Your browser doesn't support javascript.
loading
Mostrar: 20 | 50 | 100
Resultados 1 - 20 de 14.271
Filtrar
1.
Sensors (Basel) ; 21(21)2021 Oct 27.
Artigo em Inglês | MEDLINE | ID: mdl-34770422

RESUMO

Providing a stable, low-price, and safe supply of energy to end-users is a challenging task. The energy service providers are affected by several events such as weather, volatility, and special events. As such, the prediction of these events and having a time window for taking preventive measures are crucial for service providers. Electrical load forecasting can be modeled as a time series prediction problem. One solution is to capture spatial correlations, spatial-temporal relations, and time-dependency of such temporal networks in the time series. Previously, different machine learning methods have been used for time series prediction tasks; however, there is still a need for new research to improve the performance of short-term load forecasting models. In this article, we propose a novel deep learning model to predict electric load consumption using Dual-Stage Attention-Based Recurrent Neural Networks in which the attention mechanism is used in both encoder and decoder stages. The encoder attention layer identifies important features from the input vector, whereas the decoder attention layer is used to overcome the limitations of using a fixed context vector and provides a much longer memory capacity. The proposed model improves the performance for short-term load forecasting (STLF) in terms of the Mean Absolute Error (MAE) and Root Mean Squared Errors (RMSE) scores. To evaluate the predictive performance of the proposed model, the UCI household electric power consumption (HEPC) dataset has been used during the experiments. Experimental results demonstrate that the proposed approach outperforms the previously adopted techniques.


Assuntos
Aprendizado de Máquina , Redes Neurais de Computação , Previsões , Tempo , Tempo (Meteorologia)
2.
Sensors (Basel) ; 21(21)2021 Nov 08.
Artigo em Inglês | MEDLINE | ID: mdl-34770732

RESUMO

This paper proposes a post-processing method called bidirectional interpolation method for sampling-based path planning algorithms, such as rapidly-exploring random tree (RRT). The proposed algorithm applies interpolation to the path generated by the sampling-based path planning algorithm. In this study, the proposed algorithm is applied to the path created by RRT-connect and six environmental maps were used for the verification. It was visually and quantitatively confirmed that, in all maps, not only path lengths but also the piecewise linear shape were decreased compared to the path generated by RRT-connect. To check the proposed algorithm's performance, visibility graph, RRT-connect algorithm, Triangular-RRT-connect algorithm and post triangular processing of midpoint interpolation (PTPMI) were compared in various environmental maps through simulation. Based on these experimental results, the proposed algorithm shows similar planning time but shorter path length than previous RRT-like algorithms as well as RRT-like algorithms with PTPMI having a similar number of samples.


Assuntos
Robótica , Algoritmos , Simulação por Computador , Tempo
3.
Sensors (Basel) ; 21(22)2021 Nov 19.
Artigo em Inglês | MEDLINE | ID: mdl-34833791

RESUMO

An efficient energy operation strategy for the smart grid requires accurate day-ahead electricity load forecasts with high time resolutions, such as 15 or 30 min. Most high-time resolution electricity load prediction techniques deal with a single output prediction, so their ability to cope with sudden load changes is limited. Multistep-ahead forecasting addresses this problem, but conventional multistep-ahead prediction models suffer from deterioration in prediction performance as the prediction range is expanded. In this paper, we propose a novel two-stage multistep-ahead forecasting model that combines a single-output forecasting model and a multistep-ahead forecasting model to solve the aforementioned problem. In the first stage, we perform a single-output prediction based on recent electricity load data using a light gradient boosting machine with time-series cross-validation, and feed it to the second stage. In the second stage, we construct a multistep-ahead forecasting model that applies an attention mechanism to sequence-to-sequence bidirectional long short-term memory (S2S ATT-BiLSTM). Compared to the single S2S ATT-BiLSTM model, our proposed model achieved improvements of 3.23% and 4.92% in mean absolute percentage error and normalized root mean square error, respectively.


Assuntos
Eletricidade , Memória de Longo Prazo , Previsões , Tempo
4.
PLoS One ; 16(11): e0259282, 2021.
Artigo em Inglês | MEDLINE | ID: mdl-34731181

RESUMO

Infectious diseases and widespread outbreaks influence different sectors of the economy, including the stock market. In this article, we investigate the effect of EBOV and COVID-19 outbreaks on stock market indices. We employ time-varying and constant bivariate copula methods to measure the dependence structure between the infectious disease equity market volatility index (IEMV) and the stock market indices of several sectors. The results show that the financial and communication services sectors have the highest and the lowest negative dependency on IEMV during the Ebola virus (EBOV) pandemic, respectively. However, the health care and energy sectors have the highest and lowest negative dependency on IEMV during the COVID-19 outbreak, respectively. Therefore, the results confirm the heterogeneous time-varying dependency between infectious diseases and the stock market indices. The finding of our study contributes to the ongoing literature on the impact of disease outbreaks, especially the novel coronavirus outbreak on global large-cap companies in the stock market.


