Abstract
An automated computer-aided diagnosis system is developed to classify benign and malignant thyroid nodules using multi-stained fine needle aspiration biopsy (FNAB) cytological images. In the first phase, the image segmentation is performed to remove the background staining information and retain the appropriate foreground cell objects in cytological images using mathematical morphology and watershed transform segmentation methods. Subsequently, statistical features are extracted using two-level discrete wavelet transform (DWT) decomposition, gray level co-occurrence matrix (GLCM) and Gabor filter based methods. The classifiers k-nearest neighbor (k-NN), Elman neural network (ENN) and support vector machine (SVM) are tested for classifying benign and malignant thyroid nodules. The combination of watershed segmentation, GLCM features and k-NN classifier results a lowest diagnostic accuracy of 60 %. The highest diagnostic accuracy of 93.33 % is achieved by ENN classifier trained with the statistical features extracted by Gabor filter bank from the images segmented by morphology and watershed transform segmentation methods. It is also observed that SVM classifier results its highest diagnostic accuracy of 90 % for DWT and Gabor filter based features along with morphology and watershed transform segmentation methods. The experimental results suggest that the developed system with multi-stained thyroid FNAB images would be useful for identifying thyroid cancer irrespective of staining protocol used.
Similar content being viewed by others
References
Keleş A, Keleş A (2008) ESTDD: expert system for thyroid diseases diagnosis. Expert Syst Appl 34:242–246
Tsantis S, Dimitropoulos N, Cavouras D, Nikiforidis G (2009) Morphological and wavelet features towards sonographic thyroid nodules evaluation. Comput Med Imaging Graph 33:91–99
Mundasad B, Mcallister I, Carson J, Pyper PC (2006) Accuracy of fine needle aspiration cytology in diagnosis of thyroid swellings. Internet J Endocrinol 2(2). doi:10.5580/484
Karakitsos P, Cochand-Priollet B, Pouliakis A, Guillausseau PJ, Ioakim-Liossi A (1999) Learning vector quantizer in the investigation of thyroid lesions. Anal Quant Cytol Histol 21:201–208
Würflinger T, Stockhausen J, Meyer-Ebrecht D, Böcking A (2003) Robust automatic coregistration, segmentation and classification of cell nuclei in multimodal cytopathological microscopic images. Comput Med Imaging Graph 28:87–98
Lezoray O, Lecluse M (2007) Automatic segmentation and classification of cells from Bronchoalveolar lavage. Image Anal Stereol 26:111–119
Shapiro NA, Poloz TL, Shkurupij VA, Tarkov MS, Poloz VV, Demin AV (2007) Application of artificial neural network for classification of thyroid follicular tumors. Anal Quant Cytol Histol 29:87–94
Wang X, Li S, Liu H, Wood M, Chen WR, Zheng B (2008) Automated identification of analyzable metaphase chromosomes depicted on microscopic digital images. J Biomed Inform 41:264–271
Daskalakis A, Kostopoulos S, Spyridonos P, Glotsos D, Ravazoula P, Kardari M, Kalatzis I, Cavouras D, Nikiforidis G (2008) Design of a multi-classifier system for discriminating benign from malignant thyroid nodules using routinely H&E-stained cytological images. Comput Biol Med 38:196–203
Gopinath B, Gupta BR (2010) Classification of thyroid carcinoma in FNAB cytological microscopic images. Int J Healthc Inf Syst Inform 5:60–72
Gopinath B, Gupta BR (2010) Majority voting based classification of thyroid carcinoma. Procedia Comput Sci 2:265–271
Muthu Rama Krishnan M, Shah P, Chakraborty C, Ray AK (2012) Statistical analysis of textural features for improved classification of oral histopathological images. J Med Syst 36:865–881
Kelley DJ, Terris DJ, Talavera F, Kass E, Slack CL (2012) Thyroid, Papillary Carcinoma, Early. Medscape Reference. http://emedicine.medscape.com/article/849000. Accessed 31 January 2013
