A Gaussian Process Model for Unsupervised Analysis of High Dimensional Shape Data
Pages 356 - 365
Abstract
Applications of medical image analysis are often faced with the challenge of modelling high-dimensional data with relatively few samples. In many settings, normal or healthy samples are prevalent while pathological samples are rarer, highly diverse, and/or difficult to model. In such cases, a robust model of the normal population in the high-dimensional space can be useful for characterizing pathologies. In this context, there is utility in hybrid models, such as probabilistic PCA, which learns a low-dimensional model, commensurates with the available data, and combines it with a generic, isotropic noise model for the remaining dimensions. However, the isotropic noise model ignores the inherent correlations that are evident in so many high-dimensional data sets associated with images and shapes in medicine. This paper describes a method for estimating a Gaussian model for collections of images or shapes that exhibit underlying correlations, e.g., in the form of smoothness. The proposed method incorporates a Gaussian-process noise model within a generative formulation. For optimization, we derive a novel expectation maximization (EM) algorithm. We demonstrate the efficacy of the method on synthetic examples and on anatomical shape data.
References
[1]
Atkins PR et al. Quantitative comparison of cortical bone thickness using correspondence-based shape modeling in patients with cam femoroacetabular impingement J. Orthopaedic Res. 2017 35 8 1743-1753
[2]
Bhalodia R, Dvoracek LA, Ayyash AM, Kavan L, Whitaker R, and Goldstein JA Quantifying the severity of metopic craniosynostosis: a pilot study application of machine learning in craniofacial surgery J. Craniofac. Surg. 2020 31 3 697-701
[3]
Cates J, Fletcher PT, Styner M, Shenton M, and Whitaker R Karssemeijer N and Lelieveldt B Shape modeling and analysis with entropy-based particle systems Information Processing in Medical Imaging 2007 Heidelberg Springer 333-345
[4]
Chen Y, Wiesel A, Eldar YC, and Hero AO Shrinkage algorithms for MMSE covariance estimation IEEE Trans. Signal Process. 2010 58 10 5016-5029
[5]
Doan, N.T., van Lew, B., Lelieveldt, B., van Buchem, M.A., Reiber, J.H.C., Milles, J.: Deformation texture-based features for classification in Alzheimer’s disease. In: Medical Imaging 2013: Image Processing, vol. 8669, pp. 601–607. International Society for Optics and Photonics, SPIE (2013)
[6]
Fletcher PT, Lu C, Pizer SM, and Joshi S Principal geodesic analysis for the study of nonlinear statistics of shape IEEE Trans. Med. Imaging 2004 23 8 995-1005
[7]
Gerig T, Shahim K, Reyes M, Vetter T, and Lüthi M Golland P, Hata N, Barillot C, Hornegger J, and Howe R Spatially varying registration using Gaussian processes Medical Image Computing and Computer-Assisted Intervention – MICCAI 2014 2014 Cham Springer 413-420
[8]
Gu M and Shen W Generalized probabilistic principal component analysis of correlated data J. Mach. Learn. Res. 2020 21 13 1-41
[9]
Joshi SC, Miller MI, and Grenander U On the geometry and shape of brain sub-manifolds Int. J. Pattern Recogn. Artif. Intell. 1997 11 08 1317-1343
[10]
Kellogg R, Allori AC, Rogers GF, and Marcus JR Interfrontal angle for characterization of trigonocephaly: part 1: development and validation of a tool for diagnosis of metopic synostosis J. Craniofac. Surg. 2012 23 3 799-804
[11]
Kingma, D.P., Ba, J.: Adam: a method for stochastic optimization. arXiv preprint arXiv:1412.6980 (2014)
[12]
Lawrence N and Hyvärinen A Probabilistic non-linear principal component analysis with Gaussian process latent variable models J. Mach. Learn. Res. 2005 6 11 1783-1816
[13]
Lê M, Unkelbach J, Ayache N, and Delingette H Navab N, Hornegger J, Wells WM, and Frangi AF GPSSI: Gaussian process for sampling segmentations of images Medical Image Computing and Computer-Assisted Intervention – MICCAI 2015 2015 Cham Springer 38-46
[14]
Ledoit O and Wolf M A well-conditioned estimator for large-dimensional covariance matrices J. Multivariate Anal. 2004 88 2 365-411
[15]
Leys C, Klein O, Dominicy Y, and Ley C Detecting multivariate outliers: use a robust variant of the Mahalanobis distance J. Exp. Soc. Psychol. 2018 74 150-156
[16]
Liu, P.J.: Using Gaussian process regression to denoise images and remove artefacts from microarray data. University of Toronto (2007)
[17]
Lüthi M, Gerig T, Jud C, and Vetter T Gaussian process morphable models IEEE Trans. Pattern Anal. Mach. Intell. 2017 40 8 1860-1873
[18]
Oktay, O., et al.: Anatomically constrained neural networks (ACNNs): application to cardiac image enhancement and segmentation. IEEE Trans. Med. Imaging 37(2), 384–395 (2018).
