Abstract
Breast cancer is one of the leading causes of death in women. Mammography is an effective method for early detection of breast cancer. Like other medical images, mammograms demonstrate a certain degree of self-similarity over a range of scales, which can be used in classifying individuals as cancerous or non-cancerous. In this paper, we study the robust estimation of Hurst exponent (self-similarity measure) in two-dimensional images based on non-decimated wavelet transforms (NDWT). The robustness is achieved by applying a general trimean estimator on non-decimated wavelet detail coefficients of the transformed data, and the general trimean estimator is derived as a weighted average of the distribution’s median and quantiles, combining the median’s emphasis on central values with the quantiles’ attention to the extremes. The properties of the proposed estimators are studied both theoretically and numerically. Compared with other standard wavelet-based methods (Veitch and Abry (VA) method, Soltani, Simard, and Boichu (SSB) method, median based estimators MEDL and MEDLA, and Theil-type (TT) weighted regression method), our methods reduce the variance of the estimators and increase the prediction precision in most cases. We apply proposed methods to digitized mammogram images, estimate Hurst exponent, and then use it as a discriminatory descriptor to classify mammograms to benign and malignant. Our methods yield the highest classification accuracy around 65%.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Abry, P. (2003). Scaling and wavelets: An introductory walk. In Processes with long-range correlations (pp. 34–60). Berlin: Springer.
Abry, P., Gonçalvés, P., & Flandrin, P. (1995). Wavelets, spectrum analysis and 1/f processes. In: Wavelets and statistics (pp. 15–29). Springer: New York.
Abry, P., Goncalves, P., & Véhel, J. L. (2013). Scaling, fractals and wavelets. New York: Wiley.
Abry, P., Flandrin, P., Taqqu, M. S., & Veitch, D. (2000). Wavelets for the analysis, estimation and synthesis of scaling data. Self-similar network traffic and performance evaluation (pp. 39–88). New York: Wiley.
Abry, P., Flandrin, P., Taqqu, M. S., & Veitch, D. (2003). Self-similarity and long-range dependence through the wavelet lens. In Theory and applications of long-range dependence (pp. 527–556). Boston, MA: Birkhauser.
Andrews, D. F., & Hampel, F. R. (2015). Robust estimates of location: Survey and advances. Princeton: Princeton University Press.
Bala, B. K., & Audithan, S. (2014). Wavelet and curvelet analysis for the classification of microcalcifiaction using mammogram images. In Second International Conference on Current Trends in Engineering and Technology - ICCTET 2014 (pp. 517–521). https://doi.org/10.1109/ICCTET.2014.6966351
DasGupta, A. (2008). Edgeworth expansions and cumulants. In Asymptotic theory of statistics and probability (pp. 185–201). New York: Springer.
El-Naqa, I., Yang, Y., Wernick, M. N., Galatsanos, N. P., & Nishikawa, R. M. (2002). A support vector machine approach for detection of microcalcifications. IEEE Transactions on Medical Imaging, 21(12), 1552–1563.
Engel, Jr J., Bragin, A., Staba, R., & Mody, I. (2009). High-frequency oscillations: What is normal and what is not? Epilepsia, 50(4), 598–604.
Feng, C., & Vidakovic, B. (2017), Estimation of the hurst exponent using trimean estimators on nondecimated wavelet coefficients. arXiv preprint arXiv:170908775.
Franzke, C. L., Graves, T., Watkins, N. W., Gramacy, R. B., & Hughes, C. (2012). Robustness of estimators of long-range dependence and self-similarity under non-gaussianity. Philosophical Transactions of the Royal Society A, 370(1962), 1250–1267.
Gastwirth, J. L. (1966). On robust procedures. Journal of the American Statistical Association, 61(316), 929–948.
Gastwirth, J. L., & Cohen, M.L (1970) Small sample behavior of some robust linear estimators of location. Journal of the American Statistical Association, 65(330), 946–973
Gastwirth, J. L, & Rubin, H. (1969). On robust linear estimators. The Annals of Mathematical Statistics, 40(1), 24–39.
