Time–frequency analysis in infant cry classification using quadratic time frequency distributions

• Autorzy: Saraswathy J. , Hariharan M. , Khairunizam W. , Sarojini J. , Thiyagar N. , Sazali Y. , Nisha S.

Identyfikatory

DOI

10.1016/j.bbe.2018.05.002

• Języki publikacji: EN

Abstrakty

EN

This paper presents a new investigation of time–frequency (t–f) based signal processing approach using quadratic time–frequency distributions (QTFDs) namely spectrogram(SPEC), Wigner–Ville distribution (WVD), Smoothed–Wigner Ville distribution (SWVD), Choi–William distribution (CWD) and modified B-distribution (MBD) for classification of infant cry signals. t–f approaches have proved as an efficient approach for applications involving the non stationary signals. In feature extraction, a cluster of t–f based features were extracted by extending the time-domain and frequency-domain features to the joint t–f domain from the generated t–f representation. Conventional features such as mel-frequency cepstral coefficients (MFCCs) and linear prediction coefficients (LPCs) were also extracted in order to compare the effectiveness of the t–f methods. The efficacy of the extracted feature vectors was validated using probabilistic neural network (PNN) and general regression neural network (GRNN). The proposed methodology was implemented to classify different sets of binary classification problems of infant cry signals from different native. The best empirical result of above 90% was reported and revealed the good potential of t–f methods in the context of infant cry classification.

Słowa kluczowe

EN

infant cry; time-frequency analysis; quadratic time-frequency distributions; t–f based feature extraction

PL

płacz niemowlęcy; analiza czasowo-częstotliwościowa; ekstrakcja cech

• Wydawca: Nałęcz Institute of Biocybernetics and Biomedical Engineering of the Polish Academy of Sciences ; Elsevier
• Czasopismo: Biocybernetics and Biomedical Engineering
• Rocznik: 2018
• Tom: Vol. 38, no. 3
• Strony: 634--645
• Opis fizyczny: Bibliogr. 40 poz., rys., tab., wykr.

Twórcy

autor

Saraswathy J.

• School of Mechatronic Engineering, University Malaysia Perlis (UniMAP), Campus Pauh Putra, 02600 Perlis, Malaysia

autor

Hariharan M.

• Department of Biomedical Engineering, Faculty of Engineering & Technology, SRM Institute of Science and Technology, Kattankulathur, Chennai, Tamil Nadu, India

autor

Khairunizam W.

• School of Mechatronic Engineering, University Malaysia Perlis (UniMAP), Perlis, Malaysia

autor

Sarojini J.

• School of Bioprocess Engineering, University Malaysia Perlis (UniMAP), Perlis, Malaysia

autor

Thiyagar N.

• Department of Pediatrics, Hospital Sultanah Bahiyah, Kedah, Malaysia

autor

Sazali Y.

• Universiti Kuala Lumpur Malaysian Spanish Institute, Kedah, Malaysia

autor

Nisha S.

• School of Mechatronic Engineering, University Malaysia Perlis (UniMAP), Perlis, Malaysia