Gopinath B, Shanthi N (2012) Automated segmentation of ELA cancer cells in microscopic images for evaluating the cytotoxic effect of selected medicinal plants. J Med Biol Eng 32:279–286
Kimori Y (2011) Mathematical morphology-based approach to the enhancement of morphological features in medical images. J Clin Bioinforma 1:33
Hrebien M, Stec P, Nieczkowski T, Obuchowicz A (2008) Segmentation of breast cancer fine needle biopsy cytological images. Int J Appl Math Comput 18:159–170
Kimori Y, Baba N, Morone N (2010) Extended morphological processing: a practical method for automatic spot detection of biological markers from microscopic images. BMC Bioinformatics 11:373
Petushi S, Garcia FU, Haber MM, Katsinis C, Tozeren A (2006) Large-scale computations on histology images reveal grade-differentiating parameters for breast cancer. BMC Med Imaging 6:14
Kavitha MS, Asano A, Taguchi A, Kurita T, Sanada M (2012) Diagnosis of osteoporosis from dental panoramic radiographs using the support vector machine method in a computer-aided system. BMC Med Imaging 12:1
Otsu N (1979) A threshold selection method from gray-level histograms. IEEE Trans Syst Man Cybern 9:62–66
Sadeghian F, Seman Z, Ramli AR, Kahar BHA, Saripan M (2009) A framework for white blood cell segmentation in microscopic blood images using digital image processing. Biol Proced Online 11:196–206
Julesz B, Gilbert EN, Shepp LA, Frisch HL (1973) Inability of humans to discriminate between visual textures that agree in second-order statistics revisited. Perception 2:391–405
Arivazhagan S, Ganesan L (2003) Texture classification using wavelet transform. Pattern Recogn Lett 24:1513–1521
Abi-Abdallah D, Chauvet E, Bouchet-Fakri L, Bataillard A, Briguet A, Fokapu O (2006) Reference signal extraction from corrupted ECG using wavelet decomposition for MRI sequence triggering: application to small animals. Biomed Eng Online 5:11
Liu T, Zhang L, Li P, Lin H (2012) Remotely sensed image retrieval based on region-level semantic mining. EURASIP J Image Video Process 2012:4
Zhang B, Pham TD (2011) Phenotype recognition with combined features and random subspace classifier ensemble. BMC Bioinformatics 12:128
Walkowski S, Szymas J (2011) Quality evaluation of virtual slides using methods based on comparing common image areas. Diagn Pathol 6:S14
Haralick RM, Shanmugam K, Dinstein I (1973) Textural features for image classification. IEEE Trans Syst Man Cybern 3:610–621
Linder N, Konsti J, Turkki R, Rahtu E, Lundin M, Nordling S, Haglund C, Ahonen T, Pietikäinen M, Lundin J (2012) Identification of tumor epithelium and stroma in tissue microarrays using texture analysis. Diagn Pathol 7:22
Bovik A, Clark M, Geisler WS (1990) Multichannel texture analysis using localized spatial filters. IEEE Trans Pattern Anal Mach Intell 12:55–73
Hua J, Xiong Z, Lowey J, Suh E, Dougherty ER (2005) Optimal number of features as a function of sample size for various classification rules. Bioinformatics 21:1509–1515
Tahir MA, Bouridane A, Kurugollu F (2007) Simultaneous feature selection and feature weighting using hybrid Tabu search/K-nearest neighbor classifier. Pattern Recogn Lett 28:438–446
Elman JL (1990) Finding structure in time. Cogn Sci 14:179–211
Song Q (2010) On the weight convergence of Elman networks. IEEE Trans Neural Netw 21:463–480
Thammano A, Ruxpakawong P (2010) Nonlinear dynamic system identification using recurrent neural network with multi-segment piecewise-linear connection weight. Memetic Comp 2:273–282
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Gopinath, B., Shanthi, N. Computer-aided diagnosis system for classifying benign and malignant thyroid nodules in multi-stained FNAB cytological images. Australas Phys Eng Sci Med 36, 219–230 (2013). https://doi.org/10.1007/s13246-013-0199-8
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s13246-013-0199-8