Gregoriou, G. G., Gotts, S. J., Zhou, H., & Desimone, R. (2009). High-frequency, long-range coupling between prefrontal and visual cortex during attention. Science, 324(5931), 1207–1210.
Hamilton, E. K., Jeon, S., Cobo, P. R., Lee, K. S., & Vidakovic, B. (2011). Diagnostic classification of digital mammograms by wavelet-based spectral tools: A comparative study. In 2011 IEEE International Conference on Bioinformatics and Biomedicine (BIBM), IEEE (pp. 384–389).
Heath, M., Bowyer, K., Kopans, D., Moore, R., & Kegelmeyer, W. P. (2000). The digital database for screening mammography. In Proceedings of the 5th International Workshop on Digital Mammography (pp. 212–218). Medical Physics Publishing.
Jeon, S., Nicolis, O., & Vidakovic, B. (2014). Mammogram diagnostics via 2-d complex wavelet-based self-similarity measures. The São Paulo Journal of Mathematical Sciences, 8(2), 265–284.
Kang, M., & Vidakovic, B. (2017). Medl and medla: Methods for assessment of scaling by medians of log-squared nondecimated wavelet coefficients. ArXiv Preprint ArXiv:170304180.
Katul, G., Vidakovic, B., & Albertson, J. (2001). Estimating global and local scaling exponents in turbulent flows using discrete wavelet transformations. Physics of Fluids, 13(1), 241–250.
Kestener, P., Lina, J. M., Saint-Jean, P., & Arneodo, A. (2011). Wavelet-based multifractal formalism to assist in diagnosis in digitized mammograms. Image Analysis & Stereology, 20(3), 169–174.
Kolmogorov, A. N. (1940). Wienersche spiralen und einige andere interessante kurven in hilbertscen raum, cr (doklady). Academy of Sciences URSS (NS), 26, 115–118.
Mandelbrot, B. B., & Van Ness, J. W. (1968). Fractional Brownian motions, fractional noises and applications. SIAM Review, 10(4), 422–437.
Nason, G. P., & Silverman, B. W. (1995). The stationary wavelet transform and some statistical applications. In Wavelets and statistics (pp. 281–299). New York: Springer.
Netsch, T., & Peitgen, H. O. (1999). Scale-space signatures for the detection of clustered microcalcifications in digital mammograms. IEEE Transactions on Medical Imaging, 18(9), 774–786.
Nicolis, O., Ramírez-Cobo, P., & Vidakovic, B. (2011). 2d wavelet-based spectra with applications. Computational Statistics & Data Analysis, 55(1), 738–751.
Park, J., & Park, C. (2009). Robust estimation of the hurst parameter and selection of an onset scaling. Statistica Sinica, 19, 1531–1555.
Park, K., & Willinger, W. (2000). Self-similar network traffic and performance evaluation. Wiley Online Library. https://doi.org/10.1002/047120644X.
Percival, D. B., & Walden, A. T. (2006). Wavelet methods for time series analysis (vol. 4). New York: Cambridge University Press.
Ramírez-Cobo, P., & Vidakovic, B. (2013). A 2d wavelet-based multiscale approach with applications to the analysis of digital mammograms. Computational Statistics & Data Analysis, 58, 71–81.
Reiss, P. T., & Ogden, R. T. (2010). Functional generalized linear models with images as predictors. Biometrics, 66(1), 61–69.
Reiss, P. T., Ogden, R. T., Mann, J. J., & Parsey, R. V. (2005). Functional logistic regression with pet imaging data: A voxel-level clinical diagnostic tool. Journal of Cerebral Blood Flow & Metabolism, 25(1_suppl), S635–S635.
