A computer-aided system for malignancy risk assessment of nodules in thyroid US images based on boundary features

In this paper, a novel computer-based approach is proposed for malignancy risk assessment of thyroid nodules in ultrasound images. The proposed approach is based on boundary features and is motivated by the correlation which has been addressed in medical literature between nodule boundary irregularity and malignancy risk. In addition, local echogenicity variance is utilized so as to incorporate information associated with local echogenicity distribution within nodule boundary neighborhood. Such information is valuable for the discrimination of high-risk nodules with blurred boundaries from medium-risk nodules with regular boundaries. Analysis of variance is performed, indicating that each boundary feature under study provides statistically significant information for the discrimination of thyroid nodules in ultrasound images, in terms of malignancy risk. k-nearest neighbor and support vector machine classifiers are employed for the classification tasks, utilizing feature vectors derived from all combinations of features under study. The classification results are evaluated with the use of the receiver operating characteristic. It is derived that the proposed approach is capable of discriminating between medium-risk and high-risk nodules, obtaining an area under curve, which reaches 0.95.

[1]  Y. Chou,et al.  Stepwise logistic regression analysis of tumor contour features for breast ultrasound diagnosis. , 2001, Ultrasound in Medicine and Biology.

[2]  R. Jeffrey,et al.  Management of thyroid nodules detected at US: Society of Radiologists in Ultrasound consensus conference statement. , 2005, Ultrasound quarterly.

[3]  Anil K. Jain,et al.  Statistical Pattern Recognition: A Review , 2000, IEEE Trans. Pattern Anal. Mach. Intell..

[4]  D. Cavouras,et al.  Development of a support vector machine-based image analysis system for assessing the thyroid nodule malignancy risk on ultrasound. , 2005, Ultrasound in medicine & biology.

[5]  K Nakamura,et al.  Computerized analysis of the likelihood of malignancy in solitary pulmonary nodules with use of artificial neural networks. , 2000, Radiology.

[6]  P. Larsen,et al.  Prevalence and distribution of carcinoma in patients with solitary and multiple thyroid nodules on sonography. , 2006, The Journal of clinical endocrinology and metabolism.

[7]  Borut Zalik,et al.  An efficient chain code with Huffman coding , 2005, Pattern Recognit..

[8]  G. Mailloux,et al.  Computer analysis of echographic textures in hashimoto disease of the thyroid , 1986, Journal of clinical ultrasound : JCU.

[9]  Nikos Dimitropoulos,et al.  Variable Background Active Contour Model for Computer-Aided Delineation of Nodules in Thyroid Ultrasound Images , 2007, IEEE Transactions on Information Technology in Biomedicine.

[10]  M. Kallergi.  Computer-aided diagnosis of mammographic microcalcification clusters. , 2004, Medical physics.

[11]  David Casasent.  Intelligent Robots and Computer Vision XIV: Algorithms, Techniques, Active Vision, and Materials Handling , 1995 .

[12]  Mikhail J. Atallah,et al.  On Symmetry Detection , 1985, IEEE Transactions on Computers.

[13]  Ari Visa,et al.  Shape recognition of irregular objects , 1996, Other Conferences.

[14]  B. Ripley,et al.  Pattern Recognition , 1968, Nature.

[15]  Guojun Lu,et al.  Review of shape representation and description techniques , 2004, Pattern Recognit..

[16]  Frank Y. Shih,et al.  A one-pass algorithm for local symmetry of contours from chain codes , 1999, Pattern Recognit..

[17]  J. J. Janick,et al.  AACE Clinical Practice Guidelines for the Diagnosis and Management of Thyroid Nodules , 1999 .

[18]  Heinz-Otto Peitgen,et al.  The science of fractal images , 2011 .

[19]  G. van Kaick,et al.  Quantification and classification of echographic findings in the thyroid gland by computerized B-mode texture analysis. , 1989, European journal of radiology.

[20]  W J Lorenz,et al.  [The value of computer-assisted sonographic tissue characterization in focal lesions of the thyroid]. , 1989, Der Radiologe.

[21]  Primo Zingaretti,et al.  Fast Chain Coding of Region Boundaries , 1998, IEEE Trans. Pattern Anal. Mach. Intell..

[22]  Rangaraj M. Rangayyan,et al.  Fractal Analysis of Contours of Breast Masses in Mammograms , 2007, Journal of Digital Imaging.

[23]  Andrew P. Bradley,et al.  The use of the area under the ROC curve in the evaluation of machine learning algorithms , 1997, Pattern Recognit..

[24]  Herbert Freeman,et al.  On the Encoding of Arbitrary Geometric Configurations , 1961, IRE Trans. Electron. Comput..

[25]  G. Medeiros-Neto,et al.  Combined ultrasonographic and cytological studies in the diagnosis of thyroid nodules. , 1999, Biochimie.

[26]  Chih-Jen Lin,et al.  LIBSVM: A library for support vector machines , 2011, TIST.

[27]  Rafael C. González,et al.  Digital image processing, 3rd Edition , 2008 .

[28]  Huihua Kenny Chiang,et al.  Stepwise logistic regression analysis of tumor contour features for breast ultrasound diagnosis , 2001, 2001 IEEE Ultrasonics Symposium. Proceedings. An International Symposium (Cat. No.01CH37263).

[29]  Yuan Yan Tang,et al.  Feature extraction using wavelet and fractal , 2001, Pattern Recognit. Lett..

[30]  Sergios Theodoridis,et al.  Pattern Recognition, Third Edition , 2006 .

[31]  Paul Wintz,et al.  Digital image processing (2nd ed.) , 1987 .

[32]  Rangaraj M. Rangayyan,et al.  Measures of acutance and shape for classification of breast tumors , 1997, IEEE Transactions on Medical Imaging.

[33]  Nikos Dimitropoulos,et al.  Computational Characterization of Thyroid Tissue in the Radon Domain , 2007, Twentieth IEEE International Symposium on Computer-Based Medical Systems (CBMS'07).

[34]  Sven Loncaric,et al.  A survey of shape analysis techniques , 1998, Pattern Recognit..

[35]  F. Kantarcı,et al.  Concurrent routine breast and thyroid sonography for detection of thyroid tumors. , 2006, AJR. American journal of roentgenology.

[36]  Nikos Dimitropoulos,et al.  Multi-scaled morphological features for the characterization of mammographic masses using statistical classification schemes , 2007, Artif. Intell. Medicine.

[37]  Emmanuel Ifeachor,et al.  SAMPLE SIZE DETERMINATION USING ROC ANALYSIS , 2007 .

[38]  H Yamashita,et al.  Ultrasonographic characteristics of thyroid nodules: prediction of malignancy. , 2001, Archives of surgery.

[39]  Bernhard Schölkopf,et al.  Support vector learning , 1997 .