Assuntos
COVID-19/economia , Efeitos Psicossociais da Doença , Surtos de Doenças/economia , Doença pelo Vírus Ebola/economia , Comércio , Ebolavirus , Humanos , Tempo
5.
Beijing Da Xue Xue Bao Yi Xue Ban ; 53(5): 942-945, 2021 Oct 18.
Artigo em Chinês | MEDLINE | ID: mdl-34650299

RESUMO

OBJECTIVE: With the rapid development of sleep medicine, there are various methods for detecting sleep diseases. This study compared the correlation between the lightweight watch-type sleep monitor (Actiwatch) and the "gold standard" polysomnography (PSG) in the Chinese population, in order to provide a basis for clinical application. METHODS: From August 2018 to December 2019, 121 subjects who simultaneously performed sleep breathing monitoring (PSG) and wearing a watch-type sleep monitor (Actiwatch) in the Sleep Center of Peking University People's Hospital were enrolled. All subjects received PSG and Actiwatch at the same time, and filled out the sleep diary next morning. Monitoring indicators were collected for linear correlation analysis and paired t test to compare the differences. RESULTS: Under low sensitivity conditions, the correlation coefficient of total sleep time (TST) between PSG and Actiwatch was 0.53 (P < 0.05). Paired t test analysis showed that there was no significant difference between the TSTs of Actiwatch and PSG (t=-0.890, P=0.36). According to age stratification, the smaller the age, the stronger the correlation between the TSTs of Actiwatch and PSG, and the coefficient could be up to 0.92 (P < 0.05). Paired t test showed that there was no significant difference between them (t=-1.057, P=0.35). According to the stratification by diagnosis, the correlation coefficient between the TSTs of Actiwatch and PSG in normal PSG group could be as high as 0.79 (P < 0.05), the results of paired t test showed that there was no significant difference between the TSTs of Actiwatch and PSG in normal PSG group (t=-0.784, P=0.44). CONCLUSION: As a wearable home recorder, when the analysis parameters of Actiwatch were set as low sensitivity, PSG and Actiwatch had the highest TST correlation. The younger the age, the stronger correlation between the TSTs of Actiwatch and PSG. The PSG and Actiwatch subjects with normal PSG presentation had a higher TST correlation.


Assuntos
Actigrafia , Transtornos do Sono-Vigília , Humanos , Polissonografia , Reprodutibilidade dos Testes , Sensibilidade e Especificidade , Sono , Tempo
6.
Sensors (Basel) ; 21(20)2021 Oct 18.
Artigo em Inglês | MEDLINE | ID: mdl-34696120

RESUMO

In order to solve the problems of long path planning time and large number of redundant points in the rapidly-exploring random trees algorithm, this paper proposed an improved algorithm based on the parent point priority determination strategy and the real-time optimization strategy to optimize the rapidly-exploring random trees algorithm. First, in order to shorten the path-planning time, the parent point is determined before generating a new point, which eliminates the complicated process of traversing the random tree to search the parent point when generating a new point. Second, a real-time optimization strategy is combined, whose core idea is to compare the distance of a new point, its parent point, and two ancestor points to the target point when a new point is generated, choosing the new point that is helpful for the growth of the random tree to reduce the number of redundant points. Simulation results of 3-dimensional path planning showed that the success rate of the proposed algorithm, which combines the strategy of parent point priority determination and the strategy of real-time optimization, was close to 100%. Compared with the rapidly-exploring random trees algorithm, the number of points was reduced by more than 93.25%, the path planning time was reduced by more than 91.49%, and the path length was reduced by more than 7.88%. The IRB1410 manipulator was used to build a test platform in a laboratory environment. The path obtained by the proposed algorithm enables the manipulator to safely avoid obstacles to reach the target point. The conclusion can be made that the proposed strategy has a better performance on optimizing the success rate, the number of points, the planning time, and the path length.


