Conference PaperPDF Available

Abstract

Untangling pathological processes from the natural inter-individual variability is a major challenge in Alzheimer’s disease (AD): natural anatomical variability is often much larger than individual pathological alterations, hindering the identification of relevant disease markers. In a longitudinal MRI data set of subjects progressing from mild cognitive impairment (MCI) to AD, variability not only comes from baseline differences, but also from the individual dynamics of progression. Based on recent computational morphometry tools, [1] proposes to learn a hierarchical model from longitudinal shape data sets, where individual progressions are represented in terms of orthogonal spatial and temporal warps of a common reference progression. In this work, we extend this approach to 3D images to model the spatiotemporal progression of atrophy in AD starting from the prodromal stage. A long-term average scenario of progression to AD is learned, along with interpretable individual parameters of variability that we correlate with genetic and environmental co-factors.
Hierarchical modeling of Alzheimer's disease
progression from a large longitudinal MRI data set
Alexandre Bône1, Benoît Martin1, Maxime Louis1, Olivier Colliot1, Stanley Durrleman1,
and the Alzheimer’s Disease Neuroimaging Initiative (ADNI)
1ARAMIS Lab, ICM, Inserm U1127, CNRS UMR 7225, Sorbonne University, Inria, Paris,
France
Introduction
Untangling pathological processes from the natural inter-individual variability is a major
challenge in Alzheimer’s disease (AD): natural anatomical variability is often much larger
than individual pathological alterations, hindering the identification of relevant disease
markers. In a longitudinal MRI data set of subjects progressing from mild cognitive
impairment (MCI) to AD, variability not only comes from baseline differences, but also from
the individual dynamics of progression. Based on recent computational morphometry tools,
[1] proposes to learn a hierarchical model from longitudinal shape data sets, where individual
progressions are represented in terms of orthogonal spatial and temporal warps of a
common reference progression. In this work, we extend this approach to 3D images to
model the spatiotemporal progression of atrophy in AD starting from the prodromal stage. A
long-term average scenario of progression to AD is learned, along with interpretable
individual parameters of variability that we correlate with genetic and environmental
co-factors.
Methods
We selected from the ADNI database all subjects meeting the following requirements: (i)
diagnosed as MCI at a visit, (ii) diagnosed as AD at a later visit, (iii) never reverted from AD
to MCI or from MCI to cognitively normal (CN). A total of 322 subjects fulfilled those criteria,
representing 1993 visits with T1-weighted MRI (average number of visits: 5.8 (±2.4); age:
74.0 (±6.7) years; education level: 15.9 (±2.8) years; 41.2% females; 65.2% APOE-ε4
carriers; 80.9% married). Images were first processed with the longitudinal FreeSurfer
pipeline [5] through the Clinica platform [6]. The resulting skull-removed intensity-normalized
images were then affinely aligned onto the Colin27 reference brain [4] with the FSL software
[7]. To reduce the computational burden, images were finally subsampled to a size of 643
voxels. The result was given as input to the longitudinal atlas pipeline of the Deformetrica
software [2, 3] that implements the approach introduced in [1]. Outputs are composed of an
average long-term scenario, along with low-dimensional parameters encoding how each
individual differ from this normative scenario. The variability in the dynamics is represented
for each individual by: (i) an estimated onset age that encodes the variability in the onset of
the alterations, (ii) an estimated acceleration factor that encodes the variability in the pace of
progression of those alterations.
Results
Figure 1 displays the estimated long-term scenario of progression to AD. The ventricles
clearly increase in size and atrophy is visible in the insula and parietal sulci regions. Figure 2
plots the estimated individual onset ages against the actual ages at which AD diagnosis was
made. The high correlation suggests that the model successfully captured the relative stages
of development of the disease across patients. We studied the association between
estimated onset ages and co-factors (gender, APOE-ε4 carriership, marital status and
education level) through multivariate linear regression. Analysis of the estimated coefficients
revealed that female subjects develop AD significantly earlier (by 31.2 (±18.4) months,
p=0.0019), similarly to APOE-ε4 carriers vs. non-carriers (by 19.8 (±17.5) months, p=0.036)
and married subjects vs. non-married (by 45.9 (±22.8) months, p=0.00038). The same
analysis was performed for the estimated individual acceleration factors: we found that
APOE-ε4 carriers progress significantly faster than non-carriers (by a factor of 1.14 (±0.09),
p=0.010). Confidence intervals are at 95%, and p-values are FDR-corrected.
