A learning-based automatic segmentation and quantification method on left ventricle in gated myocardial perfusion SPECT imaging: A feasibility study
- 32 Downloads
The performance of left ventricular (LV) functional assessment using gated myocardial perfusion SPECT (MPS) relies on the accuracy of segmentation. Current methods require manual adjustments that are tedious and subjective. We propose a novel machine-learning-based method to automatically segment LV myocardium and measure its volume in gated MPS imaging without human intervention.
We used an end-to-end fully convolutional neural network to segment LV myocardium by delineating its endocardial and epicardial surface. A novel compound loss function, which encourages similarity and penalizes discrepancy between prediction and training dataset, is utilized in training stage to achieve excellent performance. We retrospectively investigated 32 normal patients and 24 abnormal patients, whose LV myocardial contours automatically segmented by our method were compared with those delineated by physicians as the ground truth.
The results of our method demonstrated very good agreement with the ground truth. The average DSC metrics and Hausdorff distance of the contours delineated by our method are larger than 0.900 and less than 1 cm, respectively, among all 32 + 24 patients of all phases. The correlation coefficient of the LV myocardium volume between ground truth and our results is 0.910 ± 0.061 (P < 0.001), and the mean relative error of LV myocardium volume is − 1.09 ± 3.66%.
These results strongly indicate the feasibility of our method in accurately quantifying LV myocardium volume change over the cardiac cycle. The learning-based segmentation method in gated MPS imaging has great promise for clinical use.
KeywordsMyocardial perfusion SPECT segmentation machine learning
Single-photon emission computed tomography
Myocardial perfusion SPECT
Magnetic resonance imaging
Binary cross entropy
Dice similarity coefficient
This research is supported in part by the National Cancer Institute of the National Institutes of Health under Award Number R01CA215718 and Emory Winship Cancer Institute pilot Grant. This research is also supported by the American Heart Association under Award Number 17AIREA33700016.
Dr. Kristin Higgins is consulting for Astra Zeneca, Varian, on advisory board for Genentech, and receiving research funding from RefleXion Medical. Tonghe Wang, Yang Lei, Haipeng Tang, Zhuo He, Richard Castillo, Cheng Wang, Dianfu Li, Tian Liu, Walter J. Curran, Weihua Zhou, and Xiaofeng Yang declare that they have no conflict of interest.
- 2.Holder L, Lewis S, Abrames E, Wolin EA. Review of SPECT myocardial perfusion imaging. J Am Osteopath Coll Radiol. 2016;5:5–13.Google Scholar
- 6.Paul AK, Nabi HA. Gated myocardial perfusion SPECT: basic principles, technical aspects, and clinical applications. J Nucl Med Technol. 2004;32:179–87 quiz 88-9.Google Scholar
- 9.Sharir T, Germano G, Kang X, Lewin HC, Miranda R, Cohen I, et al. Prediction of myocardial infarction versus cardiac death by gated myocardial perfusion SPECT: risk stratification by the amount of stress-induced ischemia and the poststress ejection fraction. J Nucl Med. 2001;42:831–7.Google Scholar
- 10.Sharir T, Berman DS, Waechter PB, Areeda J, Kavanagh PB, Gerlach J, et al. Quantitative analysis of regional motion and thickening by gated myocardial perfusion SPECT: normal heterogeneity and criteria for abnormality. J Nucl Med. 2001;42:1630–8.Google Scholar
- 11.Shaw LJ, Iskandrian AE. Prognostic value of gated myocardial perfusion SPECT. J Nucl Med. 2004;11:171–85.Google Scholar
- 12.Friehling M, Chen J, Saba S, Bazaz R, Schwartzman D, Adelstein EC, et al. A prospective pilot study to evaluate the relationship between acute change in left ventricular synchrony after cardiac resynchronization therapy and patient outcome using a single-injection gated SPECT protocol. Circ Cardiovasc Imaging. 2011;4:532–9.CrossRefGoogle Scholar
- 13.Faber TL, Cooke CD, Folks RD, Vansant JP, Nichols KJ, DePuey EG, et al. Left ventricular function and perfusion from gated SPECT perfusion images: an integrated method. J Nucl Med. 1999;40:650–9.Google Scholar
- 15.Germano G, Kavanagh PB, Waechter P, Areeda J, Van Kriekinge S, Sharir T, et al. A new algorithm for the quantitation of myocardial perfusion SPECT. I: technical principles and reproducibility. J Nucl Med. 2000;41:712–9.Google Scholar
- 16.Germano G, Kiat H, Kavanagh PB, Moriel M, Mazzanti M, Su HT, et al. Automatic quantification of ejection fraction from gated myocardial perfusion SPECT. J Nucl Med. 1995;36:2138–47.Google Scholar
- 22.Milletari F, Navab N, Ahmadi S. V-Net: Fully Convolutional Neural Networks for Volumetric Medical Image Segmentation. 2016 Fourth International Conference on 3D Vision (3DV); 2016. p. 565-71.Google Scholar
- 23.Varadhan R, Karangelis G, Krishnan K, Hui S. A framework for deformable image registration validation in radiotherapy clinical applications. J Appl Clin. 2013;14:4066.Google Scholar
- 26.Trobaugh GB, Wackers FJ, Sokole EB, DeRouen TA, Ritchie JL, Hamilton GW. Thallium-201 myocardial imaging: an interinstitutional study of observer variability. J Nucl Med. 1978;19:359–63.Google Scholar
- 27.Nakajima K, Higuchi T, Taki J, Kawano M, Tonami N. Accuracy of ventricular volume and ejection fraction measured by gated myocardial SPECT: comparison of 4 software programs. J Nucl Med. 2001;42:1571–8.Google Scholar