Assuntos
Algoritmos , Simulação por Computador , Robótica , Tempo
7.
Annu Rev Biomed Data Sci ; 4: 165-187, 2021 07 20.
Artigo em Inglês | MEDLINE | ID: mdl-34465177

RESUMO

Electronic health records (EHRs) are becoming a vital source of data for healthcare quality improvement, research, and operations. However, much of the most valuable information contained in EHRs remains buried in unstructured text. The field of clinical text mining has advanced rapidly in recent years, transitioning from rule-based approaches to machine learning and, more recently, deep learning. With new methods come new challenges, however, especially for those new to the field. This review provides an overview of clinical text mining for those who are encountering it for the first time (e.g., physician researchers, operational analytics teams, machine learning scientists from other domains). While not a comprehensive survey, this review describes the state of the art, with a particular focus on new tasks and methods developed over the past few years. It also identifies key barriers between these remarkable technical advances and the practical realities of implementation in health systems and in industry.


Assuntos
Mineração de Dados , Médicos , Registros Eletrônicos de Saúde , Humanos , Aprendizado de Máquina , Tempo
9.
Math Biosci Eng ; 18(5): 6819-6840, 2021 08 13.
Artigo em Inglês | MEDLINE | ID: mdl-34517559

RESUMO

In this paper, we investigate the existence of global attractors, extreme stability, periodicity and asymptotically periodicity of solutions of the delayed population model with survival rate on isolated time scales given by $ x^{\Delta} (t) = \gamma(t) x(t) + \dfrac{x(d(t))}{\mu(t)}e^{r(t)\mu(t)\left(1 - \frac{x(d(t))}{\mu(t)}\right)}, \ \ t \in \mathbb T. $ We present many examples to illustrate our results, considering different time scales.


Assuntos
Periodicidade , Taxa de Sobrevida , Tempo
10.
Int J Surg ; 94: 106098, 2021 Oct.
Artigo em Inglês | MEDLINE | ID: mdl-34509672

RESUMO

BACKGROUND: Spinal cord injuries (SCI) are a devastating condition and can lead to severe functional and psychosocial problems. However, the influence of the timing of the surgical intervention for acute SCI remains debated, with substantial variability in clinical practice. Thus, this study aims to compare the efficacy of early and late surgical intervention for acute SCI. METHODS: A systematic search of PubMed, Embase, Cochrane Library and Web of Science up to January 10, 2021 was conducted for relevant studies that compared early and late acute SCI. Neurological outcomes were assessed by American Spinal Injury Association (ASIA). Early surgery was defined as the surgical intervention within 24 h after spinal injury. The primary outcome was the change of ASIA score from baseline to follow-up time after spinal injury. Second primary outcomes were clinical outcomes including neurological improvement rate, mortality, length of stay (LOS), charges ($), complications and ASIA Impairment Scale (AIS). All statistical analyses were performed using standard statistical procedures provided in Review Manager 5.2. RESULTS: A total of 16 studies including 3977 SCI patients were identified finally. Our pooled results indicated that, compared with late surgery, patients who underwent early surgery experienced more ASIA score improvement, with pooled MDs of 2.32 points (95% CI 1.07-3.57; P = 0.0003) in total motor scores, 5.13 points (95% CI 3.94-6.32; P < 0.0001) in light touch scores, and 4.49 points (95% CI 2.22-6.76; P = 0.0001) in pin prick scores respectively. In addition, patients receiving early surgery experienced more total motor score after surgery (MD 3.30; 95% CI 0.82-5.79; P = 0.009). Patients who had early surgery also had higher neurological improvement rate (OR 1.66; 95% CI 1.19-2.31; P = 0.003), shorter LOS (MD -4.77; 95% CI -7.42 to -2.12), less charges ($) (MD -0.33; 95% CI -0.43 to -0.22), lower incidence of complications (OR 0.62; 95% CI 0.48-0.81), and higher AIS improvement rate (OR 1.71; 95% CI 1.20-2.44) respectively. CONCLUSIONS: Compared with late surgery, acute SCI patients who underwent early surgery experienced greater recovery after spinal injury, with better neurological improvement, shorter LOS, less charges and lower incidence of complications.


Assuntos
Descompressão Cirúrgica , Traumatismos da Medula Espinal , Humanos , Tempo de Internação , Traumatismos da Medula Espinal/cirurgia , Tempo , Resultado do Tratamento
11.
Elife ; 102021 09 20.
Artigo em Inglês | MEDLINE | ID: mdl-34542407

RESUMO

The flexible and efficient adaptation to dynamic, rapid changes in the auditory environment likely involves generating and updating of internal models. Such models arguably exploit connections between the neocortex and the cerebellum, supporting proactive adaptation. Here, we tested whether temporo-cerebellar disconnection is associated with the processing of sound at short timescales. First, we identify lesion-specific deficits for the encoding of short timescale spectro-temporal non-speech and speech properties in patients with left posterior temporal cortex stroke. Second, using lesion-guided probabilistic tractography in healthy participants, we revealed bidirectional temporo-cerebellar connectivity with cerebellar dentate nuclei and crura I/II. These findings support the view that the encoding and modeling of rapidly modulated auditory spectro-temporal properties can rely on a temporo-cerebellar interface. We discuss these findings in view of the conjecture that proactive adaptation to a dynamic environment via internal models is a generalizable principle.


Assuntos
Percepção Auditiva , Encéfalo/fisiologia , Cerebelo/diagnóstico por imagem , Conectoma , Lobo Temporal/diagnóstico por imagem , Tempo , Adulto , Estudos de Casos e Controles , Cerebelo/fisiopatologia , Feminino , Humanos , Imageamento por Ressonância Magnética/métodos , Masculino , Pessoa de Meia-Idade , Acidente Vascular Cerebral/fisiopatologia , Análise e Desempenho de Tarefas , Lobo Temporal/fisiopatologia
12.
Nutrients ; 13(9)2021 Aug 25.
Artigo em Inglês | MEDLINE | ID: mdl-34578819

RESUMO

BACKGROUND: Time restricted Feeding (TRF) is a dietary pattern utilized by endurance athletes, but there is insufficient data regarding its effects on performance and metabolism in this population. The purpose of this investigation was to examine the effects of a 16/8 TRF dietary pattern on exercise performance in trained male endurance runners. METHODS: A 4-week randomized crossover intervention was used to compare an 8-h TRF to a 12-h normal diet (ND) feeding window. Exercise training and dietary intake were similar across interventions. Runners completed a dual-energy X-ray absorptiometry (DXA) scan to assess body composition, a graded treadmill running test to assess substrate utilization, and ran a 10 km time trial to assess performance. RESULTS: There was a significant decrease in fat mass in the TRF intervention (-0.8 ± 1.3 kg with TRF (p = 0.05), vs. +0.1 ± 4.3 kg with ND), with no significant change in fat-free mass. Exercise carbon dioxide production (VCO2) and blood lactate concentration were significantly lower with the TRF intervention (p ≤ 0.02). No significant changes were seen in exercise respiratory exchange ratio or 10 km time trial performance (-00:20 ± 3:34 min:s TRF vs. -00:36 ± 2:57 min:s ND). CONCLUSION: This investigation demonstrated that adherence to a 4-week 16/8 TRF dietary intervention decreased fat mass and maintained fat-free mass, while not affecting running performance, in trained male endurance runners.


Assuntos
Tecido Adiposo , Desempenho Atlético/estatística & dados numéricos , Composição Corporal , Treino Aeróbico/métodos , Jejum , Corrida , Adulto , Atletas/estatística & dados numéricos , Dieta , Humanos , Masculino , Valores de Referência , Tempo , Adulto Jovem
13.
Int J Clin Pract ; 75(11): e14798, 2021 Nov.
Artigo em Inglês | MEDLINE | ID: mdl-34482615

RESUMO

BACKGROUND/AIM: Permanent daylight saving time has been implemented in Turkey since 2016. The present study determines the characteristics of road traffic collisions in the short-term, trauma severity, and whether permanent daylight saving time has an impact on these parameters. MATERIALS AND METHODS: Drivers admitted because of road traffic collisions to a tertiary care university hospital emergency service two weeks before and after the transition to wintertime in 2014 and 2015 and summertime in 2015 and 2016 as well as those admitted two weeks before and after the same period with permanent daylight saving time in 2016 and 2017 wintertime and 2017 and 2018 summertime were included in the study. Trauma severity was measured using the Injury Severity Score. RESULTS: The study analysed the data of 710 patients. There was no statistically significant difference was found between admissions in the summertime and permanent daylight saving time periods in terms of gender, time of admission, week of admission, Injury Severity Score and outcome (P > .05 for all values). CONCLUSIONS: In this study, we examined the short-term effects of daylight saving time on road traffic collisions, and demonstrated that it had no impact on the number or time of admission, trauma severity and patient outcomes. More comprehensive studies covering longer periods can be performed across the country.