Conclusions
The learned hierarchical model provides both a normative long-term scenario of AD
progression at the population level, and interpretable parameters encoding its dynamical
variability at the individual level. This scenario is in line with current medical knowledge, and
significant co-factors of progression can be identified from the individual parameters.
Acknowledgments
This work has been partly funded by grants ERC 678304, H2020 EU project 666992,
ANR-10-IAIHU-06.
References
[1]. Bône, A. et al. (2018, March). Learning distributions of shape trajectories from
longitudinal datasets: a hierarchical model on a manifold of diffeomorphisms. In Proceedings
of the IEEE Conference on Computer Vision and Pattern Recognition (pp. 9271-9280).
[2]. Bône, A. et al. (2018, September). Deformetrica 4: an open-source software for
statistical shape analysis. In International Workshop on Shape in Medical Imaging. Springer.
[3]. Durrleman, S. et al. (2014). Morphometry of anatomical shape complexes with dense
deformations and sparse parameters. NeuroImage, 101, 35-49.
[4]. Holmes, C. J. et al. (1998). Enhancement of MR images using registration for signal
averaging. Journal of computer assisted tomography, 22(2), 324-333.
[5]. Reuter, M. et al. (2012). Within-subject template estimation for unbiased longitudinal
image analysis. Neuroimage, 61(4), 1402-1418.
[6]. Samper-González, J. et al. (2018). Reproducible evaluation of classification methods in
Alzheimer's disease: framework and application to MRI and PET data. bioRxiv, 274324.
[7]. Woolrich, M. W. et al. (2009). Bayesian analysis of neuroimaging data in FSL.
Neuroimage, 45(1), S173-S186.
Figures
Thesis
This PhD proposes new Riemannian geometry tools for the analysis of longitudinal observations of neuro-degenerative subjects. First, we propose a numerical scheme to compute the parallel transport along geodesics. This scheme is efficient as long as the co-metric can be computed efficiently. Then, we tackle the issue of Riemannian manifold learning. We provide some minimal theoretical sanity checks to illustrate that the procedure of Riemannian metric estimation can be relevant. Then, we propose to learn a Riemannian manifold so as to model subject's progressions as geodesics on this manifold. This allows fast inference, extrapolation and classification of the subjects.
Full-text available
Conference Paper
Deformetrica is an open-source software for the statistical analysis of images and meshes. It relies on a specific instance of the large deformation diffeomorphic metric mapping (LDDMM) framework, based on control points: local momenta vectors offer a low-dimensional and interpretable parametrization of global diffeomorphims of the 2/3D ambient space, which in turn can warp any single or collection of shapes embedded in this physical space. Deformetrica has very few requirements about the data of interest: in the particular case of meshes, the absence of point correspondence can be handled thanks to the current or var-ifold representations. In addition to standard computational anatomy functionalities such as shape registration or atlas estimation, a bayesian version of atlas model as well as temporal methods (geodesic regression and parallel transport) are readily available. Installation instructions, tutorials and examples can be found at http://www.deformetrica.org.
Full-text available
Article
A large number of papers have introduced novel machine learning and feature extraction methods for automatic classification of AD. However, they are difficult to reproduce because key components of the validation are often not readily available. These components include selected participants and input data, image preprocessing and cross-validation procedures. The performance of the different approaches is also difficult to compare objectively. In particular, it is often difficult to assess which part of the method provides a real improvement, if any. We propose a framework for reproducible and objective classification experiments in AD using three publicly available datasets (ADNI, AIBL and OASIS). The framework comprises: i) automatic conversion of the three datasets into BIDS format, ii) a modular set of preprocessing pipelines, feature extraction and classification methods, together with an evaluation framework, that provide a baseline for benchmarking the different components. We demonstrate the use of the framework for a large-scale evaluation on 1960 participants using T1 MRI and FDG PET data. In this evaluation, we assess the influence of different modalities, preprocessing, feature types, classifiers, training set sizes and datasets. Performances were in line with the state-of-the-art. FDG PET outperformed T1 MRI for all classification tasks. No difference in performance was found for the use of different atlases, image smoothing, partial volume correction of FDG PET images, or feature type. Linear SVM and L2-logistic regression resulted in similar performance and both outperformed random forests. The classification performance increased along with the number of subjects used for training. Classifiers trained on ADNI generalized well to AIBL and OASIS. All the code of the framework and the experiments is publicly available at: https://gitlab.icm-institute.org/aramislab/AD-ML.
