Journal of radiological science and technology (대한방사선기술학회지:방사선기술과학)

Korean Society of Radiological Science (대한방사선과학회)
권호 표지
DOI QR코드

DOI QR Code

Retrospective Analysis of Cytopathology using Gray Level Co-occurrence Matrix Algorithm for Thyroid Malignant Nodules in the Ultrasound Imaging

갑상샘 악성결절의 초음파영상에서 GLCM 알고리즘을 이용한 세포병리 진단의 후향적 분석

  • Kim, Yeong-Ju (Dept. of Radiology, Inje University Haeundae Paik Hospital) ;
  • Lee, Jin-Soo (Dept. of Radiology, Inje University Haeundae Paik Hospital) ;
  • Kang, Se-Sik (Dept. of Radiological Science, College of Health Sciences, Catholic University of Pusan) ;
  • Kim, Changsoo (Dept. of Radiological Science, College of Health Sciences, Catholic University of Pusan)
  • 김영주 (인제대학교 해운대백병원 영상의학과) ;
  • 이진수 (인제대학교 해운대백병원 영상의학과) ;
  • 강세식 (부산가톨릭대학교 보건과학대학 방사선학과) ;
  • 김창수 (부산가톨릭대학교 보건과학대학 방사선학과)
  • Received : 2017.05.18
  • Accepted : 2017.06.21
  • Published : 2017.06.30
Download PDF
⟨ Previous Next ⟩

Abstract

This study evaluated the applicability of computer-aided diagnosis by retrospective analysis of GLCM algorithm based on cytopathological diagnosis of normal and malignant nodules in thyroid ultrasound images. In the experiment, the recognition rate and ROC curve of thyroid malignant nodule were analyzed using 6 parameters of GLCM algorithm. Experimental results showed 97% energy, 93% contrast, 92% correlation, 92% homogeneity, 100% entropy and 100% variance. Statistical analysis showed that the area under the curve of each parameter was more than 0.947 (p = 0.001) in the ROC curve, which was significant in the recognition of thyroid malignant nodules. In the GLCM, the cut-off value of each parameter can be used to predict the disease through analysis of quantitative computer-aided diagnosis.

본 연구는 갑상샘 초음파 영상에서 정상 및 악성결절의 세포병리 진단결과를 바탕으로 GLCM 알고리즘분석을 통한 후향적 연구를 시행하여 컴퓨터보조진단의 적용 가능성을 평가하였다. GLCM 알고리즘의 6가지 파라미터를 이용한 갑상샘 악성결절의 인식률 평가와 ROC 곡선을 분석하였다. 실험 결과는 에너지 97%, 대조도 93%, 상관관계 92%, 동질성 92%, 엔트로피 100%, 분산 100%의 높은 질환인식률을 나타내었다. ROC 곡선 분석에서 각 파라미터의 곡선아래면적이 0.947(p=0.001) 이상을 나타내어 갑상샘 악성결절의 인식에 의미가 있는 결과로 나타났다. 또한 GLCM에서 각 파라미터의 cut-off값 결정으로 정량적인 컴퓨터보조진단의 분석을 통한 질환예측이 가능할 것으로 판단된다.

Keywords

References

  1. Singh N, Jindal A: A Segmentation Method and Comparison of Classification Methods for Thyroid Ultrasound Images, International Journal of Computer Applications. 50(11), 43-49, 2012 https://doi.org/10.5120/7818-1115
  2. Stamos K, Eystratios G K, Dimitris M: A Contourlet Transform Feature Extraction Scheme for Ultrasound Thyroid Texture Classification. Engineering Intelligent Systems Special issue: Artificial Intelligence Applications and Innovations, 18(3), 2010
  3. Tan GH, Gharib H: Thyroid incidentalomas ; management approaches nodules discovered incidentally on thyroid imaging, Annals of Internal Medicine, 126(3), 226-231, 1997 https://doi.org/10.7326/0003-4819-126-3-199702010-00009
  4. Ross DS: Neopalpable thyroid nodules ; managing an epidemic. Journal of Clinical Endocrinology & Metabolism, 87(5), 1938-1940, 2002 https://doi.org/10.1210/jc.87.5.1938
  5. Hegedus L: Clinical practice ; The thyroid nodule. New England Journal of Medicine, 351, 1764-1771, 2004 https://doi.org/10.1056/NEJMcp031436
  6. Gharib H, Papini E: Thyroid nodule ; clinical importance, assessment, and treatment, Endocrinology Metabolism Clinics of North America, 36, 707-735, 2007 https://doi.org/10.1016/j.ecl.2007.04.009
  7. Marqusee E, Benson CB, Frates MC, Doubilet PM, Larsen PR: Usefulness of ultrasonography in the management of nodular thyroid disease, Annals of Internal Medicine, 133, 696-700, 2000 https://doi.org/10.7326/0003-4819-133-9-200011070-00011
  8. Cho YH: A Performance Improvement of GLCM Based on Nonuniform Quantization Method ; Journal of Korean Institute of Intelligent Systems, 25(2), 133-138, 2015 https://doi.org/10.5391/JKIIS.2015.25.2.133
  9. Gireesha HM, Nanda S: Thyroid Nodule Segmentation and Classification in Ultrasound Images, International Journal of Engineering Research & Technology, 3(5), 2252-2256, 2014
  10. Shrikant D. Kale, Krushil M. Punwatkar: Texture Analysis of Ultrasound Medical Images for Diagnosis of Thyroid Nodule Using Support Vector Machine, International Journal of Computer Science and Mobile Computing, 2(10), 71-77, 2013
  11. Nitish Z, Vrushsen P: GLCM Textural Features for Brain Tumor Classification, International Journal of Computer Science, 9(3), 2012
  12. Poonguzhali S, Ravindran G: Automatic Classification of Focal Lesions in Ultrasound Liver Image Using Combined Texture Features, Information Technology Journal, 7(1), 205-209, 2008 https://doi.org/10.3923/itj.2008.205.209
  13. Bag BE: Texture analysis of thyroid nodules in ultrasound image for computer aided diagnostic system, Graduate school of Yonsei University, 2016
  14. Cho JY, Kim CS, Kang SS, Ko SJ, Ye SY: Image Analysis of Computer Aided Diagnosis using Gray Level Co-occurrence Matrix in the Ultrasonography for Benign Prostate Hyperplasia. Journal Of The Korea Contents Association., 15( 3), 184-191, 2015
  15. Park YO, Jung EK, Park YJ, Nam SR, Jung JY,, Kim HJ: The Evaluation of CR and DDR chest image using ROC analysis. Journal of the Korean Society of Radiology, 1(1), 25-30, 2007