Vücut Seslerinden Bölge Tanımlanması için İdeal Kayıt Süresinin Belirlenmesinde MFCC ve GTCC Özniteliklerinin Etkisinin Karşılaştırılması

İnsan vücudunun durumu hakkında bilgi almak için yapılabilecek en hızlı yöntemlerden birisi vücut seslerini analiz etmektir. Seslerin dijital ortama aktarılabilmesi bu analizi kolaylaştırmaktadır. Bu çalışmada kalp, akciğer ve karın bölgelerinden alınan ses verilerinden bölge tespiti yapılmıştır. Eğitimde 12 kişiden alınan 4000 örnekleme frekansına sahip 20s lik veriler kullanılmıştır. Veriler 9 farklı saniyede incelenmiştir. Her bir saniye için tüm veriler bölünmüş ve eğitim için hazırlanmıştır. MFCC ve GTCC kullanılarak öznitelikler çıkarılmış ve bu öznitelikler CNN modelinde eğitilmiştir. MFCC ve GTCC katsayılarının sonuçlar üzerindeki etkisi kıyaslanmıştır. Eğitimde en iyi sonuç %98 ile 1,5 saniyelik kayıtlardan alınan MFCC katsayısından, validationlarda ise en iyi sonuç %85 ile 1 saniyelik kayıtların MFCC katsayılarından elde edilmiştir. Genel validation sonuçlarına bakıldığında MFCC sonuçlarının daha başarılı olduğu görülmüştür.

Anahtar Kelimeler:

Artificial intelligence, body sounds, convolution nneural network, gammatone cepstral coefficients, mel-frequency cestrum coefficients

Comparison of the Effect of MFCC and GTCC Features on Determining the Ideal Recording Time for Body Sound Location Identification

One of the fastest ways to get information about the state of the human body is to analyse body sounds. The ability to transfer sounds to a digital medium facilitates this analysis. In this study, zone detection was performed from the sound data obtained from the heart, lung, and abdominal regions. 20s data with a sampling frequency of 4000 from 12 men were used in the training. The data was analysed in 9 different seconds. All data for each second is divided and prepared for training. Features were extracted using MFCC and GTCC and these features were trained in CNN model. The effect of MFCC and GTCC coefficients on the results was compared. In training, the best result was obtained from the MFCC coefficient obtained from 1.5-second recordings with 98%, and in validations, the best result was obtained from MFCC coefficients of 1-second recordings with 85%. Looking at the general validation results, it was seen that the MFCC results were more successful.

Keywords:

Artificial intelligence, body sounds, convolution nneural network, gammatone cepstral coefficients, mel-frequency cestrum coefficients,

PDF

___

Aziz, S., Khan, M. U., Shakeel, M., Mushtaq, Z., Khan, A. Z. 2019. "An Automated System towards Diagnosis of Pneumonia using Pulmonary Auscultations". MACS 2019 - 13th International Conference on Mathematics, Actuarial Science, Computer Science and Statistics, Proceedings.
Bahoura, M. and Ezzaidi, H., 2013. ‘‘Hardware implementation of MFCC feature extraction for respiratory sounds analysis,’’ in Proc. 8th Int. Workshop Syst., Signal Process. Appl. (WoSSPA), May 2013, pp. 226–229
Bahoura, M. and Pelletier, C., 2004. ‘‘Respiratory sounds classification using cepstral analysis and Gaussian mixture models,’’ in Proc. 26th Annu. Int. Conf. IEEE Eng. Med. Biol. Soc., Sep. 2004, pp. 9–12.
Bardou, D., Zhang, K., Ahmad, S. M. 2018. "Lung sounds classification using convolutional neural networks". Artificial Intelligence in Medicine, 88, 58–69.
Cheng, S., Wang, C., Yue, K., Li, R., Shen, F., Shuai, W., … Dai, L. 2022. "Automated sleep apnea detection in snoring signal using long short-term memory neural networks". Biomedical Signal Processing and Control, 71(PB), 103238.
Dar, J. A., Srivastava, K. K., Lone, S. A. 2022. "Jaya Honey Badger optimization-based deep neuro-fuzzy network structure for detection of (SARS-CoV) Covid-19 disease by using respiratory sound signals". International Journal of Intelligent Computing and Cybernetics.
Jayalakshmy, S., Sudha, G. F. 2021. "GTCC-based BiLSTM deep-learning framework for respiratory sound classification using empirical mode decomposition". Neural Computing and Applications, 33(24), 17029–17040.
Kutlu Y, Karaca G. Recognition of turkish command to play chess game using cnn. Akıllı Sistemler ve Uygulamaları Dergisi (Journal of Intelligent Systems with Applications) 2022; 5(1): 71-73.
Lella, K. K., Pja, A. 2022. "Automatic diagnosis of COVID-19 disease using deep convolutional neural network with multi-feature channel from respiratory sound data: Cough, voice, and breath". Alexandria Engineering Journal, 61(2), 1319–1334.
Mayorga, P., Druzgalski, C., Morelos, R. L., Gonzalez, O. H. and Vidales, J. 2010. ‘‘Acoustics based assessment of respiratory diseases using GMM classification,’’ in Proc. Annu. Int. Conf. IEEE Eng. Med. Biol., Aug. 2010, pp. 6312–6316.
Meng, F., Shi, Y., Wang, N., Cai, M., & Luo, Z. (2020). Detection of Respiratory Sounds Based on Wavelet Coefficients and Machine Learning. IEEE Access, 8, 155710–155720.
Mridha, K., Sarkar, S., Kumar, D. 2021. "Respiratory Disease Classification by CNN using MFCC". 2021 IEEE 6th International Conference on Computing, Communication and Automation, ICCCA 2021, 517–523.
Pittner, S., and Kamarthi, S. V. 1999. ‘‘Feature extraction from wavelet coefficients for pattern recognition tasks,’’ IEEE Trans. Pattern Anal. Mach. Intell., vol. 21, no. 1, pp. 83–88, Jan. 1999.
Shaharum, S. M., Sundaraj, K., Aniza, S., Palaniappan, R., Helmy, K. 2019. "A performance comparison of wheeze feature extraction methods for asthma severity levels classification". 2018 9th IEEE Control and System Graduate Research Colloquium, ICSGRC 2018 - Proceeding, (August), 145–150.
Tamas, W., Notton, G., Paoli, C., Nivet, M. L., Voyant, C. 2016. "Hybridization of air quality forecasting models using machine learning and clustering: An original approach to detect pollutant peaks". Aerosol and Air Quality Research, 16(2), 405–416.
Winursito, A., Hidayat, R. and Bejo, A. 2018. "Improvement of MFCC feature extraction accuracy using PCA in Indonesian speech recognition," 2018 International Conference on Information and Communications Technology (ICOIACT), 2018, pp. 379-383.