Assuntos
Acidentes de Trânsito , Ferimentos e Lesões , Serviço Hospitalar de Emergência , Hospitalização , Humanos , Tempo , Turquia/epidemiologia , Ferimentos e Lesões/epidemiologia , Ferimentos e Lesões/terapia
15.
BMC Public Health ; 21(1): 1673, 2021 09 15.
Artigo em Inglês | MEDLINE | ID: mdl-34521376

RESUMO

BACKGROUND: The aim of this study was to investigate bidirectional associations between (prolonged) sitting time and sleep duration in 12- to 14-year-old adolescents using a between-subjects and within-subjects analyses approach. METHODS: Observational data were used from 108 adolescents (53% girls; mean age 12.9 (SD 0.7) years) from six schools in Flanders, Belgium. The Axivity AX3 triaxial accelerometer, worn on the thigh, was used to assess daily total sitting time and daily time spent in sedentary bouts of ≥30 min (as a proxy for prolonged sitting time). The Fitbit Charge 3 was used to assess nightly sleep duration. Both monitors were worn on schooldays only (ranging from 4 to 5 days). Linear mixed models were conducted to analyse the associations, resulting in four models. In each model, the independent variable (sleep duration, sitting time or prolonged sitting time) was included as within- as well as between-subjects factor. RESULTS: Within-subjects analyses showed that when the adolescents sat more and when the adolescents spent more time sitting in bouts of ≥30 min than they usually did on a given day, they slept less during the following night (p = 0.01 and p = 0.05 (borderline significant), respectively). These associations were not significant in the other direction. Between-subjects analyses showed that adolescents who slept more on average, spent less time sitting (p = 0.006) and less time sitting in bouts of ≥30 min (p = 0.004) compared with adolescents who slept less on average. Conversely, adolescents who spent more time sitting on average and adolescents who spent more time sitting in bouts of ≥30 min on average, slept less (p = 0.02 and p = 0.003, respectively). CONCLUSIONS: Based on the between-subjects analyses, interventions focusing on reducing or regularly breaking up sitting time could improve adolescents' sleep duration on a population level, and vice versa. However, the within-subjects association was only found in one direction and suggests that to sleep sufficiently during the night, adolescents might limit and regularly break up their sitting time the preceding day. TRIAL REGISTRATION: Data have been used from our trial registered at ClinicalTrials.gov ( NCT04327414 ; registered on March 11, 2020).


Assuntos
Comportamento Sedentário , Postura Sentada , Adolescente , Criança , Feminino , Humanos , Masculino , Instituições Acadêmicas , Sono , Tempo
16.
Clin Immunol ; 232: 108859, 2021 11.
Artigo em Inglês | MEDLINE | ID: mdl-34563685

RESUMO

Changes in immune response of children with congenital toxoplasmosis (CT) regarding infection evolution and therapeutic intervention was addressed. Infants with CT presented increased counts of monocytes, CD3-CD16-CD56High, CD3+CD56+ and CD4+ T-cells 1-year after treatment onset (TOXO1-yearAT). Smaller numbers of CD3-CD16-CD56+ and TCRγδ+ T-cells were specifically observed in infants with retinochoroidal lesions (L(+)). When infants were classified based on the baseline status, expansion of CD3-CD16-CD56High and CD4+ T-cells were observed in L(+) who had active, active/cicatricial or cicatricial lesions. Infants who had active or active/cicatricial lesions display augmented numbers of monocytes, CD3-CD16+CD56+, CD3+CD56+, CD8+DR+ and TCRγδ+ T-cells and those with active/cicatricial or cicatricial at baseline displayed increase in CD14+CD64+ monocytes. Moreover, all L(+) had increased IFN-γ+ and IL-10+ CD4+ T-cells, while L(-) had increased ratios of TNF+, IFN-γ+ and IL-4+ NK-cells upon antigen-specific stimulation. Persistent alterations in leukocytes in TOXO1-yearAT suggest long-term sequels in the immune system of infants with CT.


