Rotate to view desktop version ...


All publications
FastSurfer - A fast and accurate deep learning based neuroimaging pipeline. L.Henschel,...M.Reuter. under review 2019. [arXiv].
FatSegNet: A Fully Automated Deep Learning Pipeline for Adipose Tissue Segmentation on Abdominal Dixon MRI. S.Estrada,...M.Reuter. Magn Reson Med. 2019 (open access). [arXiv] [DOI-Link].
Competition vs. Concatenation in Skip Connections of Fully Convolutional Networks. S.Estrada,...M.Reuter. MICCAI 2018 Workshop. [arXiv].
Complex Fully Convolutional Neural Networks for MR Image Reconstruction. M.Dedmari,...M.Reuter. MICCAI 2018 Workshop. [arXiv].
Advantages of cortical surface reconstruction using submillimeter 7 T MEMPRAGE. N.Zaretskaya et al.. NeuroImage 165:11-26, 2018. [DOI-Link].
GRAPPA reconstructed wave-CAIPI MP-RAGE at 7 Tesla. JM.Schwarz, ...M.Reuter, T.Stöcker. Magn Reson Med. 80(6):2427-2438, 2018. [DOI-Link].
A Longitudinal Imaging Genetics Study of Neuroanatomical Asymmetry in Alzheimer's Disease. C.Wachinger, ...M.Reuter, A.Riekmann. Biological Psychiatry 84(7):522-530, 2018. [DOI-Link].
Longitudinal MRI data analysis in presence of measurement error but absence of replicates. C.Ranjan et al.. IISE Trans on Healthcare Sys Eng 8(2):117-130, 2018. [DOI-Link].
Object and scene memory are differentially associated with CSF markers of Alzheimer's disease and MRI volumetry. D.Berron et al.. Alzheimer's & Dementia 13(7):P1553-P1554, 2017. [DOI-Link].
DeepNAT: Deep convolutional neural network for segmenting neuroanatomy. C.Wachinger, M.Reuter, T.Klein. NeuroImage 170:434-445, 2017. [DOI-Link].
High-resolution magnetic resonance imaging reveals nuclei of the human amygdala: manual segmentation to automatic atlas. Z.M.Saygin, et al.. NeuroImage 155:370-382, 2017. [DOI-Link].
Latent Processes Governing Neuroanatomical Change in Aging and Dementia. C.Wachinger, A.Rieckmann, M.Reuter. MICCAI 2017.
Mindboggling morphometry of human brains. A.Klein, et al.. PLoS Computational Biology 2017; 13(3):e1005350. [DOI-Link].
Mid-space-independent deformable image registration. I.Aganj, J.E.Iglesias, M.Reuter,.... NeuroImage 152:158-170, 2017. [DOI-Link].
Whole-brain Analysis Reveals Increased Neuroanatomical Asymmetries in Dementia for Hippocampus and Amygdala. C.Wachinger,... M.Reuter. Brain 139(12):3253-3266, 2016. [DOI-Link].
Multidimensional Heritability Analysis of Neuroanatomical Shape. T.Ge, M.Reuter,.... Nature Communications 7:13291, 2016. [DOI-Link].
Bayesian longitudinal segmentation of hippocampal substructures in brain MRI using subject-specific atlases. J.Iglesias,... M.Reuter. NeuroImage 141:542-555, 2016. [DOI-Link].
Domain adaptation for Alzheimer's disease diagnostics. C.Wachinger, M.Reuter. NeuroImage 139:470-479, 2016. [PDF] [DOI-Link] [PubMed]
Shape-based Classification and Domain Adaptation for Alzheimer's Disease Diagnostics. C.Wachinger,M.Reuter. Human Brain Mapping 2016, Geneva. Abstract / Poster.
Joint reconstruction of white-matter pathways from longitudinal diffusion MRI data with anatomical priors. A.Yendiki, M.Reuter.... NeuroImage 127:277-286, 2016. [DOI-Link].
Prospective motion correction with volumetric navigators (vNavs) reduces the bias and variance in brain morphometry induced by subject motion. MD.Tisdall*, M.Reuter* ...(*equal contribution). NeuroImage 127:12-22, 2016. [DOI-Link].
Responses of the human brain to mild de- and rehydration explored in vivo by 1H-MR imaging and spectroscopy. A.Biller, et al.. American Journal of Neuroradiology 36:2277-2284, 2015. [PDF] [DOI-Link] [PubMed] details

This is the first study simultaneously evaluating changes in brain tissue fluid, metabolites, volume, and cortical thickness. Our results reflect cellular volume regulatory mechanisms at a macroscopic level and emphasize that it is essential to control for hydration levels in studies on brain morphometry and metabolism in order to avoid confounding the findings. hide

Mid-space-independent symmetric data term for pairwise deformable image registration. I.Aganj, et al.. MICCAI, 2015 [PDF] [DOI-Link].
BrainPrint - A Compact Shape Description of the Brain for Identification and Classification. C.Wachinger,...,M.Reuter. Human Brain Mapping 2015, Honolulu. Abstract / Poster.
Impact of varying acquisition parameters on cortical thickness/volume derived from MEMPRAGE scans. R.Mair, M.Reuter, A.v.d.Kouwe. Human Brain Mapping 2015, Honolulu. Abstract / Poster.
Characterization of cortical surface reconstruction for sub-millimeter 7T MPRAGE using FreeSurfer. N.Zaretskaya, B.Fischl, M.Reuter, .... Human Brain Mapping 2015, Honolulu. Abstract / Poster.
Assessing atrophy measurement techniques in dementia: Results from the MIRIAD atrophy challenge. D.Cash, et al.. NeuroImage 123:149-164, 2015 [DOI-Link].
BrainPrint: A Discriminative Characterization of Brain Morphology. C.Wachinger, ..., M.Reuter. NeuroImage 109:232-248, 2015. [PDF] [BibTex] [DOI-Link] [PubMed] details