[19]
Purisha Z, Jidling C, Wahlström N, Schön TB, and Särkkä S Probabilistic approach to limited-data computed tomography reconstruction Inverse Probl. 2019 35 10 105004
[20]
Roweis, S.: EM algorithms for PCA and SPCA. In: Proceedings of the 10th International Conference on Neural Information Processing Systems, pp. 626–632, NIPS 1997. MIT Press, Cambridge (1997)
[21]
Rueckert D, Frangi AF, and Schnabel JA Automatic construction of 3-D statistical deformation models of the brain using nonrigid registration IEEE Trans. Med. Imaging 2003 22 8 1014-1025
[22]
Styner M et al. Framework for the statistical shape analysis of brain structures using SPHARM-PDM Insight J. 2006 1071 242-250
[23]
Styner M and Gerig G Insana MF and Leahy RM Medial models incorporating object variability for 3D shape analysis Information Processing in Medical Imaging 2001 Heidelberg Springer 502-516
[24]
Beiging ET, Morris A, Wilson BD, McGann CJ, Marrouche NF, and Cates J Left atrial shape predicts recurrence after atrial fibrillation catheter ablation J. Cardiovasc. Electrophysiol. 2018 29 7 966-972
[25]
Tipping ME and Bishop CM Probabilistic principal component analysis J. R. Stat. Soc. Ser. B 1999 61 3 611-622
[26]
Uebersax JS and Grove WM A latent trait finite mixture model for the analysis of rating agreement Biometrics 1993 49 3 823-835
[27]
Wood BC et al. What’s in a name? Accurately diagnosing metopic craniosynostosis using a computational approach Plastic Reconstr. Surg. 2016 137 1 205-213
Index Terms
- A Gaussian Process Model for Unsupervised Analysis of High Dimensional Shape Data
Index terms have been assigned to the content through auto-classification.
Recommendations
Gaussian Process Subspace Prediction for Model Reduction
Subspace-valued functions arise in a wide range of problems, including parametric reduced order modeling (PROM), parameter reduction, and subspace tracking. In PROM, each parameter point can be associated with a subspace, which is used for Petrov--Galerkin ...
Shared Linear Encoder-based Gaussian Process Latent Variable Model for Visual Classification
MM '18: Proceedings of the 26th ACM international conference on MultimediaMulti-view learning has shown its powerful potential in many applications and achieved outstanding performances compared with the single-view based methods. In this paper, we propose a novel multi-view learning model based on the Gaussian Process Latent ...
Enhanced algorithm for high-dimensional data classification
Graphical abstractIllustration of the decision hyperplanes generated by TSSVM, MCVSVM, and LMLP on an artificial dataset. Display Omitted HighlightsIn the case of the singularity of the within-class scatter matrix, the drawbacks of both MCVSVM and LMLP ...
Comments
Information & Contributors
Information
Published In
Sep 2021
722 pages
ISBN:978-3-030-87588-6
DOI:10.1007/978-3-030-87589-3
© Springer Nature Switzerland AG 2021.
Publisher
Springer-Verlag
Berlin, Heidelberg
Publication History
Published: 27 September 2021
Author Tags
Qualifiers
- Article
Contributors
Other Metrics
Bibliometrics & Citations
Bibliometrics
Article Metrics
- 0Total Citations
- 0Total Downloads
- Downloads (Last 12 months)0
- Downloads (Last 6 weeks)0
Reflects downloads up to 03 Oct 2024
Other Metrics
Citations
View Options
View options
Get Access
Login options
Check if you have access through your login credentials or your institution to get full access on this article.
Sign in