Feature Extraction From Parametric Time–Frequency Representations for Heart Murmur Detection
- 443 Downloads
The detection of murmurs from phonocardiographic recordings is an interesting problem that has been addressed before using a wide variety of techniques. In this context, this article explores the capabilities of an enhanced time–frequency representation (TFR) based on a time-varying autoregressive model. The parametric technique is used to compute the TFR of the signal, which serves as a complete characterization of the process. Parametric TFRs contain a large quantity of data, including redundant and irrelevant information. In order to extract the most relevant features from TFRs, two specific approaches for dimensionality reduction are presented: feature extraction by linear decomposition, and tiling partition of the t–f plane. In the first approach, the feature extraction was carried out by means of eigenplane-based PCA and PLS techniques. Likewise, a regular partition and a refined Quadtree partition of the t–f plane were tested for the tiled-TFR approach. As a result, the feature extraction methodology presented, which searches for the most relevant information immersed on the TFR, has demonstrated to be very effective. The features extracted were used to feed a simple k-nn classifier. The experiments were carried out using 45 phonocardiographic recordings (26 normal and 19 records with murmurs), segmented to extract 548 representative individual beats. The results using these methods point out that better accuracy and flexibility can be accomplished to represent non-stationary PCG signals, showing evidences of improvement with respect to other approaches found in the literature. The best accuracy obtained was 99.06 ± 0.06%, evidencing high performance and stability. Because of its effectiveness and simplicity of implementation, the proposed methodology can be used as a simple diagnostic tool for primary health-care purposes.
KeywordsHeart sounds Feature extraction Time–frequency representation Time-varying autoregressive model Murmur detection
Bayesian information criterion
Continuous wavelet transform
Principal component analysis
Partial least squares
The authors would like to acknowledge Dr. Ana Maria Matijasevic and Dr. Guillermo Agudelo who are working with Universidad de Caldas for organizing the acquisition of the PCG data. This research was carried out under grants: “Centro de Investigación e Innovación de Excelencia ARTICA,” funded by COLCIENCIAS; and TEC2006-12887-C02 from the Ministry of Science and Technology of Spain.
- 3.Avendano-Valencia, D., F. Martinez-Tabares, D. Acosta-Medina, I. Godino-Llorente, and G. Castellanos-Dominguez. TFR-based feature extraction using PCA Approaches for discrimination of heart murmurs. Proceedings of the 31th IEEE EMBS Annual International Conference (EMBC’09), 2009.Google Scholar
- 9.Deng, J., J. Yao, J. Dewald, and P. Julius. Classification of the intention to generate a shoulder versus elbow torque by means of a time frequency synthesized spatial patterns BCI algorithm. J. Neural Eng. 2(4):131–138, 2005.Google Scholar
- 13.Kaipio, J., and M. Juntunen. Deterministic regression smoothness priors TVAR modelling. Proc. IEEE ICASSP 99, 1999, 1693–1696.Google Scholar
- 14.Kanai, H., N. Chubachi, and Y. Koiwa. A time-varying AR modeling of heart wall vibration. In: Proceedings on International Conference of the Acoustics, Speech, and Signal Processing, ICASSP 95, edited by EEE Computer Society, 1995, pp. 941–944.Google Scholar
- 16.Nandagopal, D., J. Mazumbar, and R. Bogner. Spectral analysis of second heart sound in children by selective linear prediction coding. Med. Biol. Eng. Comput. 22:229–239, 1985.Google Scholar