We introduce BrainPrint, a compact and discriminative representation of brain morphology to study the similarity between brains. BrainPrint captures shape information of an ensemble of cortical and subcortical structures by solving the eigenvalue problem of the 2D and 3D Laplace-Beltrami operator on triangular (boundary) and tetrahedral (volumetric) meshes. We highlight four applications for BrainPrint in this article: (i) subject identication, (ii) age and sex prediction, (iii) brain asymmetry analysis, and (iv) potential genetic influences on brain morphology. The properties of BrainPrint require the derivation of new algorithms to account for the heterogeneous mix of brain structures with varying discriminative power. We conduct experiments on three datasets, including over 3000 MRI scans from the ADNI database, 436 MRI scans from the OASIS dataset, and 236 MRI scans from the VETSA twin study. All processing steps for obtaining the compact representation are fully automated, making this processing framework particularly attractive for handling large datasets. hide

Standardized evaluation of algorithms for computer-aided diagnosis of dementia based on structural MRI: the CADDementia challenge. E.Bron, et al.. NeuroImage 111:562-579, 2015 [DOI-Link] [PubMed].
Head Motion during MRI Acquisition Reduces Gray Matter Volume and Thickness Estimates. M.Reuter*, MD.Tisdall* ...(*equal contribution). NeuroImage 107:107-115, 2015. [PDF] [BibTex] [DOI-Link] [PubMed] details

Here we show that cortical gray matter volume and thickness estimates based on MRI (FreeSurfer 5.3, VBM8 SPM, and FSL Siena 5.0.7) are biased by head motion with an average apparent volume loss of roughly 0.7%/mm/min of subject motion. Effects vary across regions and remain significant after excluding scans that fail a rigorous quality check. In view of these results, the interpretation of reported morphometric effects of movement disorders or other conditions with increased motion tendency may need to be revisited: effects may be overestimated when not controlling for head motion. Furthermore, drug studies with hypnotic, sedative, tranquilizing, or neuromuscular-blocking substances may contain spurious effects of reduced atrophy or brain growth simply because they affect motion distinct from true effects of the disease or therapeutic process. hide

Multi-Modal Robust Inverse-Consistent Linear Registration. C.Wachinger, ..., M.Reuter. Human Brain Mapping Journal 36(4):1365-1380, 2015. [PDF] [BibTex] [DOI-Link [PubMed]
Avoiding symmetry-breaking spatial non-uniformity in deformable image registration via a quasi-volume-preserving constraint. I.Aganj, M.Reuter, M.R.Sabuncu, B.Fischl. NeuroImage 106:238-251, 2015 [DOI-Link] [PubMed].
BrainPrint: Identifying Subjects by their Brain. C.Wachinger, P.Golland, M.Reuter. In Proc MICCAI, LNCS 8675:41-48, 2014. [PDF] [BibTex] [DOI-Link] [PubMed]
BrainPrint in the Computer-Aided Diagnosis of Alzheimer's Disease. C.Wachinger, ..., M.Reuter. In Proc MICCAI Challenge on Computer-Aided Diagnosis of Dementia Based on Structural MRI Data p.129-138, 2014. [PDF] [BibTex]
Gaussian Process interpolation for uncertainty estimation in image registration. C.Wachinger, P.Golland, M.Reuter, W.Wells. In Proc MICCAI, 17(Pt 1):267-274, 2014.
Effects of Sex Chromosome Dosage on Corpus Callosum Morphology in Supernumerary Sex Chromosome Aneuploidies. B.S.C.Wade, S.H.Joshi, M.Reuter, et al. Biology of Sex Differences 5:16, 2014 [DOI-Link].
Cross-validation of serial optical coherence scanning and diffusion tensor imaging: A study on neural fiber maps in human medulla oblongata. H.Wang, J.Zhu, M.Reuter, et al. NeuroImage 100:395-404, 2014. [BibTex] [DOI-Link] [PubMed]
Event Time Analysis of Longitudinal Neuroimage Data. M.R.Sabuncu, J.Bernal-Rusiel, M.Reuter, et al. NeuroImage 97:9-18, 2014. [BibTex] [DOI-Link] [PubMed] details

This paper presents a method for the statistical analysis of the associations between longitudinal neuroimaging measurements, e.g., of cortical thickness, and the timing of a clinical event of interest, e.g., disease onset. The proposed approach consists of two steps: (i) a linear mixed effects (LME) model to capture temporal variation in serial imaging data and (ii) the extended Cox regression model to examine the relationship between time-dependent imaging measurements and the timing of the event of interest. We provide a quantitative and objective empirical evaluation of the statistical performance of the proposed method on longitudinal data from subjects suffering from Mild Cognitive Impairment (MCI) at baseline. hide

Head Motion in MRI Causes Bias in Structural Brain Measurements. M.Reuter, et al. Human Brain Mapping 2014, Hamburg. Abstract / Poster.
Shape Analysis of 101 Healty Human Brains. A.Klein, et al. Human Brain Mapping 2014, Hamburg. Abstract / Poster.
Comparison of Subcortical Morphometry in Alzheimer's Disease and HIV + Subjects. B.S.C.Wade, et al. Human Brain Mapping 2014, Hamburg. Abstract / Poster.
Cortical Thickness/Volume Measured with Variable Acceleration in Young and Elderly Populations. R.Mair, M.Reuter, A.v.d.Kouwe. Human Brain Mapping 2014, Hamburg. Abstract / Poster.
Impact of MRI head placement on glioma response assessment. M.Reuter, et al. Journal of Neuro-Oncology 118(1):123-129, 2014. [PDF] [BibTex] [DOI-Link] [PubMed] details

