Abstract: Visual recognition of cardiac images is important for cardiac pathology diagnosis and treatment. Due to the limited availability of annotated datasets, traditional methods usually extract features directly from two-dimensional slices of three-dimensional (3D) heart images, followed by pathological classification. This process may not ensure the overall anatomical consistency in 3D heart. A new method for classification of cardiac pathology is therefore proposed based on 3D parametric model reconstruction. First, 3D heart models are reconstructed based on multiple 3D volumes of cardiac imaging data at the end-systole (ES) and end-diastole (ED) phases. Next, based on these reconstructed 3D hearts, 3D parametric models are constructed through the statistical shape model (SSM), and then the heart data are augmented via the variation in shape parameters of one 3D parametric model with visual knowledge constraints. Finally, shape and motion features of 3D heart models across two phases are extracted to classify cardiac pathology. Comprehensive experiments on the automated cardiac diagnosis challenge (ACDC) dataset of the Statistical Atlases and Computational Modelling of the Heart (STACOM) workshop confirm the superior performance and efficiency of this proposed approach.

基于三维参数模型重建的心脏病理视觉识别

[1]Ammar A, Bouattane O, Youssfi M, 2021. Automatic cardiac cine MRI segmentation and heart disease classification. Comput Med Imag Graph, 88:101864.

[2]Attar R, Pereañez M, Bowles C, et al., 2019. 3D cardiac shape prediction with deep neural networks: simultaneous use of images and patient metadata. Proc 22^nd Int Conf on Medical Image Computing and Computer-Assisted Intervention, p.586-594.

[3]Bai WJ, Oktay O, Rueckert D, 2016. Classification of myocardial infarcted patients by combining shape and motion features. Proc 6^th Int Workshop on Statistical Atlases and Computational Models of the Heart, p.140-145.

[4]Bernard O, Lalande A, Zotti C, et al., 2018. Deep learning techniques for automatic MRI cardiac multi-structures segmentation and diagnosis: is the problem solved? IEEE Trans Med Imag, 37(11):2514-2525.

[5]Bernardino G, Benkarim O, Sanz-de la Garza M, et al., 2020. Handling confounding variables in statistical shape analysis-application to cardiac remodelling. Med Image Anal, 65:101792.

[6]Besl PJ, McKay ND, 1992. A method for registration of 3-D shapes. IEEE Trans Patt Anal Mach Intell, 14(2):239-256.

[7]Biffi C, Cerrolaza JJ, Tarroni G, et al., 2019. 3D high-resolution cardiac segmentation reconstruction from 2D views using conditional variational autoencoders. Proc IEEE 16^th Int Symp on Biomedical Imaging, p.1643-1646.

[8]Brock A, Donahue J, Simonyan K, 2019. Large scale GAN training for high fidelity natural image synthesis. https://arxiv.org/abs/1809.11096

[9]Cerqueira MD, Weissman NJ, Dilsizian V, et al., 2002. Standardized myocardial segmentation and nomenclature for tomographic imaging of the heart—a statement for healthcare professionals from the Cardiac Imaging Committee of the Council on Clinical Cardiology of the American Heart Association. Circulation, 105(4):539-542.

[10]Cetin I, Sanroma G, Petersen SE, et al., 2017. A radiomics approach to computer-aided diagnosis with cardiac cine-MRI. Proc 8^th Int Workshop on Statistical Atlases and Computational Models of the Heart, p.82-90.

[11]Chang YK, Jung C, 2020. Automatic cardiac MRI segmentation and permutation-invariant pathology classification using deep neural networks and point clouds. Neurocomputing, 418:270-279.

[12]Charles RQ, Su H, Mo KC, et al., 2017. PointNet: deep learning on point sets for 3D classification and segmentation. Proc IEEE Conf on Computer Vision and Pattern Recognition, p.77-85.

[13]Chen X, Ravikumar N, Xia Y, et al., 2021. Shape registration with learned deformations for 3D shape reconstruction from sparse and incomplete point clouds. Med Imag Anal, 74:102228.

[14]Cootes TF, Taylor CJ, Cooper DH, et al., 1995. Active shape models—their training and application. Comput Vis Imag Underst, 61(1):38-59.

