Vocalization removal for improved automatic segmentation of dual-axis swallowing accelerometry signals.

Automatic segmentation of dual-axis swallowing accelerometry signals can be severely affected by strong vocalizations. In this paper, a method based on periodicity detection is proposed to detect and remove such vocalizations. Periodic signal components are detected using conventional speech processing techniques and information from both axes are combined to improve vocalization detection accuracy. Experiments with 408 healthy subjects performing dry, wet, and wet chin tuck swallows show that the proposed method attains an average 95.3% sensitivity and 96.3% specificity. When applied in conjunction with an automatic segmentation algorithm, it is observed that segmentation accuracy improves by approximately 55%. These results encourage further development of medical devices for the detection of swallowing difficulties.

[1]  Kuldip K. Paliwal,et al.  Speech Coding and Synthesis , 1995 .

[2]  Tom Chau,et al.  A radial basis classifier for the automatic detection of aspiration in children with dysphagia , 2006, Journal of NeuroEngineering and Rehabilitation.

[3]  Tom Chau,et al.  Baseline Characteristics of Dual-Axis Cervical Accelerometry Signals , 2010, Annals of Biomedical Engineering.

[4]  Ryo Ishida,et al.  Hyoid Motion During Swallowing: Factors Affecting Forward and Upward Displacement , 2002, Dysphagia.

[5]  J. Richards,et al.  ``Wet Voice'' as a Predictor of Penetration and Aspiration in Oropharyngeal Dysphagia , 2000, Dysphagia.

[6]  Tom Chau,et al.  Segmentation of Dual-Axis Swallowing Accelerometry Signals in Healthy Subjects With Analysis of Anthropometric Effects on Duration of Swallowing Activities , 2009, IEEE Transactions on Biomedical Engineering.

[7]  Tom Chau,et al.  Anthropometric and Demographic Correlates of Dual-Axis Swallowing Accelerometry Signal Characteristics: A Canonical Correlation Analysis , 2010, Dysphagia.

[8]  Tom Chau,et al.  An Online Swallow Detection Algorithm Based on the Quadratic Variation of Dual-Axis Accelerometry , 2010, IEEE Transactions on Signal Processing.

[9]  J. Korpáš,et al.  Analysis of the cough sound: an overview. , 1996, Pulmonary pharmacology.

[10]  A. Rademaker,et al.  Incidence and Patient Characteristics Associated with Silent Aspiration in the Acute Care Setting , 1999, Dysphagia.

[11]  Daniel Berckmans,et al.  Assessing the sound of cough towards vocality , 2001, MAVEBA.

[12]  P. Marik,et al.  Aspiration pneumonia and dysphagia in the elderly. , 2003, Chest.

[13]  N. P. Reddy,et al.  Measurements of acceleration during videofluorographic evaluation of dysphagic patients. , 2000, Medical engineering & physics.

[14]  Youngsun Kim,et al.  Maximum Hyoid Displacement in Normal Swallowing , 2008, Dysphagia.

[15]  J. Hidler,et al.  Journal of Neuroengineering and Rehabilitation Quantification of Functional Weakness and Abnormal Synergy Patterns in the Lower Limb of Individuals with Chronic Stroke , 2006 .

[16]  P Piirilä,et al.  Objective assessment of cough. , 1995, The European respiratory journal.

[17]  T. Chau,et al.  Investigating the stationarity of paediatric aspiration signals , 2005, IEEE Transactions on Neural Systems and Rehabilitation Engineering.

[18]  Tom Chau,et al.  Swallow segmentation with artificial neural networks and multi-sensor fusion. , 2009, Medical engineering & physics.

[19]  Ronald J. Baken,et al.  Clinical measurement of speech and voice , 1987 .

[20]  T Chau,et al.  Time and time–frequency characterization of dual-axis swallowing accelerometry signals , 2008, Physiological measurement.

[21]  Amitava Das,et al.  Hybrid fuzzy logic committee neural networks for recognition of swallow acceleration signals , 2001, Comput. Methods Programs Biomed..