Diagnosis of progressive disease or (partial) response during tumor treatment is based on manual size estimates of enhancing tumor area. We show that the manual area measure is highly sensitive to changes in head placement in the MRI scanner, with a root mean squared error of 22%. In our simulated study, the majority of subjects would have been incorrectly diagnosed with progressive disease without any true anatomical changes. These results highlight the urgent need for revised and more reliable response assessment criteria such as increased slice resolution, 3D volume analysis and percent change computation with respect to an average of patient specific longitudinal measurements instead of a single measurement to define progression or response. hide

The effect of amyloid pathology and glucose metabolism on cortical volume loss over time in Alzheimer's disease. S.Adriaanse, et al. European Journal of Nuclear Medicine and Molecular Imaging 41(6):1190-1198, 2014. [BibTex] [DOI-Link]
MRI Parcellation of Ex Vivo Medial Temporal Lobe. J.Augustinack, C.Magnain, M.Reuter, et al. NeuroImage 93(2):252-259, 2014. [BibTex] [DOI-Link] [PubMed] details

Brain mapping of mesocortical areas affected with neurofibrillary tangle pathology early in Alzheimer's disease progression provides not only an accurate description for location of these areas but also supplies spherical coordinates that allows comparison between other ex vivo cases and larger in vivo datasets. We have identified several cytoarchitectonic features in the medial temporal lobe with high resolution ex vivo MRI, such as the entorhinal layer II 'islands', perirhinal layer II-III columns as well as lamina of the hippocampus. Localization of Brodmann area's 28 and 35 demonstrates MRI based area boundaries validated with multiple methods and histological stains. Based on our findings, both myelin and Nissl staining relate to contrast in ex vivo MRI. Precise brain mapping serves to create modern atlases for cortical areas, allowing accurate localization with important applications to detecting early disease processes. hide

Whole brain mapping of water pools and molecular dynamics with rotating frame MR relaxation using gradient modulated low-power adiabatic pulses. O.C.Andronesi, H.Bhat, M.Reuter, et al. NeuroImage 89:92-109, 2014. [BibTex] [DOI-Link] [PubMed]
Quantitative comparison of cortical surface reconstructions from MP2RAGE and multi-echo MPRAGE data at 3 and 7 T. K.Fujimoto, et al. NeuroImage 90:60-73, 2014. [BibTex] [DOI-Link] [PubMed]
PRECREST: A Phase II Prevention and Biomarker Trial of Creatine in At-Risk for Huntington's disease. H.D.Rosas, et al. Neurology, online, 2014. [BibTex] [DOI-Link] [PubMed] details

We describe a design that preserves the autonomy of subjects not wanting genetic testing while including controls for assessing the specificity of treatment effects. Our results demonstrate the feasibility of prevention trials for HD and the safety of high-dose creatine, provide possible evidence of disease modification, support future studies of creatine, and illustrate the value of prodromal biomarkers. hide

Blockface histology with optical coherence tomography: A comparison with Nissl staining. C.Magnain, J.Augustinack, M.Reuter, et al. NeuroImage 84:524-533, 2014. [BibTex] [DOI-Link] [PubMed] details

Spectral domain optical coherence tomography (SD-OCT) is a high resolution imaging technique that generates excellent contrast based on intrinsic optical properties of the tissue, such as neurons and fibers. The SD-OCT data acquisition is performed directly on the tissue block, diminishing the need for cutting, mounting and staining and thus excluding related artifacts and distortions. We utilized SD-OCT to visualize the laminar structure of the isocortex and compared cortical cytoarchitecture with the gold standard Nissl staining, both qualitatively and quantitatively. Our results suggest that SD-OCT contains information comparable to standard histological stains such as Nissl in terms of distinguishing cortical layers and architectonic areas. We propose that SD-OCT can be used to reliably generate 3D reconstructions of multiple cubic centimeters of cortex that can be used to accurately and semi-automatically perform standard histological analyses. hide

Spatiotemporal Linear Mixed Effects Modeling for the Mass-univariate Analysis of Longitudinal Neuroimage Data. J.L.Bernal-Rusiel, M.Reuter, et al. NeuroImage 81:358-370, 2013. [BibTex] [DOI-Link] [PubMed] details

We present an extension of the Linear Mixed Effects (LME) modeling approach to be applied to the mass-univariate analysis of longitudinal neuroimaging (LNI) data. The proposed method, called spatiotemporal LME or ST-LME, exploits the spatial structure in image data. We instantiated ST-LME for the analysis of cortical surface measurements (e.g. thickness) computed by FreeSurfer. We validate the proposed ST-LME method and provide a quantitative and objective empirical comparison with alternative methods using (ADNI) and (OASIS) datasets. Our experiments revealed that ST-LME offers a dramatic gain in statistical power and repeatability of findings, while providing good control of the false positive rate. hide

Symmetric Non-Rigid Image Registration via an Adaptive Quasi-Volume-Preserving Constraint. I.Aganj, M.Reuter, et al. In Proc. of the Int. Symp. Biomedical Imaging 2013, San Francisco. [PDF].
Placement in Scanner Causes Significant Diagnostic Errors in Brain Tumor Treatment Assessment. M.Reuter, et al. Human Brain Mapping 2013, Seattle. Abstract / Poster. [PDF] details

