Nuclear Medicine and Molecular Imaging

The Korean Society of Nuclear Medicine (대한핵의학회)
권호 표지

Diagnostic Performance of Combined Single Photon Emission Computed Tomographic Scintimammography and Ultrasonography Based on Computer-Aided Diagnosis for Breast Cancer

유방 SPECT 및 초음파 컴퓨터진단시스템 결합의 유방암 진단성능

  • Hwang, Kyung-Hoon (Department of Nuclear Medicine, Gachon University Gil Medical Center) ;
  • Lee, Jun-Gu (Department of Radiology, Seoul National University, College of Medicine) ;
  • Kim, Jong-Hyo (Department of Radiology, Seoul National University, College of Medicine) ;
  • Lee, Hyung-Ji (CAD Impact, Inc.) ;
  • Om, Kyong-Sik (CAD Impact, Inc.) ;
  • Lee, Byeong-Il (Department of Nuclear Medicine, Chonnam National University Hospital) ;
  • Choi, Duck-Joo (Department of Internal Medicine, Gachon University Gil Medical Center) ;
  • Choe, Won-Sick (Department of Nuclear Medicine, Gachon University Gil Medical Center)
  • 황경훈 (가천의과학대학교 길병원 핵의학과) ;
  • 이준구 (서울대학교 의과대학 방사선과) ;
  • 김종효 (서울대학교 의과대학 방사선과) ;
  • 이형지 (캐드임팩트(주)) ;
  • 엄경식 (캐드임팩트(주)) ;
  • 이병일 (전남대학교병원 핵의학과) ;
  • 최덕주 (가천의과학대학교 길병원 내과) ;
  • 최원식 (가천의과학대학교 길병원 핵의학과)
  • Published : 2007.06.30
Download PDF
⟨ Previous Next ⟩

Abstract

Purpose: We investigated whether the diagnostic performance of SPECT scintimammography (SMM) can be improved by adding computer-aided diagnosis (CAD) of ultrasonography (US). Materials and methods: We reviewed breast SPECT SMM images and corresponding US images from 40 patients with breast masses (21 malignant and 19 benign tumors). The quantitative data of SPECT SMM were obtained as the uptake ratio of lesion to contralateral normal breast. The morphologic features of the breast lesions on US were extracted and quantitated using the automated CAD software program. The diagnostic performance of SPECT SMM and CAD of US alone was determined using receiver operating characteristic (ROC) curve analysis. The best discriminating parameter (D-value) combining SPECT SMM and the CAD of US was created. The sensitivity, specificity and accuracy of combined two diagnostic modalities were compared to those of a single one. Results: Both SPECT SMM and CAD of US showed a relatively good diagnostic performance (area under curve = 0.846 and 0.831, respectively). Combining the results of SPECT SMM and CAD of US resulted in improved diagnostic performance (area under curve =0.860), but there was no statistical differerence in sensitivity, specificity and accuracy between the combined method and a single modality. Conclusion: It seems that combining the results of SPECT SMM and CAD of breast US do not significantly improve the diagnostic performance for diagnosis of breast cancer, compared with that of SPECT SMM alone. However, SPECT SMM and CAD of US may complement each other in differential diagnosis of breast cancer.

목적: 유방암의 감별진단에서 기존의 유방 초음파 검사나 핵의학 유방SPECT의 진단성능에는 한계가 있다. 저자들은 초음파 컴퓨터진단시스템(CAD: computer aided diagnosis)의 적용에 의하여 유방 SPECT의 진단성능이 향상되는지를 알아보았다. 대상 및 방법: 유방초음파 및 유방 SPECT(Tc-99m tetrofosmin)를 시행하고 수술후 확진된 여자환자 40명(21명:악성종양, 19명:양성병변)의 영상자료를 분석하였다. 유방초음파영상을 컴퓨터분석 소프트웨어를 이용하여 병변의 경계를 분리한 후, 영상의 형태학적 특성들을 추출하였다. 초음파영상에서 추출된 형태학적 특성 중에서 감별능력이 있는 것으로 판단된 특성들을 골라 정량화하였다. 정량화된 형태학적 특성값들을 유방SPECT에서 구한 병변 대 반대측 유방의 방사능비와 판별분석에 의하여 결합하여 새로운 파라메터인 D-수치를 산출하였다. 유방SPECT의 병변 방사능비, 유방초음파 컴퓨터진단시스템의 악성확률 및 두가지를 결합한 D-수치에 대하여 수신자판단특성곡선(ROC curve) 분석을 이용하여 최적 판별 수치(cut-off value)를 구하고 이에 의한 유방암 진단의 예민도, 특이도 및 정확도를 계산하여 유방 SPECT과 초음파 컴퓨터진단시스템의 결합에 의한 진단성능을 기존의 유방 SPECT의 진단성능과 비교하였다. 결과: ROC curve분석상에서 유방암 진단에 대한 성능은 유방초음파의 컴퓨터 분석시스템 및 유방SPECT 각각 모두 우수하였다(area under curve=0.831 and 0.846). 두 결과를 통계적인 방법으로 결합하였을 때 ROC curve분석의 area under curve는(0.860) 향상되었으나, 최적 판별 수치(cut-off value)에 의한 유방암 진단의 예민도, 특이도 및 정확도에는 통계적인 차이는 없었다. 결론: 유방초음파의 컴퓨터분석시스템의 결과를 유방 SPECT에 적용하여 유방암의 진단성능을 향상시킬 수 있었지만 통계적으로는 유의하지 못하였다. 향후 추가적인 연구가 필요할 것으로 보인다.

