Quantitative Assessment Technology of Small Animal Myocardial Infarction PET Image Using Gaussian Mixture Model

다중가우시안혼합모델을 이용한 소동물 심근경색 PET 영상의 정량적 평가 기술

  • Woo, Sang-Keun (Molecular Imaging Research Center, Korea Institute of Radiological and Medical Sciences) ;
  • Lee, Yong-Jin (Molecular Imaging Research Center, Korea Institute of Radiological and Medical Sciences) ;
  • Lee, Won-Ho (Molecular Imaging Research Center, Korea Institute of Radiological and Medical Sciences) ;
  • Kim, Min-Hwan (Molecular Imaging Research Center, Korea Institute of Radiological and Medical Sciences) ;
  • Park, Ji-Ae (Molecular Imaging Research Center, Korea Institute of Radiological and Medical Sciences) ;
  • Kim, Jin-Su (Molecular Imaging Research Center, Korea Institute of Radiological and Medical Sciences) ;
  • Kim, Jong-Guk (Molecular Imaging Research Center, Korea Institute of Radiological and Medical Sciences) ;
  • Kang, Joo-Hyun (Molecular Imaging Research Center, Korea Institute of Radiological and Medical Sciences) ;
  • Ji, Young-Hoon (Division of Radiation Cancer Research, Korea Institute of Radiological and Medical Sciences) ;
  • Choi, Chang-Woon (Department of Nuclear Medicine, Korea Institute of Radiological and Medical Sciences) ;
  • Lim, Sang-Moo (Department of Nuclear Medicine, Korea Institute of Radiological and Medical Sciences) ;
  • Kim, Kyeong-Min (Molecular Imaging Research Center, Korea Institute of Radiological and Medical Sciences)
  • 우상근 (한국원자력의학원 방사선의학연구소 분자영상연구부) ;
  • 이용진 (한국원자력의학원 방사선의학연구소 분자영상연구부) ;
  • 이원호 (한국원자력의학원 방사선의학연구소 분자영상연구부) ;
  • 김민환 (한국원자력의학원 방사선의학연구소 분자영상연구부) ;
  • 박지애 (한국원자력의학원 방사선의학연구소 분자영상연구부) ;
  • 김진수 (한국원자력의학원 방사선의학연구소 분자영상연구부) ;
  • 김종국 (한국원자력의학원 방사선의학연구소 분자영상연구부) ;
  • 강주현 (한국원자력의학원 방사선의학연구소 분자영상연구부) ;
  • 지영훈 (한국원자력의학원 방사선의학연구소 방사선암연구부) ;
  • 최창운 (한국원자력의학원 방사선의학연구소 원자력병원 핵의학과) ;
  • 임상무 (한국원자력의학원 방사선의학연구소 원자력병원 핵의학과) ;
  • 김경민 (한국원자력의학원 방사선의학연구소 분자영상연구부)
  • Received : 2011.02.15
  • Accepted : 2011.03.02
  • Published : 2011.03.31

Abstract

Nuclear medicine images (SPECT, PET) were widely used tool for assessment of myocardial viability and perfusion. However it had difficult to define accurate myocardial infarct region. The purpose of this study was to investigate methodological approach for automatic measurement of rat myocardial infarct size using polar map with adaptive threshold. Rat myocardial infarction model was induced by ligation of the left circumflex artery. PET images were obtained after intravenous injection of 37 MBq $^{18}F$-FDG. After 60 min uptake, each animal was scanned for 20 min with ECG gating. PET data were reconstructed using ordered subset expectation maximization (OSEM) 2D. To automatically make the myocardial contour and generate polar map, we used QGS software (Cedars-Sinai Medical Center). The reference infarct size was defined by infarction area percentage of the total left myocardium using TTC staining. We used three threshold methods (predefined threshold, Otsu and Multi Gaussian mixture model; MGMM). Predefined threshold method was commonly used in other studies. We applied threshold value form 10% to 90% in step of 10%. Otsu algorithm calculated threshold with the maximum between class variance. MGMM method estimated the distribution of image intensity using multiple Gaussian mixture models (MGMM2, ${\cdots}$ MGMM5) and calculated adaptive threshold. The infarct size in polar map was calculated as the percentage of lower threshold area in polar map from the total polar map area. The measured infarct size using different threshold methods was evaluated by comparison with reference infarct size. The mean difference between with polar map defect size by predefined thresholds (20%, 30%, and 40%) and reference infarct size were $7.04{\pm}3.44%$, $3.87{\pm}2.09%$ and $2.15{\pm}2.07%$, respectively. Otsu verse reference infarct size was $3.56{\pm}4.16%$. MGMM methods verse reference infarct size was $2.29{\pm}1.94%$. The predefined threshold (30%) showed the smallest mean difference with reference infarct size. However, MGMM was more accurate than predefined threshold in under 10% reference infarct size case (MGMM: 0.006%, predefined threshold: 0.59%). In this study, we was to evaluate myocardial infarct size in polar map using multiple Gaussian mixture model. MGMM method was provide adaptive threshold in each subject and will be a useful for automatic measurement of infarct size.