Brain tumor treatment response is assessed by manual 2D size measurements of the enhancing tumor taken by a radiologist. Here we show the high variability of these measurements caused by different head placement (and slice orientations) in the scanner. With no anatomical changes, all subjects in this study would have been characterized as 'progresive disease' in spite of no anatomical changes. hide

Robust registration of multi-modal images. C.Wachinger, et al. Human Brain Mapping 2013, Seattle. Abstract / Poster. [PDF] details

This abstract/poster describes a novel methods for highly accurate cross modal/sequence registration in the presence of outliers (such as jaw, neck, skull strip difference or differences caused by enhancing tumor). Another application is the registration of histology images containing artifacts such as rips, tears and local deformations from the slicing procedure. The method will be made available as an extension to mri_robust_register (see [Reuter..NI 2010]) in future FreeSurfer releases. hide

Increasing Statistical Power by Modeling Spatiotemporal Correlations in Longitudinal Neuroimage Data. J.L.Bernal-Rusiel, et al. Human Brain Mapping 2013, Seattle. Abstract / Poster.
Mapping T1rho and T2rho relaxation across the whole brain with robust MRI sequences. O.Andronesi, et al. Human Brain Mapping 2013, Seattle. Abstract / Poster.
Quantitative Validation of Morphometric Data from a Rapid 2-Minute Multi-Echo MPRAGE Scan. R.Mair, et al. Human Brain Mapping 2013, Seattle. Abstract / Poster.
Cytoarchitecture of cortex imaged by Optical Coherence Tomography. C.Magnain, et al. Human Brain Mapping 2013, Seattle. Abstract / Poster.
Longitudinal FreeSurfer for Reliable Imaging Biomarkers. M.Reuter, HD.Rosas, B.Fischl. MICCAI NIBAD'12 Challenge, 12 pages, Nice, 2012. [PDF] [BibTex] details

We have presented an unbiased longitudinal processing framework previously. Here we discuss an extension to improve results in subjects with large ventricles. Furthermore, we demonstrate methods to investigate longitudinal data via scatter plots, as well as linear models and non-linear flow lines. Using normalized brain volume to estimate disease progression, we find increased atrophy rates in several structures in advanced disease stages. We also highlight how linear fits into percent volume changes can support QA and detection of early disease effects. Finally we show improved surface placement accuracy when using longitudinal image processing in cases with low image quality relative to independent processing. hide

Statistical Analysis of Longitudinal Neuroimage Data with Linear Mixed Effects Models. J.L.Bernal-Rusiel, D.N.Greve, M.Reuter et al. NeuroImage 66:249-260, 2012. [BibTex] [DOI-Link] [PubMed] details

Longitudinal neuroimaging (LNI) studies are rapidly becoming more prevalent and growing in size. However, widely used methods for the statistical analysis are often sub-optimal. Linear Mixed Effects (LME) modeling, a mature approach well known in the statistics community, offers a powerful and versatile framework for analyzing real-life LNI data. This article presents the theory behind LME models, contrasts it with other popular approaches (repeated measures ANOVA or annualized atrophy rates) in the context of LNI, provides empirical evaluation of the performance, and is accompanied with an array of computational tools that will be made freely available through FreeSurfer. hide

Exploring the biomarker cascade model: Relating cortical volume loss to [11C]PIB, [18F]FDG and MMSE over time in Alzheimer's disease patients and normal controls. S.Adriaanse, K.v.Dijk, R.Ossenkoppele, M.Reuter et al. Alzheimer's & Dementia 8(4):P348, 2012. [DOI-Link] details

Structural brain changes in the form of neuronal atrophy are important diagnostic features of Alzheimer's disease (AD) when patients present with symptoms. Amyloid-plaque formation and hypometabolism are believed to precede atrophy and followed by cognitive decline. The present longitudinal study examined how cortical volume loss over time was related to AD pathology and global cognition in both AD patients and controls. Brain volume loss over time was most strongly related to change over time in metabolism as well as to lower MMSE score at follow-up in AD patients. It also seemed to be related to amyloid plaque formation at both time-points in healthy controls, which may be an indicator of early subclinical pathological brain changes in cognitively normal subjects. hide

Registration of Histology and MRI using Blockface as Intermediate Space. M.Reuter, P.Sand, et al. Human Brain Mapping 2012, Beijing. Abstract / Poster. [PDF] details

Registering histological images with MR data is difficult because of geometric differences caused by physical deformations as well as tissue appearance. Here we use blockface images (top surface of tissue block) as intermediate space. We register the blockface volume with an MRI (3D) and then individual histological images with MR slices (2D). hide

Robust and Accurate Contralateral Registration for Pose Normalization and Tumor Segmentation. M.Reuter, H.D.Rosas, B.Fischl. Human Brain Mapping 2012, Beijing. Abstract / Poster. [PDF] details

Radiotherapy or surgery of brain tumors require prior tumor segmentation. Automatic tumor segmentation and quantitative analysis poses challenging computational problems. We aim at providing sophisticated registration procedures, expected to improve segmentation results in existing segmentation approaches and show that joint probabilities in cross-modal registration and outlier regions in robust contralateral registration can further be used to guide automatic tumor segmentation. Moreover inverse consitent contralateral registration into a half-way space results in an upright and straight head position (pose normalization). hide

A comparison of methods for non-rigid 3D shape retrieval. Z.Lian, A.Godil et al. Pattern Recognition, 46(1):449-461, 2013. [DOI-Link] details