[15]Dryden IL, Mardia KV, 1998. Statistical Shape Analysis. John Wiley & Sons, London, UK, p.663-669.

[16]Frangi AF, Rueckert D, Schnabel JA, et al., 2002. Automatic construction of multiple-object three-dimensional statistical shape models: application to cardiac modeling. IEEE Trans Med Imag, 21(9):1151-1166.

[17]Gao ZF, Zhang HY, Wang DF, et al., 2018. Robust recovery of myocardial kinematics using dual H_∞ criteria. Multim Tools Appl, 77(17):23043-23071.

[18]Gjesdal O, Bluemke DA, Lima JA, 2011. Cardiac remodeling at the population level—risk factors, screening, and outcomes. Nat Rev Cardiol, 8(12):673-685.

[19]Isensee F, Jaeger PF, Full PM, et al., 2018. Automatic cardiac disease assessment on cine-MRI via time-series segmentation and domain specific features. Proc 8^th Int Workshop on Statistical Atlases and Computational Models of the Heart, p.120-129.

[20]Jolliffe IT, 2002. Principal component analysis. J Mark Res, 87(4):513.

[21]Karras T, Laine S, Aila T, 2019. A style-based generator architecture for generative adversarial networks. Proc IEEE/CVF Conf on Computer Vision and Pattern Recognition, p.4396-4405.

[22]Khened M, Alex V, Krishnamurthi G, 2018. Densely connected fully convolutional network for short-axis cardiac cine MR image segmentation and heart diagnosis using random forest. Proc 8^th Int Workshop on Statistical Atlases and Computational Models of the Heart, p.140-151.

[23]Liu HF, Shi PC, 2009. Maximum a posteriori strategy for the simultaneous motion and material property estimation of the heart. IEEE Trans Biomed Eng, 56(2):378-389.

[24]Pan YH, 1996. The synthesis reasoning. Patt Recogn Artif Intell, 9(3):201-208 (in Chinese).

[25]Pan YH, 2019. On visual knowledge. Front Inform Technol Electron Eng, 20(8):1021-1025.

[26]Pan YH, 2020. Multiple knowledge representation of artificial intelligence. Engineering, 6(3):216-217.

[27]Pan YH, 2021a. Miniaturized five fundamental issues about visual knowledge. Front Inform Technol Electron Eng, 22(5):615-618.

[28]Pan YH, 2021b. On visual understanding. Front Inform Technol Electron Eng, early access.

[29]Parajuli N, Lu A, Duncan JS, 2016. Left ventricle classification using active shape model and support vector machine. Proc 6^th Int Workshop on Statistical Atlases and Computational Models of the Heart, p.154-161.

[30]Rodero C, Strocchi M, Marciniak M, et al., 2021. Linking statistical shape models and simulated function in the healthy adult human heart. PLoS Comput Biol, 17(4):e1008851.

[31]Suinesiaputra A, Ablin P, Albà X, et al., 2018. Statistical shape modeling of the left ventricle: myocardial infarct classification challenge. IEEE J Biomed Health Inform, 22(2):503-515.

[32]Thermos S, Liu X, O’Neil A, et al., 2021. Controllable cardiac synthesis via disentangled anatomy arithmetic. Proc 24^th Int Conf on Medical Image Computing and Computer-Assisted Intervention, p.160-170.

[33]Van Dijck C, Wirix-Speetjens R, Jonkers I, et al., 2018. Statistical shape model-based prediction of tibiofemoral cartilage. Comput Methods Biomech Biomed Eng, 21(9):568-578.

[34]Wolterink JM, Leiner T, Viergever MA, et al., 2018. Automatic segmentation and disease classification using cardiac cine MR images. Proc 8^th Int Workshop on Statistical Atlases and Computational Models of the Heart, p.101-110.

[35]Zheng Q, Delingette H, Ayache N, 2019. Explainable cardiac pathology classification on cine MRI with motion characterization by semi-supervised learning of apparent flow. Med Imag Anal, 56:80-95.

[36]Zhou QY, Park J, Koltun V, 2016. Fast global registration. Proc 14^th European Conf on Computer Vision, p.766-782.