전통적으로 심근 생존능을 식별하고 심근 관류를 정확히 평가하기 위한 도구로 핵의학영상이 이용되고 있으나 경색영역을 정의하기에는 어려움이 있다. 이에 본 연구에서는 극성지도의 분포를 분석하여 특성에 맞는 적응적 임계값을 이용하여 심근경색 모델을 정량적으로 평가하고자 하였다. 쥐 심근경색 모델은 왼쪽 관상동맥을 결찰시켜 제작하였다. 소동물PET 영상은 37 MBq $^{18}F$-FDG를 쥐의 꼬리정맥에 주사한 후 60분 섭취 후 Siemens Inveon SPECT/PET 스캐너를 이용하여 20분 동안 ECG 신호와 함께 획득하였고, OSEM 2D 알고리즘을 이용하여 재구성하였다. PET 영상의 심근 극성지도는 Siemens QGS 소프트웨어에 적합한 형식으로 변환 후 자동으로 심근 벽을 설정하여 작성하였다. 심근경색영역의 기준데이터는 TTC 염색으로 설정하였으며 전체 좌심실대비 염색된 영역의 백분율로 획득하였다. 최적의 임계값 설정을 위해 절대치 설정 방법, Otsu 알고리즘, 다중가우시안혼합모델(Multi Gaussian mixture model, MGMM)을 이용하여 평가하였다. 절대치 설정 방법은 10~90%까지 10%단위로 미리 정의 된 임계값을 이용하였고, Otsu 알고리즘은 영상 내에서 두 군집의 분산을 최대로 하는 임계값으로 설정하였다. MGMM 방법은 영상의 화소 강도를 분석하여 여러 개의 가우시안 분포함수(MGMM2, $\cdots$ MGMM4)로 반복 수행하여 최적의 가우시안 분포를 구하여 적응적 임계값을 설정하였다. 극성지도 평가지표는 각각의 알고리즘에서 측정된 임계값을 이용하여 이진화하고 전체 극성지도와 경색영역의 백분율로 획득한 후, TTC 염색으로 획득된 기준데이터와의 차이를 비교하였다. 그 차이는 절대치 방법의 20%에서 $7.04{\pm}3.44%$, 30%에서 $3.87{\pm}2.09%$, 40%에서 $2.15{\pm}2.07%$이었다. Otsu 방법은 $3.56{\pm}4.16%$이었으며 MGMM 방법은 $2.29{\pm}1.94%$이었다. 소동물 PET 극성지도에서는 30% 임계값이 조직학적 데이터와 비교하여 가장 작은 차이를 보였다. 그러나 TTC 염색으로 측정한 크기가 10% 이하에서는 MGMM 방법이 절대치 방법보다 작은 차이를 보였다(MGMM: 0.006%, 절대치방법: 0.59%). 이 연구에서는 심근경색 모델 평가를 위하여 생체영상 극성지도에서 다중가우시안혼합모델을 이용하여 평가하고자 하였다. MGMM은 사용자의 선택 없이도 자동적으로 영상 특성을 고려하여 적응적 임계값을 찾아주는 방법으로 극성지도에서 심근경색을 평가하는데 도움이 될 것으로 기대된다.