The aim of this paper is to measure and compare the performance of state-of-the-art methods for non-rigid 3D shape retrieval. The paper develops a new benchmark consisting of 600 non-rigid 3D watertight meshes, which are equally classified into 30 categories, to carry out experiments for 11 different algorithms, whose retrieval accuracies are evaluated using 6 commonly-utilized measures. hide

Within-Subject Template Estimation for Unbiased Longitudinal Image Analysis. M.Reuter, N.J.Schmansky, H.D.Rosas, B.Fischl. NeuroImage 61(4):1402-1418, 2012. [PDF] [BibTex] [DOI-Link] [PubMed] details

Challenges in longitudinal image analysis 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 for FreeSurfer, based on unbiased, robust, within-subject template creation, for automatic surface reconstruction and segmentation of brain MRI of arbitrarily many time points. 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. hide

Volumetric Navigators for Prospective Motion Correction and Selective Reacquisition in Neuroanatomical MRI. M.D.Tisdall, A.T.Hess, M.Reuter et al. Magnetic Resonance in Medicine 68(2):389-399, 2012. [BibTex] [DOI-Link] [PubMed] details

We introduce a novel method of prospectively compensating for subject motion in neuroanatomical imaging. Short 3D EPI volumetric navigators (vNavs) are embedded in a long 3D sequence, and the resulting image volumes registered to provide an estimate of the subject's location in the scanner, giving a motion-corrected sequence with no extra time requirements. We also demonstrate motion-driven selective reacquisition of k-space to further compensate for subject motion. The improved final output and lack of additional hardware requirements make this method suitable for high-throughput environments. hide

A tale of two factors: What determines the rate of progression in Huntington's disease? A longitudinal MRI study. H.D.Rosas, M.Reuter et al. Movement Disorders 26(9):1691-1697, 2011. [BibTex] [DOI-Link] [PubMed] details

We evaluated progression in Huntington's disease using a novel longitudinal magnetic resonance imaging analysis. Patients who developed symptoms earlier demonstrated the most rapid rates of atrophy compared with those who developed symptoms during middle age or more advanced age. Rates of regional brain atrophy seem to be influenced by the age of onset of Huntington's disease symptoms and are only partially explained by CAG repeat length. These findings suggest that other genetic, epigenetic, and environmental factors play important roles in neurodegeneration in Huntington's disease. hide

Comparison of cortical surface reconstructions from MP2RAGE data at 3T and 7T. K.Fujimoto, J. Polimeni et al. Human Brain Mapping 2011, Quebec City. Abstract / Poster.
Unbiased Longitudinal Processing of Structural MRI Data in FreeSurfer. M.Reuter, H.D.Rosas, B.Fischl. Human Brain Mapping 2011, Quebec City. Abstract / Poster. [PDF]
The dynamics of cortical and hippocampal atrophy in Alzheimer's disease. M.R.Sabuncu et al. Archives of Neurology 68(8):1040-1048, 2011. [BibTex] [PDF] [DOI-Link] [PubMed] details

Measure: We used automated tools to compute annual longitudinal atrophy in the hippocampus and cortical regions targeted in AD. We used Mini-Mental State Examination scores as a measure of cognitive performance. We performed a cross-subject analysis of atrophy rates and acceleration on individuals with an AD-like cerebrospinal fluid molecular profile. Conclusion: Alzheimer disease specific cortical thinning and hippocampal volume loss are consistent with a sigmoidal pattern, with an acceleration phase during the early stages of the disease. Clinical trials should carefully consider the nonlinear behavior of these AD biomarkers. hide

SHREC’11 Track: Shape Retrieval on Non-rigid 3D Watertight Meshes. Z.Lian, A.Godil et al. In Proceedings of the Eurographics Workshop on 3D Object Retrieval, pp. 79-88, 2011. [PDF] [BibTex] [EG Pub] [SHREC'11-NonRigid] details

Non-rigid 3D shape retrieval has become an important research topic in content-based 3D object retrieval. The aim of this track is to measure and compare the performance of non-rigid 3D shape retrieval methods implemented by different participants around the world. The track is based on a new non-rigid 3D shape benchmark, which contains 600 watertight triangle meshes that are equally classified into 30 categories. In this track, 25 runs have been submitted by 9 groups and their retrieval accuracies were evaluated using 6 commonly-utilized measures. Our method ShapeDNA (introduced at SPM in 2005, see below) has reached the second place together with a method by Lian et al., who also organized this competition. The first place was reached by Smeets et al. hide

Avoiding Asymmetry-Induced Bias in Longitudinal Image Processing. M.Reuter, B.Fischl. NeuroImage 57(1):19-21, invited comment, 2011. [PDF] [BibTex] [DOI-Link] [PubMed] details

Longitudinal image processing procedures frequently transfer or pool information across time within subject, with the dual goals of reducing the variability and increasing the accuracy of the derived measures. In this note, we discuss common difficulties in longitudinal image processing, focusing on the introduction of bias, and describe the approaches we have taken to avoid them in the FreeSurfer longitudinal processing stream. hide

Selective Disruption of the Cerebral Neocortex in Alzheimer's Disease. R.S.Desikan, M.R. Sabuncu, N.J. Schmansky, M. Reuter, H.J. Cabral et al. PloS ONE 5(9):e12853, 2010. [BibTex] [PDF] [DOI-Link] [PubMed] details