Shen, H., Zhu. Z., & Lee, T. C. (2007). Robust estimation of the self-similarity parameter in network traffic using wavelet transform. Signal Processing, 87(9), 2111–2124.
Sheng, H., Chen, Y., & Qiu, T. (2011). On the robustness of hurst estimators. IET Signal Processing, 5(2), 209–225.
Soltani, S., Simard, P., & Boichu, D. (2004). Estimation of the self-similarity parameter using the wavelet transform. Signal Processing, 84(1), 117–123.
Theil, H. (1992). A rank-invariant method of linear and polynomial regression analysis. In Henri Theils contributions to economics and econometrics (pp. 345–381). The Netherlands: Springer.
Tukey, J. W. (1977). Exploratory data analysis (Vol. 2). Reading, MA: Addison-Wesley.
Vidakovic, B. (2009). Statistical modeling by wavelets (Vol. 503). New York: Wiley.
Wang, T. C., & Karayiannis, N. B. (1998). Detection of microcalcifications in digital mammograms using wavelets. IEEE Transactions on Medical Imaging, 17(4), 498–509.
Wood, A. T., & Chan, G. (1994). Simulation of stationary gaussian processes in [0, 1] d. Journal of Computational and Graphical Statistics, 3(4), 409–432.
Woods, T., Preeprem, T., Lee, K., Chang, W., & Vidakovic, B. (2016). Characterizing exons and introns by regularity of nucleotide strings. Biology Direct, 11(1), 6.
Zhou, B. (1996). High-frequency data and volatility in foreign-exchange rates. Journal of Business and Economic Statistics, 14(1), 45–52.
Zhou, H., Li, L., & Zhu, H. (2013). Tensor regression with applications in neuroimaging data analysis. Journal of the American Statistical Association, 108(502), 540–552.
Zipunnikov, V., Caffo, B., Yousem, D. M., Davatzikos, C., Schwartz, B. S., Crainiceanu. C. (2011). Functional principal component model for high-dimensional brain imaging. NeuroImage, 58(3), 772–784.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Appendix
Appendix
1.1 Proof of Theorem 5.1
Proof
A single wavelet coefficient in a non-decimated wavelet transform of a 2-D fBm of size N × N with Hurst exponent H is normally distributed, with variance depending on its level j. The four coefficients in each set
are assumed to be independent and follow the same normal distribution.
Then the mid-energies in D j defined in (5.9) and (5.8) can be readily shown to have exponential distribution with scale parameter \(\lambda _j=\sigma ^2\cdot 2^{-\left (2H+2\right )j}\). Therefore at each detail level j, the mid-energies in D j are i.i.d. \( \mathcal {E}xp\left (\lambda _j^{-1}\right )\), and when applying general trimean estimator \(\hat {\mu }_{j}\) on D j, following the derivation in Sect. 5.3, we have
and
therefore, the asymptotic distribution of \(\hat {\mu }_{j,i}\) is normal with mean
and variance
Since the Hurst exponent can be estimated as
where \(\hat {\beta }\) is the regression slope in the least square linear regression on pairs \(\left (j, \log _2\left (\hat {\mu }_{j}\right )\right )\) from level J 1 to J 2, J 1 ≤ j ≤ J 2. It can be easily derived that \(\hat {\beta }\) is a linear combination of \(\log _2\left (\hat {\mu }_{j}\right )\),
We can check that \(\sum _{j=J_1}^{J_2}a_j=0\) and \(\sum _{j=J_1}^{J_2}a_j j=1\). Also, if \(X\sim \mathcal {N}(\mu , \sigma ^2)\), the approximate expectation and variance of g(X) are
based on which we calculate
Therefore
and
where the asymptotic variance V 1 is a constant number independent of simple size N and level j,
and
