RESUMO
The recognition of human activities (HAR) using wearable device data, such as smartwatches, has gained significant attention in the field of computer science due to its potential to provide insights into individuals' daily activities. This article aims to conduct a comparative study of deep learning techniques for recognizing activities of daily living (ADL). A mapping of HAR techniques was performed, and three techniques were selected for evaluation, along with a dataset. Experiments were conducted using the selected techniques to assess their performance in ADL recognition, employing standardized evaluation metrics, such as accuracy, precision, recall, and F1-score. Among the evaluated techniques, the DeepConvLSTM architecture, consisting of recurrent convolutional layers and a single LSTM layer, achieved the most promising results. These findings suggest that software applications utilizing this architecture can assist smartwatch users in understanding their movement routines more quickly and accurately.
Assuntos
Atividades Cotidianas , Aprendizado Profundo , Humanos , Reconhecimento Psicológico , Benchmarking , MovimentoRESUMO
Motion analysis is an area with several applications for health, sports, and entertainment. The high cost of state-of-the-art equipment in the health field makes it unfeasible to apply this technique in the clinics' routines. In this vein, RGB-D and RGB equipment, which have joint tracking tools, are tested with portable and low-cost solutions to enable computational motion analysis. The recent release of Google MediaPipe, a joint inference tracking technique that uses conventional RGB cameras, can be considered a milestone due to its ability to estimate depth coordinates in planar images. In light of this, this work aims to evaluate the measurement of angular variation from RGB-D and RGB sensor data against the Qualisys Tracking Manager gold standard. A total of 60 recordings were performed for each upper and lower limb movement in two different position configurations concerning the sensors. Google's MediaPipe usage obtained close results compared to Kinect V2 sensor in the inherent aspects of absolute error, RMS, and correlation to the gold standard, presenting lower dispersion values and error metrics, which is more positive. In the comparison with equipment commonly used in physical evaluations, MediaPipe had an error within the error range of short- and long-arm goniometers.
Assuntos
Movimento , Esportes , Fenômenos Biomecânicos , Movimento (Física) , BenchmarkingRESUMO
BACKGROUND: Interactive systems are being developed with the intention to help in the engagement of patients on various therapies. Amid the recent technological advances, Kinect™ from Microsoft (Redmond, WA) has helped pave the way on how user interaction technology facilitates and complements many clinical applications. In order to examine the actual status of Kinect developments for rehabilitation, this article presents a systematic review of articles that involve interactive, evaluative, and technical advances related to motor rehabilitation. MATERIALS AND METHODS: Systematic research was performed in the IEEE Xplore and PubMed databases using the key word combination "Kinect AND rehabilitation" with the following inclusion criteria: (1) English language, (2) page number >4, (3) Kinect system for assistive interaction or clinical evaluation, or (4) Kinect system for improvement or evaluation of the sensor tracking or movement recognition. Quality assessment was performed by QualSyst standards. RESULTS: In total, 109 articles were found in the database research, from which 31 were included in the review: 13 were focused on the development of assistive systems for rehabilitation, 3 in evaluation, 3 in the applicability category, 7 on validation of Kinect anatomic and clinical evaluation, and 5 on improvement techniques. Quality analysis of all included articles is also presented with their respective QualSyst checklist scores. CONCLUSIONS: Research and development possibilities and future works with the Kinect for rehabilitation application are extensive. Methodological improvements when performing studies on this area need to be further investigated.
Assuntos
Terapia por Exercício/métodos , Interface Usuário-Computador , Jogos de Vídeo , Humanos , Destreza Motora/fisiologia , Movimento , Recuperação de Função Fisiológica , Jogos de Vídeo/psicologiaRESUMO
AIM: This research aims to analyze the acute effect of incremental inspiratory loads on respiratory pattern and on the predominant activity frequency of inspiratory muscle, taking into account differences in gender responses. Optoelectronic Plethysmography was performed during loads in 39 healthy subjects (20 women), placing 89 markers on the thoracic-abdominal wall to obtain total and regional volumes. Surface electromyography (SEMG) was taken simultaneously on the Sternocleidomastoid and Diaphragm muscles, to calculate the predominant muscle activity frequency through wavelet analysis. Inspiratory loads were performed using Threshold(®)with 2 min of breathing at different levels, ranging from a load of 10 cmH(2)O plus 5 cmH(2)O to 40 cmH(2)O or fatigue. RESULTS: Inspiratory Time increased during loads. Total and compartmental volumes increased with different regions, changing at different loads. These changes in volume occur earlier in women (20 cmH(2)O) than in men (30 cmH(2)O). The predominant activity frequency of Sternocleidmastoid muscle decreased at 30 cmH(2)O, while Diaphragm activity decreased at 40 cmH(2)O. CONCLUSION: The acute effects of incremental inspiratory loads are increases of total and regional volumes and inspiratory time. As for muscle activity, the predominant activity frequency declined in Sternocleidomastoid and Diaphragm muscles, but at different loads. Such respiratory and SEMG patterns and gender differences should be considered when clinical interventions are performed.