Alzheimer's disease (AD) and its transitional state mild cognitive impairment (MCI) are characterized by amyloid plaque and tau neurofibrillary tangle (NFT) deposition within the cerebral neocortex and neuronal loss within the hippocampal formation. However, the precise relationship between pathologic changes in neocortical regions and hippocampal atrophy is largely unknown. This longitudinal study combines structural MRI scans and automated image analysis tools (longitudinal FreeSurfer) with change of the above measures. We examined the relationship between the presence of Alzheimer's pathology, gray matter thickness of select neocortical regions, and hippocampal volume in cognitively normal older participants and individuals with MCI and AD (n = 724). Conclusion: Cortical A beta and tau pathology affects gray matter thinning within select neocortical regions and potentially contributes to downstream hippocampal degeneration. Neocortical Alzheimer's pathology is evident even amongst older asymptomatic individuals suggesting the existence of a preclinical phase of dementia. hide

Highly Accurate Inverse Consistent Registration: A Robust Approach. M. Reuter, H.D. Rosas, B. Fischl. NeuroImage 53(4):1181-1196, 2010. [PDF] [BibTex] [DOI-Link] [PubMed] details

The registration of images is a task that is at the core of many applications in computer vision. In computational neuroimaging where the automated segmentation of brain structures is frequently used to quantify change, a highly accurate registration is necessary for motion correction of images taken in the same session, or across time in longitudinal studies where changes in the images can be expected. This paper, inspired by Nestares and Heeger (2000), presents a method based on robust statistics to register images in the presence of differences, such as jaw movement, differential MR distortions and true anatomical change. The approach we present guarantees inverse consistency (symmetry), can deal with different intensity scales and automatically estimates a sensitivity parameter to detect outlier regions in the images. The resulting registrations are highly accurate due to their ability to ignore outlier regions and show superior robustness with respect to noise, to intensity scaling and outliers when compared to state-of-the-art registration tools such as FLIRT (in FSL) or the coregistration tool in SPM. hide

Accurate Inverse Consistent Robust Registration. M. Reuter, H.D. Rosas, B. Fischl. Human Brain Mapping 2010, Barcelona. Abstract / Poster. [PDF] details

In order to accurately study disease effects or disease-modifying therapies using computational anatomy, anatomical change needs to be quantified in medical images. Since the object of interest is usually located differently in each acquired image, geometric transformations are necessary to register the input images into a common space. A highly accurate registration is desirable for motion correction of images taken in the same session, across time in longitudinal studies or simply to initialize non-linear warps. This work describes a method based on robust statistics to register images in the presence of differences, such as jaw movement, differential MR distortions and atrophy. Our approach guarantees inverse consistency (symmetry) and can deal with different intensity scales. The resulting registrations are highly accurate and show superior robustness with respect to noise, intensity scaling and outliers when compared to state-of-the-art registration tools such as FLIRT (in FSL) or the coregistration tool in SPM. hide

Unbiased Robust Template Estimation for Longitudinal Analysis in FreeSurfer. M. Reuter, H.D. Rosas, B. Fischl. Human Brain Mapping 2010, Barcelona. Abstract / Poster. [PDF] details

Compared with cross-sectional studies, a longitudinal design can significantly reduce the confounding effect of inter-individual morphological variability by using each subject as his or her own control. As a result, longitudinal imaging studies are increasing in popularity in various aspects of neuroscience. Changes in gray matter that makes up the cortical sheet are for example manifested in aging, Alzheimer's disease, Huntington's disease, multiple sclerosis and schizophrenia. In vivo cortical thickness measures could be useful as marker of disease progression or onset. Longitudinal imaging-based biomarkers are thus of great potential utility in evaluating the efficiency of disease-modifying therapies. For these reasons, developing more robust and reliable measures of cortical, subcortical and white matter atrophy may have a profound clinical impact. The current methods that utilize cross-sectional approaches, in which images are processed individually introduce the natural variability of the brain as a confound. We sought to develop and validate a longitudinal approach that takes advantage of intra-subject longitudinal acquired scans to improve the sensitivity and reliability of automatic neuro-imaging morphometic measures. hide

Hierarchical Shape Segmentation and Registration via Topological Features of Laplace-Beltrami Eigenfunctions. M. Reuter. International Journal of Computer Vision 89(2):287-308, 2010 (online 08/09). [PDF] [BibTex] [DOI-Link] [Pubget] details

This work introduces a method to hierarchically segment articulated shapes into meaningful parts and to register these parts across populations of near-isometric shapes (e.g. head, arms and legs of humans in different body postures). The method exploits the isometry invariance of eigenfunctions of the Laplace-Beltrami operator and uses topological features (level sets at important saddles) for the segmentation. Concepts from persistent homology are employed for a hierarchical representation, for the elimination of topological noise and for the comparison of eigenfunctions. The obtained parts can be registered via their spectral projection across a population of near isometric shapes. This work also presents the highly accurate computation of eigenfunctions and eigenvalues with cubic FEM on triangle meshes and discusses the construction of persistence diagrams from the Morse-Smale complex as well as the relation to size functions. hide

US Patent Application (US-2009-0169050-A1). F.-E. Wolter, M. Reuter and N. Peinecke. Method for Characterization of Objects, pending, July 2009.
Discrete Laplace-Beltrami Operators for Shape Analysis and Segmentation. M. Reuter, S. Biasotti, D. Giorgi, G. Patane, M. Spagnuolo. SMI'09, Computers & Graphics 33(3):381-390, 2009. [PDF] [BibTex] [DOI-Link] details

In this paper, we first analyze different common discretizations of the Laplace-Beltrami operator (geometric Laplacians, linear and cubic FEM operators) in terms of the correctness of their eigenfunctions with respect to the continuous case. We then present the family of segmentations induced by the nodal sets of the eigenfunctions, discussing its meaningfulness for shape understanding. hide

