Using a stethoscope, the health care provider may hear normal breathing sounds, decreased or absent breath sounds, and abnormal breath sounds. Hinton, G., et al.: Deep neural networks for acoustic modeling in speech recognition: the shared views of four research groups. Normal lung sounds occur in all parts of the chest area, including above the collarbones and at the bottom of the rib cage. In: Proceedings of APSIPA 2020 Asia Pacific Signal and Information Processing Association Annual Summit and Conference, pp. Yamashita, M.: Construction of effective HMMs for classification between normal and abnormal respiration. In: Proceedings of ICASSP 2014, IEEE International Conference on Acoustics, Speech, and Signal Processing, pp. Yamashita, M., Himeshima, M., Matsunaga, S.: Robust classification between normal and abnormal lung sounds using adventitious-sound and heart-sound models. In: Proceedings of ICA 2010, International Congress on Acoustics, vol. Yamamoto, H., Matsunaga, S., Yamashita, M., Yamauchi K., Miyahara, S.: Classification between normal and abnormal respiratory sounds based on stochastic approach. In: Proceedings of ICASSP 2009, IEEE International Conference on Acoustics, Speech, and Signal Processing, pp. Matsunaga, S., Yamauchi, K., Yamashita, M., Miyahara, S.: Classification between normal and abnormal respiratory sounds based on maximum likelihood approach. Taplidou, S.A., Hadjileontiadis, L.J.: Wheeze detection based on time-frequency analysis of breath sounds. Marshall, A., Boussakta, S.: Signal analysis of medical acoustic sounds with applications to chest medicine. In: EMBC 2001, International Conference of the IEEE Engineering in Medicine and Biology Society (2001) Kahya, Y.P., Yerer, S., Cerid, O.: A wavelet-based instrument for detection of crackles in pulmonary sounds. Gavriely, N., Cugell, D.W.: Breath Sounds Methodology. The results proved the effectiveness of the proposed method. By selecting a suitable number of hidden layers and units for the DNN-HMM, the classification rate was increased (91.26%). In this paper, we present the construction of a DNN-HMM with high accuracy by selecting a suitable acoustic feature and setting the number of hidden layers and units for the DNN-HMM. However, in the case of lung sound, we cannot use a DNN because the amount of training data is small. In speech recognition, the accuracy was improved using a deep neural network (DNN)-HMM. However, the classification rate between normal and abnormal respiration was low (86.53%). For this purpose, we expressed the acoustic features of normal lung sound for healthy subjects and abnormal lung sound for patients by using the Gaussian mixture model (GMM)-hidden Markov model (HMM) and distinguished between normal and abnormal lung sounds. In this study, we aim to achieve the automatic detection of abnormal sounds from auscultatory sound. If it sounds like 'aa' when auscultating, it may indicate pleural effusion of lung consolidation. Have patient repeated 'ee' sound during auscultation of the lung fields. In many situations, abnormal sounds termed adventitious sounds are part of the lung sound of a subject suffering from a pulmonary disease. - Equal inspiratory and expiratory sounds.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |