Avaliação de representações simbólicas de movimentos articulares humanos úteis para sistemas de telemedicina
DOI:
https://doi.org/10.59681/2175-4411.v18.2026.1535Keywords:
Estudos de Séries Temporais, Reconhecimento Automatizado de Padrão, TelemedicinaAbstract
Objective: To evaluate the performance of SAX and SFA methods for the symbolic representation of time series from human joint movements, considering three alphabet sizes (three, seven, and ten symbols) and three types of movement. Methods: Eight healthy volunteers performed elbow flexion/extension, shoulder abduction/adduction, and shoulder circumduction, with acceleration data collected using a Samsung Galaxy A51 smartphone attached to the wrist. Signals were smoothed using a Gaussian filter (σ = 2), automatically segmented, and converted into symbolic words (10 symbols) using both SAX and SFA methods. Pairwise 4×4 method comparisons for each movement and alphabet size were performed using the Friedman test or repeated-measures ANOVA, depending on data distribution. Additional 9×9 comparisons among the best methods were also conducted. Results and Discussion: No statistically significant differences were observed in any analysis. For flexion/extension and abduction/adduction, all data were analyzed using the Friedman test (p-value = 0.3916). For circumduction, alphabet sizes of three and seven symbols also showed no differences, as did ten symbols, with ANOVA also showing no statistically significant difference (p-value = 0.0517). The 9×9 comparison of best methods also revealed no differences (p-value = 0.4544). Conclusion: Both methods showed equivalent performance, with SAX recommended due to its lower computational complexity (O(n)) compared to SFA (O(n log n)), making it more suitable for applications with processing constraints or real-time analysis requirements.
Downloads
References
Brasil. Ministério da Saúde. LER e DORT são as doenças que mais acometem os trabalhadores, aponta estudo [Internet]. Brasília: Ministério da Saúde; 2019 [citado 2025 nov 5]. Disponível em: https://www.gov.br/saude/pt-br/assuntos/noticias/2019/abril/ler-e-dort-sao-as-doencas-que-mais-acometem-os-trabalhadores-aponta-estudo.
Wang X, Wang Y, Wu J. Position-aware indoor human activity recognition using multisensors embedded in smartphones. Sensors. 2024;24(11):3367.
Baroudi L, Barton K, Cain SM, Shorter KA. Classification of human walking context using a single-point accelerometer. Sci Rep. 2024;14:3039.
[Autor1], [Autor2], [Autor3], [Autor4], [Autor5], [Autor6], [Autor7]. [Título do artigo anonimizado]. [Periódico] [Internet]. [Ano];[Volume]([Número]):[Páginas]. Disponível em: [URL]. DOI: [DOI].
Borges RC, Parreira WD, da Mata DL, Lage RM. Protótipo acessível para monitoramento cardíaco utilizando fotopletismografia: implementação e perspectivas. J Health Inform. 2024;16(Especial). doi:10.59681/2175-4411.v16.iEspecial.2024.1324.
Lanazi M, Aldahr RS, Ilyas M. Human activity recognition through smartphone inertial sensors with ML approach. Eng Technol Appl Sci Res. 2024;14(1):12780-7.
Lin J, Keogh EJ, Lonardi S, Pazzani A. A symbolic representation of time series, with implications for streaming data mining. Proceedings of the 8th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2003;2-11.
Schäfer P, Höggqvist M. SFA: A symbolic Fourier approximation and index for similarity search in high dimensional datasets. Data Min Knowl Discov. 2012;29(2):228-42.
Shieh J, Keogh E. iSAX: Indexing and mining terabyte-sized time series. Proceedings of the 14th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2008;623-31.
Fulcher BD, Jones NS. Highly comparative feature-based time-series classification. IEEE Trans Knowl Data Eng. 2014;26(12):3026-37.
Sistema S – Sistema S em Telemedicina [software]. Informações de autoria e registro omitidas para revisão cega.
Hamill J, Knutzen KM, Derrick TR. Biomecânica básica dos movimentos humanos. 4ª ed. Barueri (SP): Manole; 2016.
Hamming RW. Error detecting and error correcting codes. Bell Syst Tech J. 1950;29(2):147-60.
D’Agostino RB. Tests for normal distribution. In: Goodness-of-fit techniques. New York: Marcel Dekker; 1986. p. 367-419.
Fisher RA. Statistical methods for research workers. 1st ed. Edinburgh: Oliver and Boyd; 1925.
Friedman M. The use of ranks to avoid the assumption of normality implicit in the analysis of variance. J Am Stat Assoc. 1937;32(200):675-701.
Nemenyi PB. Distribution-free multiple comparisons. Princeton: Princeton University; 1963. Ph.D. Thesis.
Phinyomark A, Khushaba RN, Scheme E. Feature extraction and selection for myoelectric control based on wearable EMG sensors. Sensors (Basel). 2017;17(5):1105. doi:10.3390/s17051105.
Sousa Lima W, Batista LV, Keogh E, Povinelli RJ, Batista GEAPA. Human Activity Recognition Based on Symbolic Representations. Sensors (Basel). 2018;18(11):4045. doi:10.3390/s18114045.
Philip S, Cao Y, Li M. Sensor Based Time Series Classification of Body Movement. In: Bio-Inspired Models of Network, Information, and Computing Systems (BIONETICS 2010). Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering. Springer; 2012. p. 303-309. doi:10.1007/978-3-642-32615-8_30.
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Ana Paula Merencia, Huei Diana Lee, Weber Shoity Resende Takaki, Alexandre Peiter Ferraz, Wu Feng Chung, Newton Spolaôr

This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
Submission of a paper to Journal of Health Informatics is understood to imply that it is not being considered for publication elsewhere and that the author(s) permission to publish his/her (their) article(s) in this Journal implies the exclusive authorization of the publishers to deal with all issues concerning the copyright therein. Upon the submission of an article, authors will be asked to sign a Copyright Notice. Acceptance of the agreement will ensure the widest possible dissemination of information. An e-mail will be sent to the corresponding author confirming receipt of the manuscript and acceptance of the agreement.