Keywords

References

  1. Di Carli MF, Dorbala S, Meserve J, El Fakhri G, Sitek A, Moore SC: Clinical myocardial perfusion PET/CT. J Nucl Med 48:783-793 (2007) https://doi.org/10.2967/jnumed.106.032789
  2. Nieman K, Shapiro MD, Ferencik M, et al: Reperfused myocardial infarction: contrast-enhanced 64-Section CT in comparison to MR imaging. Radiology 247:49-56 (2008) https://doi.org/10.1148/radiol.2471070332
  3. Koszegi Z, Balkay L, Galuska L, Varga J, Hegedus I, Fulop T: Holistic polar map for integrated evaluation of cardiac imaging results. Comput Med Imaging Graph 31:577-586 (2007) https://doi.org/10.1016/j.compmedimag.2007.06.008
  4. Higuchi T, Nekolla SG, Jankaukas A, et al: Characterization of normal and infarcted rat myocardium using a combination of small-animal PET and clinical MRI. J Nucl Med 48:288-294 (2007)
  5. Kudo T, Fukuchi K, Annala AJ, et al: Noninvasive measurement of myocardial activity concentrations and perfusion defect sizes in rats with a new small-animal positron emission tomograph. Circulation. 106:118-123 (2002) https://doi.org/10.1161/01.CIR.0000020221.28996.78
  6. Buther F, Dawood M, Stegger L, et al: List mode-driven cardiac and respiratory gating in PET. J Nucl Med 50:674-681 (2009) https://doi.org/10.2967/jnumed.108.059204
  7. Stegger L, Hoffmeier AN, Schafers KP, et al: Accurate noninvasive measurement of infarct size in mice with high-resolution PET. J Nucl Med 47:1837-1844 (2006)
  8. Kuwabara Y, Watanabe S, Nakaya J, et al: Functional evaluation of myocardial viability by 99mTc tetrofosmin gated SPECT--a quantitative comparison with 18F fluorodeoxyglucose positron emission CT (18F FDG PET). Ann Nucl Med 13:135-140 (1999) https://doi.org/10.1007/BF03164852
  9. Maruyama A, Hasegawa S, Paul AK, et al: Myocardial viability assessment with gated SPECT Tc-99m tetrofosmin % wall thickening: comparison with F-18 FDG-PET. Ann Nucl Med 16:25-32 (2002) https://doi.org/10.1007/BF02995288
  10. Kudo T, Fukuchi K, Annala AJ, et al: Noninvasive measurement of myocardial activity concentrations and perfusion defect sizes in rats with a new small-animal positron emission tomograph. Circulation 106:118-123 (2002) https://doi.org/10.1161/01.CIR.0000020221.28996.78
  11. Giorgetti A, Marzullo P, Sambuceti G, et al: Base-line/post-nitrate Tc-99m tetrofosmin mismatch for the assessment of myocardial viability in patients with severe left ventricular dysfunction: comparison with baseline Tc-99m tetrofosmin scintigraphy/ FDG PET imaging. J Nucl Cardiol 11:142-151 (2004) https://doi.org/10.1016/j.nuclcard.2003.12.011
  12. Thomas D, Bal H, Arkles J, et al: Noninvasive assessment of myocardial viability in a small animal model: comparison of mri, spect, and pet. Magn Reson Med 59:252-259 (2008) https://doi.org/10.1002/mrm.21445
  13. Sherif HM, Saraste A, Weidl E, et al: Evaluation of a novel 18F-labeled positron-emission tomography perfusion tracer for the assessment of myocardial infarct size in rats. Circ Cardiovasc Imaging 2:77-84 (2009) https://doi.org/10.1161/CIRCIMAGING.108.815423
  14. Lautamaki R, Schuleri KH, Sasano T, et al: Integration of infarct size, tissue perfusion, and metabolism by hybrid cardiac positron emission tomography/computed tomography: evaluation in a porcine model of myocardial infarction. Circ Cardiovasc Imaging 2:299-305 (2009) https://doi.org/10.1161/CIRCIMAGING.108.846253
  15. Slomka PJ, Fieno D, Thomson L, et al: Automatic detection and size quantification of infarcts by myocardial perfusion SPECT: clinical validation by delayed-enhancement MRI. J Nucl Med 46:728-735 (2005)
  16. Rosendahl L, Blomstrand P, Heiberg E, et al: Computer- assisted calculation of myocardial infarct size shortens the evaluation time of contrast-enhanced cardiac MRI. Clin Physiol Funct Imaging 28:1-7 (2008)
  17. Woo SK, Lee TS, Kim KM, et al: Anesthesia condition for 18F-FDG imaging of lung metastasis tumors using small animal PET. Nucl Med Biol 35:143-150 (2008) https://doi.org/10.1016/j.nucmedbio.2007.10.003
  18. Otsu N: A threshold selection method from gray-level histograms. IEEE Trans Sys Man Cyber 9:62-66 (1979)
  19. Huang ZK, Chau KW: A new image thresholding method based on Gaussian mixture model. Appl Math Comput 205:899-907 (2008) https://doi.org/10.1016/j.amc.2008.05.130