Full-text available
Conference Paper
We propose a method to learn a distribution of shape trajectories from longitudinal data, i.e. the collection of individual objects repeatedly observed at multiple time-points. The method allows to compute an average spatiotemporal trajectory of shape changes at the group level, and the individual variations of this trajectory both in terms of geometry and time dynamics. First, we formulate a non-linear mixed-effects statistical model as the combination of a generic statistical model for manifold-valued longitudinal data, a deformation model defining shape trajectories via the action of a finite-dimensional set of diffeomorphisms with a manifold structure, and an efficient numerical scheme to compute parallel transport on this manifold. Second, we introduce a MCMC-SAEM algorithm with a specific approach to shape sampling, an adaptive scheme for proposal variances , and a log-likelihood tempering strategy to estimate our model. Third, we validate our algorithm on 2D simulated data, and then estimate a scenario of alteration of the shape of the hippocampus 3D brain structure during the course of Alzheimer's disease. The method shows for instance that hippocampal atrophy progresses more quickly in female subjects, and occurs earlier in APOE4 mutation carriers. We finally illustrate the potential of our method for classifying pathological trajectories versus normal ageing.
Article
We propose a generic method for the statistical analysis of collections of anatomical shape complexes, namely sets of surfaces that were previously segmented and labeled in a group of subjects. The method estimates an anatomical model, the template complex, that is representative of the population under study. Its shape reflects anatomical invariants within the dataset. In addition, the method automatically places control points near the most variable parts of the template complex. Vectors attached to these points are parameters of deformations of the ambient 3D space. These deformations warp the template to each subject's complex in a way that preserves the organization of the anatomical structures. Multivariate statistical analysis is applied to these deformation parameters to test for group differences. Results of the statistical analysis are then expressed in terms of deformation patterns of the template complex, and can be visualized and interpreted. The user needs only to specify the topology of the template complex and the number of control points. The method then automatically estimates the shape of the template complex, the optimal position of control points and deformation parameters. The proposed approach is completely generic with respect to any type of application and well adapted to efficient use in clinical studies, in that it does not require point correspondence across surfaces and is robust to mesh imperfections such as holes, spikes, inconsistent orientation or irregular meshing.
Article
Longitudinal image analysis has become increasingly important in clinical studies of normal aging and neurodegenerative disorders. Furthermore, there is a growing appreciation of the potential utility of longitudinally acquired structural images and reliable image processing to evaluate disease modifying therapies. Challenges have been related to the variability that is inherent in the available cross-sectional processing tools, to the introduction of bias in longitudinal processing and to potential over-regularization. In this paper we introduce a novel longitudinal image processing framework, based on unbiased, robust, within-subject template creation, for automatic surface reconstruction and segmentation of brain MRI of arbitrarily many time points. We demonstrate that it is essential to treat all input images exactly the same as removing only interpolation asymmetries is not sufficient to remove processing bias. We successfully reduce variability and avoid over-regularization by initializing the processing in each time point with common information from the subject template. The presented results show a significant increase in precision and discrimination power while preserving the ability to detect large anatomical deviations; as such they hold great potential in clinical applications, e.g. allowing for smaller sample sizes or shorter trials to establish disease specific biomarkers or to quantify drug effects.
Article
With the advent of noninvasive neuroimaging, a plethora of digital human neuroanatomical atlases has been developed. The accuracy of these atlases is constrained by the resolution and signal-gathering powers of available imaging equipment. In an attempt to circumvent these limitations and to produce a high resolution in vivo human neuroanatomy, we investigated the usefulness of intrasubject registration for post hoc MR signal averaging. Twenty-seven high resolution (7 x 0.78 and 20 x 1.0 mm3) T1-weighted volumes were acquired from a single subject, along with 12 double echo T2/proton density-weighted volumes. These volumes were automatically registered to a common stereotaxic space in which they were subsampled and intensity averaged. The resulting images were examined for anatomical quality and usefulness for other analytical techniques. The quality of the resulting image from the combination of as few as five T1 volumes was visibly enhanced. The signal-to-noise ratio was expected to increase as the root of the number of contributing scans to 5.2, n = 27. The improvement in the n = 27 average was great enough that fine anatomical details, such as thalamic subnuclei and the gray bridges between the caudate and putamen, became crisply defined. The gray/white matter boundaries were also enhanced, as was the visibility of any finer structure that was surrounded by tissue of varying T1 intensity. The T2 and proton density average images were also of higher quality than single scans, but the improvement was not as dramatic as that of the T1 volumes. Overall, the enhanced signal in the averaged images resulted in higher quality anatomical images, and the data lent themselves to several postprocessing techniques. The high quality of the enhanced images permits novel uses of the data and extends the possibilities for in vivo human neuroanatomy.