Bibliografia

• [1] Douglas PS, Hill PS. The crying baby: what approach? Curr Opin Pediat 2011;23(5):523–9.
• [2] Cry translator. 2013©Biloop Technologic, S. L.
• [3] Baby cry analyzer. WhyCryMini®, Dimple Inc.
• [4] Varallyay G. Future prospects of the application of the infant cry in the medicine. Electr Eng 2006;50:47–62.
• [5] Lederman D, Cohen A, Zmora E, Wermke K, Hauschildt S, Stellzig-Eisenhauer A. Classification of cries of infants with cleft-palate using parallel hidden markov models. Med Biol Eng Comput 2008;46:965–75.
• [6] Fuhr T, Reetz H, Wegener C. Comparison of supervised-learning models for infant cry classification/Vergleich von Klassifikationsmodellen zur Säuglingsschreianalyse. Int J Health Profess 2015;2(1):4–15.
• [7] Abou-Abbas L, Tadj C, Fersaie HA. A fully automated approach for baby cry signal segmentation and boundary detection of expiratory and inspiratory episodes. J Acoust Soc Am 2017;142(3):1318–31.
• [8] Naithani G, Kivinummi J, Virtanen T, Tammela O, Peltola MJ, Leppänen JM. Automatic segmentation of infant cry signals using hidden Markov models. EURASIP J Audio Speech Music Process 2018;2018(1):1.
• [9] Robb MP, Crowell DH, Dunn-Rankin P. Sudden infant death syndrome: cry characteristics. Int J Pediat Otorhinolaryngol 2013;77(8):1263–7.
• [10] Mittal VK. Discriminating the infant cry sounds due to pain vs. discomfort towards assisted clinical diagnosis. 7th Workshop on Speech and Language Processing for Assistive Technologies, SLPAT; 2016. pp. 37–42.
• [11] Kheddache T. Characterization of pathologic cries of newborns based on fundamental frequency estimation. Engineering 2013;5:272–6.
• [12] Fuamenya NA, Robb MP, Wermke K. Noisy but effective: crying across the first 3 months of life. J Voice 2015;29 (3):281–6.
• [13] Vempada RR, Kumar B, Rao KS. Characterization of infant cries using spectral and prosodic features. Proceedings of the National Conference in Communications; 2012. p. 1–5.
• [14] Várallyay Jr G. The melody of crying. Int J Pediat Otorhinolaryngol 2007;71(11):1699–708.
• [15] Orozco J, Garcia C. Detecting pathologies from infant cry applying scaled conjugate gradient neural networks. Eur Symp Artif Neural Networks 2003;349–54.
• [16] Hariharan M, Chee LS, Yaacob S. Analysis of infant cry through weighted linear prediction cepstral coefficients and probabilistic neural network. J Med Syst 2012;36:1309–15.
• [17] Reyes Galaviz OF, Cano Ortiz S, Reyes Garcia C. Evolutionary-neural system to classify infant cry units for pathologies identification in recently born babies. Mexican Int Conf Artif Intell 2009;330–5.
• [18] Rosales Perez A, Reyes García CA, Gonzalez JA, Reyes-Galaviz OF, Escalante HJ, Orlandi S. Classifying infant cry patterns by the genetic selection of a fuzzy model. Biomed Signal Process Control 2015;17:38–46.
• [19] Santiago Sanchez K, Reyes Garcia CA, Gomez Gil P. Type-2 fuzzy sets applied to pattern matching for the classification of cries of infants under neurological risk. Lect Notes Comput Sci 2009;5754:201–10.
• [20] Alaie HF, Abou-Abbas L, Tadj C. Cry-based infant pathology classification using GMMs. Speech Commun 2016;77:28–52.
• [21] Onu CC, Udeogu I, Ndiomu E, Kengni U, Precup D, Sant'anna GM, et al. Ubenwa: cry-based diagnosis of birth asphyxia; 2017, ar*Xiv preprint ar*Xiv:1711.06405.
• [22] Zabidi A, Yassin IM, Hassan HA, Ismail N, Hamzah MMAM, Rizman ZI, et al. Detection of asphyxia in infants using deep learning convolutional neural network (CNN) trained on Mel frequency cepstrum coefficient (MFCC) features extracted from cry sounds. J Fund Appl Sci 2017;9(3S):768–78.
• [23] Boashash B, Boubchir L. On the selection of time–frequency features for improving the detection and classification of newborn EEG seizure signals and other abnormalities. Neural Inform Process 2012;634–43.
• [24] Boashash B, Khlif MS, Ben-Jabeur T, East CE, Colditz PB. Passive detection of accelerometer-recorded fetal movements using a time–frequency signal processing approach. Digit Signal Process 2014;25:134–55.
• [25] Zahedi F, Huang H. Time–frequency analysis of electromechanical impedance (EMI) signature for physics-based damage detections using piezoelectric wafer active sensor (PWAS). Smart Mater Struct 2017;26(5):055010.
• [26] Mijovic B, Silva M, Van den Bergh BRH, Allegaert K, Aerts JM, Berckmans D, et al. Assessment of pain expression in infant cry signals using Empirical Mode Decomposition. Methods Inf Med 2010;49(5):448–52.
• [27] Hariharan M, Yaacob S, Awang SA. Pathological infant cry analysis using wavelet packet transform and probabilistic neural network. Exp Syst Appl 2011;12:15377–82.
• [28] Hariharan M, Sindhu R, Yaacob S. Normal and hypoacoustic infant cry signal classification using time–frequency analysis and general regression neural network. Comput Methods Prog Biomed 2012;108:559–69.
• [29] Hariharan M, Saraswathy J, Sindhu R, Khairunizam W, Yaacob S. Infant cry classification to identify asphyxia using time–frequency analysis and radial basis neural networks. Exp Syst Appl 2012;39:9515–23.
• [30] Saraswathy J, Hariharan M, Khairunizam W, Yaacob S, Thiyagar N. Infant cry classification: time frequency analysis. International Conference in Control System, Computing and Engineering. 2013. pp. 499–504.
• [31] Saraswathy J, Hariharan M, Nadarajaw T, Khairunizam W, Yaacob S. Optimal selection of mother wavelet for accurate infant cry classification. Aust Phys Eng Sci Med 2014;37:439–56.
• [32] Abou-Abbas L, Montazeri L, Gargour C, Tadj C. On the use of EMD for automatic newborn cry segmentation. International Conference on Advances in Biomedical Engineering (ICABME). 2015. pp. 262–5.
• [33] Abou-Abbas L, Tadj C, Gargour C, Montazeri L. Expiratory and inspiratory cries detection using different signals' decomposition techniques. J Voice 2017;31(2):259–313.
• [34] Baeck HE, Souza MN. Longitudinal study of the fundamental frequency of hunger cries along the first 6 months of healthy babies. J Voice 2007;21:551–9.
• [35] Mampe B, Friederici AD, Christophe A, Wermke K. Newborns' cry melody is shaped by their native language. Curr Biol 2009;19:1994–7.
• [36] Wermke K, Robb MP. Fundamental frequency of neonatal crying: does body size matter? J Voice 2010;24:388–94.
• [37] Rautava L, Lempinen A, Ojala S, Parkkola R, Rikalainen H, Lapinleimu H, et al. Acoustic quality of cry in very-lowbirth-weight infants at the age of 1 1/2 years. Early Hum Dev 2007;83:5–12.
• [38] Barajas-Montiel SE, Reyes-Garcia CA, Arch-Tirado E, Mandujano M. Improving baby caring with automatic infant cry recognition. Lect Notes Comput Sci 2006;4061:691–8.
• [39] Kheddache T. Resonance frequencies behavior in pathologic cries of newborns. J Voice 2014;29:1–12.
• [40] www.time-frequency.net.

Uwagi

PL

Opracowanie rekordu w ramach umowy 509/P-DUN/2018 ze środków MNiSW przeznaczonych na działalność upowszechniającą naukę (2018).