Keywords

References

  1. Lim HW. National Cancer Screening Project in Korea. J Korean Society Breast Screening 2005;2:446
  2. Stewart SL, King JB, Thompson TD, Friedman C, Wingo PA. Cancer mortality surveillanceUnited States, 1990-2000. MMWR Surveill Summ 2004;53:1-108
  3. Kopans DB. The positive predictive value of mammography. AJR 1992;158:521-6 https://doi.org/10.2214/ajr.158.3.1310825
  4. Taillefer R. Breast scintigraphy. In: Ell PJ and Gambhir SS, editors. Nuclear Medicine in Clinical Diagnosis and Treatment. 3rd ed.. Edinburgh: Churchill Livingstone ; 2004. p. 17594
  5. Chen DR, Chang RF, Huang YL. Computer-aided diagnosis pplied to US of solid breast nodules by using neural networks. Radiology 1999;213:407-12 https://doi.org/10.1148/radiology.213.2.r99nv13407
  6. Drukker K, Giger ML, Vybornyl CJ, Mendelson EB. Computerized detection and classification of cancer on breast ultrasound. Acad Radiol 2004;11:526-35 https://doi.org/10.1016/S1076-6332(03)00723-2
  7. Lee YH, Fam AT. An edge gradient enhancing adaptive order statistic filter. IEEE Trans Acoust Speech Signal Processing 1987;5:680-95
  8. Ostu N. A threshold selection method from gray-level histogram. IEEE Trans Syst Man Cybern 1979;9:62-6 https://doi.org/10.1109/TSMC.1979.4310076
  9. Hu MK. Visual Pattern Recognition by Moment Invariants. IRE Trans on Information Theory 1962;2:179-87
  10. Om KS, Moon WK. Method of searching nodule image independent of contrast and brightness and of determining malignancy or benignancy. Patent number 0461264, 2004
  11. Jackson VP. Management of solid breast nodules; What is the role of sonography? Radiology 1995;196;14-5 https://doi.org/10.1148/radiology.196.1.7784557
  12. Erickson BJ, Bartholmai B. Computer-aided detection and diagnosis at the start of the third millennium. J Dig Imaging 2002;15:59-68 https://doi.org/10.1007/s10278-002-0011-x
  13. Summers RM. Roadmaps for advancement of radiologic computeraided detection in the 21st century. Radiology 2003;229:11-13 https://doi.org/10.1148/radiol.2291030010
  14. Blickstein I, Goldchmit R, Starno SD, Goldman RD, Barzili N. Echogenicity of fibroadenoma and carcinoma of the breast. Quantitative comparison using gain-assisted densitometric evaluation of sonograms. J Ultrasound Med 1995;14: 661-64
  15. Huber S, Danes J, Zuna I, Teubner J, Medl M, Delorme S. Relevance of sonographic B-mode criteria and computer- aided ultrasonic tissue characterization in differential/ diagnosis of solid breast masses. Ultrasound Med Biol 2000; 26:1243-52 https://doi.org/10.1016/S0301-5629(00)00274-X
  16. Bongers V, Borel Rinkes IH, Sie-Go DM, Pijnappel R, de Hooge P and van Rijk PP. Detection of malignant breast tumors in dense breast tissue: results of Tc-99m tetrofosmin scintimammography related to surgery. Eur J Surg Oncol 1999;25:152-6 https://doi.org/10.1053/ejso.1998.0618
  17. Spanu A, Dettori G, Nuvoli G, Porcu A, Falchi A, Cottu P, et al. Tc-99m tetrofosmin SPECT in the detection of both primary breast cancer and axillary lymph node metastasis. Eur J Nucl Med 2001;28:1781-94 https://doi.org/10.1007/s00259-001-0657-5
  18. buscombe JR, Cwikla JB, Holloway B, Hilson AJ. Prediction of the Usefulness of Combined Mammography and Scintimammography in Suspected Primary Breast Cancer Using ROC Curves. J Nucl Med 2001;42:3-8
  19. Bone B, Wiberg K, Szabo BK, Szakos A, Danielsson R. Comparison of Tc-99m Sestamibi Scintimammography and dynamic MR Imaging as adjuncts to mammography in the diagnosis of breast cancer. Acta Radiologica 2003;44:28-34 https://doi.org/10.1034/j.1600-0455.2003.00012.x
  20. Kupinski MA, Edwards DC, Giger ML, Metz CE. Ideal observer approximation using Bayesian classification neural networks. IEEE Trans Med Imaging 2001;20:886-99 https://doi.org/10.1109/42.952727