1.2 Proof of Theorem 5.2
Proof
We have stated that each mid-energy in D j follows \(\mathcal {E}xp\left (\lambda _j^{-1}\right )\) with scale parameter \(\lambda _j=\sigma ^2\cdot 2^{-\left (2H+2\right )j}\). If we denote the kth element in \(\log \left (D_{j}\right )\) as y j,k for \(k=1,\ldots ,\frac {M^2}{2}\) and j = 1, …, J, the pdf and cdf of y j,k are
and
The p-quantile can be obtained by solving \(F\left (y_p\right )=1-e^{-\lambda _j^{-1}e^{y_p}}=p\), and \(y_p=\log \left (-\lambda _j\log \left (1-p\right )\right )\). Then it can be shown that \(f\left (y_p\right )=-\left (1-p\right )\log \left (1-p\right )\). When applying the general trimean estimator \(\hat {\mu }_{j}\) on \(\log \left (D_{j}\right )\), following the derivation in Sect. 5.3, we get
and
thus, the asymptotic distribution of \(\hat {\mu }_{j,i}\) is normal with mean
and variance
where
and
Since the Hurst exponent can be estimated as
where \(\hat {\beta }\) is the regression slope in the least square linear regressions on pairs \(\left (j, \hat {\mu }_{j}\right )\) from level J 1 to J 2, J 1 ≤ j ≤ J 2. It can be easily derived that \(\hat {\beta }\) is a linear combination of \(\hat {\mu }_{j}\),
Again, we can check that \(\sum _{j=J_1}^{J_2}a_j=0\) and \(\sum _{j=J_1}^{J_2}a_j j=1\). Therefore
and
where the asymptotic variance V 2 is a constant number independent of simple size N and level j,
and q(J 1, J 2) is given in Eq. (5.13).
1.3 Proof of Lemma 5.2
Proof
When applying Tukey’s trimean estimator \(\hat {\mu }_{j}^T\) on D j, following the derivation in Sect. 5.3.1, we have
and
therefore, the asymptotic distribution of \(\hat {\mu }_{j}^T\) is normal with mean
and variance
When applying Gastwirth estimator \(\hat {\mu }_{j}^G\) on D j, following the derivation in Sect. 5.3.2, we have
and
therefore, the asymptotic distribution of \(\hat {\mu }_{j}^G\) is normal with mean
and variance
Based on Eq. (5.28), we have
where the asymptotic variances \(V^T_1\) and \(V^G_1\) are constant numbers,
The function q(J 1, J 2) is the same as Eq. (5.13) in Theorem 5.1.
1.4 Proof of Lemma 5.3
Proof
When applying Tukey’s trimean estimator \(\hat {\mu }_{j}^T\) on \(\log \left (D_{j}\right )\), following the derivation in Sect. 5.3.1, we have
and
therefore, the asymptotic distribution of \(\hat {\mu }_{j}^T\) is normal with mean
and variance
When applying Gastwirth estimator \(\hat {\mu }_{j}^G\) on \(\log \left (D_{j,i}\right )\), following the derivation in Sect. 5.3.2, we have
and
therefore, the asymptotic distribution of \(\hat {\mu }_{j}^G\) is normal with mean
and variance
Based on Eq. (5.30), we can easily derive
where the asymptotic variances \(V^T_2\) and \(V^G_2\) are constant numbers,
The function q(J 1, J 2) is provided in Eq. (5.13).
Rights and permissions
Copyright information
© 2018 Springer Nature Switzerland AG
About this chapter
Cite this chapter
Feng, C., Mei, Y., Vidakovic, B. (2018). Mammogram Diagnostics Using Robust Wavelet-Based Estimator of Hurst Exponent. In: Zhao, Y., Chen, DG. (eds) New Frontiers of Biostatistics and Bioinformatics. ICSA Book Series in Statistics. Springer, Cham. https://doi.org/10.1007/978-3-319-99389-8_5
Download citation
DOI: https://doi.org/10.1007/978-3-319-99389-8_5
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-99388-1
Online ISBN: 978-3-319-99389-8
eBook Packages: Mathematics and StatisticsMathematics and Statistics (R0)