An official website of the United States government
Sitting is the most common status of modern human beings. Some sitting postures may bring health issues. To prevent the harm from bad sitting postures, a local sitting posture recognition system is desired with low power consumption and low computing overhead. The system should also provide good user experience with accuracy and privacy. This paper reports a novel posture recognition system on an office chair that can categorize seven different health-related sitting postures. The system uses six flex sensors, an Analog to Digital Converter (ADC) board and a Machine Learning algorithm of a two-layer Artificial Neural Network (ANN) implemented on a Spartan-6 Field Programmable Gate Array (FPGA). The system achieves 97.78% accuracy with a floating-point evaluation and 97.43% accuracy with the 9-bit fixed-point implementation. The ADC control logic and the ANN are constructed with a maximum propagation delay of 8.714 ns. The dynamic power consumption is 7.35 mW when the sampling rate is 5 Sample/second with the clock frequency of 5 MHz.
more »
« less
FPGA-based smart chair recognition system using flex sensors
Sitting is the most common status of modern human beings. Many people are sitting with bad posture which may lead to postural pain. Especially for people with short term-disabilities, sitting in the right posture is very important. In this research, we propose a posture recognition system on an office chair that can categorize different health-related sitting postures to prevent harm from bad sitting postures. The smart chair system consists of an array of five flex sensors integrated into an FPGA board. The output of the system is the classification result of the sitting posture. In this paper, several health-related sitting postures are selected. The sitting postures are: 1-sit straight; 2-left recline; 3-right recline; 4-lounge;5-lean backward; 6-cross left leg; 7-cross right leg. The proposed reconfigurable framework will be integrated as part of smart ambient assisted living with user-centered health monitoring system.
more »
« less
- Award ID(s):
- 1924117
- PAR ID:
- 10351827
- Date Published:
- Journal Name:
- IEEE 15th Dallas Circuit And System Conference (DCAS)
- Page Range / eLocation ID:
- 1 to 2
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
null (Ed.)A popular posture for using wireless technology is reclined sitting, with the trunk rotated posteriorly to the hips. This position decreases the head’s gravitational moment; however, the head angle relative to the trunk is similar to that of upright sitting when using a tablet in the lap. This study compared cervical extensor musculotendon length changes from neutral among 3 common sitting postures and maximum neck flexion while using a tablet. Twenty-one participants had radiographs taken in neutral, full-flexion, and upright, semireclined, and reclined postures with a tablet in their lap. A biomechanical model was used to calculate subject-specific normalized musculotendon lengths for 27 cervical musculotendon segments. The lower cervical spine was more flexed during reclined sitting, but the skull was more flexed during upright sitting. Normalized musculotendon length increased in the reclined compared with an upright sitting position for the C4-C6/7 (deep) and C2-C6/7 (superficial) multifidi, semispinalis cervicis (C2-C7), and splenius capitis (Skull-C7). The suboccipital ( R 2 = .19–.71) and semispinalis capitis segment length changes were significantly correlated with the Skull-C1 angle (0.24–0.51). A semireclined reading position may be an ideal sitting posture to reduce the head’s gravitational moment arm without overstretching the assessed muscles.more » « less
-
Monitoring posture and movement accurately and efficiently is essential for both physical therapy and athletic training evaluation and interventions. Motion Tape (MT), a self-adhesive wearable skin-strain sensor made of piezoresistive graphene nanosheets (GNS), has demonstrated promise in capturing low back posture and movements. However, to address some of its limitations, this work explores alternative materials by replacing GNS with multi-walled carbon nanotubes (MWCNT). This study aimed to characterize the electromechanical properties of MWCNT-based MT. Cyclic load tests for different peak tensile strains ranging from 1% to 10% were performed on MWCNT-MT made with an aqueous ink of 2% MWCNT. Additional tests to examine load rate sensitivity and fatigue were also conducted. After characterizing the properties of MWCNT-MT, a human subject study with 10 participants was designed to test its ability to capture different postures and movements. Sets of six sensors were made from each material (GNS and MWCNT) and applied in pairs at three levels along each side of the lumbar spine. To record movement of the lower back, all participants performed forward flexion, left and right bending, and left and right rotation movements. The results showed that MWCNT-MT exceeded GNS-MT with respect to consistency of signal stability even when strain limits were surpassed. In addition, both types of MT could assess lower back movements.more » « less
-
Reaching movements performed from a crouched body posture require a shift of body weight from both arms to one arm. This situation has remained unexamined despite the analogous load requirements during step initiation and the many studies of reaching from a seated or standing posture. To determine whether the body weight shift involves anticipatory or exclusively reactive control, we obtained force plate records, hand kinematics, and arm muscle activity from 11 healthy right-handed participants. They performed reaching movements with their left and right arm in two speed contexts, “comfortable” and “as fast as possible,” and two postural contexts, a less stable knees-together posture and a more stable knees-apart posture. Weight-shifts involved anticipatory postural actions (APAs) by the reaching and stance arms that were opposing in the vertical axis and aligned in the side-to-side axis similar to APAs by the legs for step initiation. Weight-shift APAs were correlated in time and magnitude, present in both speed contexts, more vigorous with the knees placed together, and similar when reaching with the dominant and nondominant arm. The initial weight-shift was preceded by bursts of muscle activity in the shoulder and elbow extensors (posterior deltoid and triceps lateral) of the reach arm and shoulder flexor (pectoralis major) of the stance arm, which indicates their causal role; leg muscles may have indirectly contributed but were not recorded. The strong functional similarity of weight-shift APAs during crouched reaching to human stepping and cat reaching suggests that they are a core feature of posture-movement coordination. NEW & NOTEWORTHY This work demonstrates that reaching from a crouched posture is preceded by bimanual anticipatory postural adjustments (APAs) that shift the body weight to the stance limb. Weight-shift APAs are more robust in an unstable body posture (knees together) and involve the shoulder and elbow extensors of the reach arm and shoulder flexor of the stance arm. This pattern mirrors the forelimb coordination of cats reaching and humans initiating a step.more » « less
-
Sleep is a vital physiological state that significantly impacts overall health. Continuous monitoring of sleep posture, heart rate, respiratory rate, and body movement is crucial for diagnosing and managing sleep disorders. Current monitoring solutions often disrupt natural sleep due to discomfort or raise privacy and instrumentation concerns. We introduce PillowSense, a fabric-based sleep monitoring system seamlessly integrated into a pillowcase. PillowSense utilizes a dual-layer fabric design. The top layer comprises conductive fabrics for sensing electrocardiogram (ECG) and surface electromyogram (sEMG), while the bottom layer features pressure-sensitive fabrics to monitor sleep location and movement. The system processes ECG and sEMG signals sequentially to infer multiple sleep variables and incorporates an adversarial neural network to enhance posture classification accuracy. We fabricate prototypes using off-the-shelf hardware and conduct both lab-based and in-the-wild longitudinal user studies to evaluate the system's effectiveness. Across 151 nights and 912.2 hours of real-world sleep data, the system achieves an F1 score of 88% for classifying seven sleep postures, and clinically-acceptable accuracy in vital sign monitoring. PillowSense's comfort, washability, and robustness in multi-user scenarios underscore its potential for unobtrusive, large-scale sleep monitoring.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- https://doi.org/10.1109/DCAS53974.2022.9845620