Assuntos
Antiprotozoários/efeitos adversos , Linfócitos/efeitos dos fármacos , Monócitos/efeitos dos fármacos , Toxoplasmose Congênita/tratamento farmacológico , Toxoplasmose Congênita/imunologia , Feminino , Humanos , Lactente , Recém-Nascido , Masculino , Fenótipo , Pirimetamina/efeitos adversos , Sulfadiazina/efeitos adversos , Tempo
17.
Bioessays ; 43(11): e2100159, 2021 11.
Artigo em Inglês | MEDLINE | ID: mdl-34585760

RESUMO

A paradigm shift in the human chronotoxicity of xenobiotics would study two-sided desynchronized phenomena of interfacial interactions between cyclic or periodic environmental insults and the endogenous response and recovery profile. These systems-based networks are under the influence of well-synchronized biological clocks and their metabolic regulators. This perspective argues in favor of addressing the concept of synchronization in studies involving critical life windows of susceptibility, or circadian rhythms, or 24-hour (periodic) diurnal rhythms and answering whether these disruptions in synchronization would affect response and recovery or disease phenotypes associated with environmental insults, e.g., xenobiotics. Synchronization or synchrony is defined as the totality of elements that appear during the same time period within a system, including the network of interactions between the system's elements. Desynchronized interfaces during critical life windows or in time-repeated exposure events would likely lead to initiating a cascade of adverse health effects associated with differentiated disease phenotypes.


Assuntos
Relógios Circadianos , Ritmo Circadiano , Humanos , Fenótipo , Tempo , Xenobióticos/toxicidade
19.
Clin Pediatr (Phila) ; 60(9-10): 418-426, 2021 09.
Artigo em Inglês | MEDLINE | ID: mdl-34342242

RESUMO

Pediatrician Screening, Brief Intervention, and Referral to Treatment (SBIRT) practices vary widely, though little is known about the correlates of SBIRT implementation. Using data from a national sample of US pediatricians who treat adolescents (n = 250), we characterized self-reported utilization rates of SBIRT among US pediatricians and identified provider- and practice-level characteristics and barriers associated with SBIRT utilization. All participants completed an electronic survey querying the demographics, practice patterns, and perceived barriers related to SBIRT practices. Our results showed that 88% of respondents reported screening for substance use annually, but only 26% used structured/validated screening instruments. Furthermore, 40% of respondents provided evidence-based brief interventions, and only 11% implemented all core SBIRT practices. Common barriers (eg, confidentiality and insufficient time) and unique provider- and setting-specific barriers to implementation were identified. These findings indicate that although most pediatricians deliver some SBIRT components in their practice, few implement the full SBIRT model, and barriers persist.


Assuntos
Intervenção na Crise/métodos , Programas de Rastreamento/métodos , Pediatras/estatística & dados numéricos , Padrões de Prática Médica/estatística & dados numéricos , Encaminhamento e Consulta/estatística & dados numéricos , Transtornos Relacionados ao Uso de Substâncias/diagnóstico , Transtornos Relacionados ao Uso de Substâncias/terapia , Adolescente , Comportamento do Adolescente , Confidencialidade , Intervenção na Crise/estatística & dados numéricos , Humanos , Transtornos Relacionados ao Uso de Substâncias/prevenção & controle , Tempo , Estados Unidos
20.
Sensors (Basel) ; 21(15)2021 Jul 21.
Artigo em Inglês | MEDLINE | ID: mdl-34372186

RESUMO

We propose a physical activity recognition and monitoring framework based on wearable sensors during maternity. A physical activity can either create or prevent health issues during a given stage of pregnancy depending on its intensity. Thus, it becomes very important to provide continuous feedback by recognizing a physical activity and its intensity. However, such continuous monitoring is very challenging during the whole period of maternity. In addition, maintaining a record of each physical activity, and the time for which it was performed, is also a non-trivial task. We aim at such problems by first recognizing a physical activity via the data of wearable sensors that are put on various parts of body. We avoid the use of smartphones for such task due to the inconvenience caused by wearing it for activities such as "eating". In our proposed framework, a module worn on body consists of three sensors: a 3-axis accelerometer, 3-axis gyroscope, and temperature sensor. The time-series data from these sensors are sent to a Raspberry-PI via Bluetooth Low Energy (BLE). Various statistical measures (features) of this data are then calculated and represented in features vectors. These feature vectors are then used to train a supervised machine learning algorithm called classifier for the recognition of physical activity from the sensors data. Based on such recognition, the proposed framework sends a message to the care-taker in case of unfavorable situation. We evaluated a number of well-known classifiers on various features developed from overlapped and non-overlapped window size of time-series data. Our novel dataset consists of 10 physical activities performed by 61 subjects at various stages of maternity. On the current dataset, we achieve the highest recognition rate of 89% which is encouraging for a monitoring and feedback system.


Assuntos
Dispositivos Eletrônicos Vestíveis , Algoritmos , Exercício Físico , Feminino , Humanos , Gravidez , Reconhecimento Psicológico , Tempo
SELEÇÃO DE REFERÊNCIAS
DETALHE DA PESQUISA
...