Laplace-Beltrami Eigenvalues and Topological Features of Eigenfunctions for Statistical Shape Analysis.
Invited Paper:
M. Reuter, F.-E. Wolter, M. Shenton, M. Niethammer. Computer-Aided Design 41(10):739-755, 2009. [PDF] [BibTex] [DOI-Link] [PubMed] [Pubget] details

This paper proposes the use of the surface based Laplace-Beltrami and the volumetric Laplace eigenvalues and -functions as shape descriptors for the comparison and analysis of shapes. These spectral measures are isometry invariant and therefore allow for shape comparisons with minimal shape pre-processing. In particular, no registration, mapping, or remeshing is necessary. The discriminatory power of the 2D surface and 3D solid methods is demonstrated at a population of female caudate nuclei (a subcortical gray matter structure of the brain, involved in memory function, emotion processing, and learning) of normal control subjects and of subjects with schizotypal personality disorder. The behavior and properties of the Laplace-Beltrami eigenvalues and -functions are discussed extensively for both the Dirichlet and Neumann boundary condition showing advantages of the Neumann vs. the Dirichlet spectra in 3D. Furthermore, topological analyses employing the Morse-Smale complex (on the surfaces) and the Reeb graph (in the solids) are performed on selected eigenfunctions, yielding shape descriptors, that are capable of localizing geometric properties and detecting shape differences by indirectly registering topological features such as critical points, level sets and integral lines of the gradient field across subjects. The use of these topological features of the Laplace-Beltrami eigenfunctions in 2D and 3D for statistical shape analysis is novel. hide

Global Medical Shape Analysis Using the Volumetric Laplace Spectrum. M. Reuter, M. Niethammer, F.-E. Wolter, S. Bouix, M. Shenton. Proceedings of the 2007 Int. Conf. on Cyberworlds, NASA-GEM Workshop, IEEE Computer Society, pp.417-426, 2007. [PDF] [BibTex] [DOI-Link] details

This paper proposes to use the volumetric Laplace spectrum as a global shape descriptor for medical shape analysis. The approach allows for shape comparisons using minimal shape preprocessing. In particular, no registration, mapping, or remeshing is necessary. All computations can be performed directly on the voxel representations of the shapes. The discriminatory power of the method is tested on a population of female caudate shapes (subcortical gray matter structure of the brain, involved in memory function, emotion processing, and learning) of normal control subjects and of subjects with schizotypal personality disorder. The behavior and properties of the volumetric Laplace spectrum are discussed extensively for both the Dirichlet and Neumann boundary condition showing advantages of the Neumann spectra. Both, the computations of spectra on 3D voxel data for the purpose of shape matching as well as the use of the Neumann spectrum for shape analysis are completely new. hide

Global Medical Shape Analysis using the Laplace-Beltrami Spectrum. M. Niethammer, M. Reuter, F.-E. Wolter, S. Bouix, N. Peinecke, M.-S. Koo, M. Shenton. MICCAI07, 10th International Conference on Medical Image Computing and Computer Assisted Intervention, Part I, LNCS 4791, pp. 850-857, 2007. [PDF] [BibTex] [DOI-Link] [PubMed] details

This paper proposes to use the Laplace-Beltrami spectrum (LBS) as a global shape descriptor for medical shape analysis. The approach allows for shape comparisons using minimal shape preprocessing. In particular, no registration, mapping, or remeshing is necessary. The discriminatory power of the method is tested on a population of female caudate shapes of normal control subjects and of subjects with schizotypal personality disorder. hide

Can one hear Shape?. M. Reuter. PAMM Proceedings of GAMM07 and ICIAM07, Vol 7, Issue 1, 6th International Congress of Industrial and Applied Mathematics, SIAM, October 2008. [PDF] [BibTex] [DOI-Link] details

The question "Can one hear the shape of a drum" has been asked in several contexts before (e.g., by Bers and Kac). It is a pictorial way of asking if the eigenvalues of the Laplacian on a given domain completely characterize its shape, in other words, if the spectrum is a complete shape descriptor (which it is not in general). In this talk we will give an overview on how the computation of the spectra can be accomplished using FEM for manifolds in 2D and 3D (e.g. iso-surfaces, boundary representations, solid bodies, vector fields...) with the Dirichlet and Neumann boundary condition. We demonstrate that it is computational feasible to numerically extract geometric properties (volume, area, boundary length and even the Euler characteristic) from the first eigenvalues. Since the spectrum contains geometrical information and since it is an isometry invariant and therefore independent of the object's representation, parametrization, spatial position, and optionally of its size, it is optimally suited to be used as a fingerprint (Shape-DNA) in contemporary computer graphics applications like database retrieval, quality assessment, and shape matching in fields like CAD, medicine or engineering. hide

Solving Nonlinear Polynomial Systems in the Barycentric Bernstein Basis. M. Reuter, T. Mikkelsen, E. Sherbrooke, T. Maekawa, N. Patrikalakis. The Visual Computer 24(3):187-200, January 2007. [PDF] [BibTex] [DOI-Link] details

We present a method for solving arbitrary systems of N nonlinear polynomials in n variables over an n-dimensional simplicial domain based on polynomial representation in the barycentric Bernstein basis and subdivision. The roots are approximated to arbitrary precision by iteratively constructing a series of smaller bounding simplices. We use geometric subdivision to isolate multiple roots within a simplex. An algorithm implementing this method in rounded interval arithmetic is described and analyzed. We find that when the total order of polynomials is close to the maximum order of each variable, an iteration of this solver algorithm is asymptotically more efficient than the corresponding step in a similar algorithm which relies on polynomial representation in the tensor product Bernstein basis. We also discuss various implementation issues and identify topics for further study. hide

Geometric Modeling for Engineering Applications. F.-E. Wolter, M. Reuter and N. Peinecke. Encyclopedia of Computational Mechanics, vol 1, chap 16, 2007. [BibTex] [DOI-Link]
International Patent Application (PCT/EP2007/000418). M. Reuter, F.-E. Wolter and N. Peinecke. Shape Optimization and efficient FEM computation employing the Medial Axis, 2007. details

This patent describes a medial axis (MA) representation and parametrization of shapes for an accurate representation of the domain with possibly curved boundary. The MA represenation can be used to model, mesh and optimize the shape, due to its skeletal structure that follows the trend of the object and describes the local thickness. The proposed method closes the design optimization cycle and bridges the gap between the design, mesh construction and FEM analysis components for a complete and efficient automation. hide

Laplace Spectra as Fingerprints for Image Recognition. N. Peinecke, F.-E. Wolter and M. Reuter. Computer-Aided Design 39(6):460-476, June 2007. [PDF] [BibTex] [DOI-Link] details

In the area of image retrieval from data bases and for copyright protection of large image collections there is a growing demand for unique but easily computable fingerprints for images. These fingerprints can be used to quickly identify every image within a larger set of possibly similar images. This paper introduces a novel method to automatically obtain such fingerprints from an image. It is based on a re-interpretation of an image as a Riemannian manifold. This representation is feasible for gray value images and color images. We discuss the use of the spectrum of eigenvalues of different variants of the Laplace operator as a fingerprint and show the usability of this approach in several use cases. Contrary to existing works in this area we do not only use the discrete Laplacian, but also with a particular emphasis the underlying continuous operator. This allows better results in comparing the resulting spectra and deeper insights in the problems arising. We show how the well known discrete Laplacian is related to the continuous Laplace-Beltrami operator. Furthermore we introduce the new concept of solid height functions to overcome some potential limitations of the method. hide

Laplace Spectra for Shape Recognition. M. Reuter. Books on Demand, ISBN 3-8334-5071-1, July 2006. [BibTex] [Amazon US] [Amazon DE] details

This books gives a detailed overview on the mathematical background of the Laplace Beltrami operator (LBO) for a Riemannian manifold. A few analytical computations are presented for special domains in 2D and 3D. Furthermore the numerical computation of the spectra of the LBO are described using a special method for planar domains and the FEM method with up to cubic form functions for the general setup in 2D and 3D. Since the spectrum of the LBO is an isometry invariant, it is invariant under translation, rotation, and invariant under a change of parametrization. It is therefore possible to construct complex objects by gluing several parameter spaces to each other. Some examples are presented, also employing the medial axis as a tool to guide the parametrization of any 2D domain in a way that ensures good meshes without any error at the boundary. Furthermore this book describes the extraction of geometric data from the spectrum. For this purpose the rapid convergence of the Heat trace is shown. Finally several applications and examples are presented to employ the spectrum in the field of shape recognition. Examples of isospectral 3D solids are shown, that can be distinguished by the spectrum of their boundary surface. The robustness is demonstrated when objects on different mesh resolutions are successfully compared. It is shown how the Neumann boundary condition can be used instead of the Dirichlet condition in cases where the boundary is not supposed to play an important role. finally several complex objects are matched successfully employing their Laplace-Beltrami spectrum. hide

Laplace-Beltrami spectra as "Shape-DNA" of surfaces and solids.
Most Cited Paper Award: [2009_most_cited_cad.pdf]
M. Reuter, F.-E. Wolter and N. Peinecke. Computer-Aided Design 38(4):342-366, 2006. [PDF] [BibTex] [DOI-Link] details

This paper describes in detail a method to extract fingerprints of any surface or solid object by taking the eigenvalues of its respective Laplace-Beltrami operator. Since the spectrum is an isometry invariant it is independent of the objects representation including parametrization, spatial position and of its representation (e.g. NURBS or any parametrized or implicitly represented surface or even for polyhedra). Therefore checking if two objects are isometric needs no prior alignment (registration / localization) of the objects, but only a comparison of their (normalized) spectra, the Shape-DNA. In this paper we describe the computation of the Shape-DNA and their comparison. We also give an overview on implementation issues like meshing, numerics and the used "Atlas" data structure allowing principally to compute eigenvalues of topologically complex objects by gluing surface patches. Exploiting the isometry invariance of the Laplace-Beltrami operator we succeed in computing eigenvalues for smoothly bounded objects without discretization errors caused by approximation of the boundary. Furthermore we show the rapid convergence of the heat trace series and demonstrate that it is computationally feasible to extract geometrical data such as the volume, the boundary length and even the Euler characteristic from the numerically calculated fingerprint data. This fact not only confirms the exactness of our computed eigenvalues, but also underlines the geometrical importance of the spectrum. With the help of the here described Shape-DNA it is possible to support copyright protection, database retrieval and quality assessment of digital data representing surfaces and solids. hide

Laplace-Spectra as Fingerprints for Shape Matching. M. Reuter, F.-E. Wolter and N. Peinecke. Proceedings of the ACM Symposium on Solid and Physical Modeling, pp.101-106, June 2005. [PDF] [BibTex] [DOI-Link] details

This paper introduces a method to extract fingerprints of any surface or solid object by taking the eigenvalues of its respective Laplace-Beltrami operator. Using an object's spectrum (i.e. the family of its eigenvalues) as a fingerprint for its shape is motivated by the fact that the related eigenvalues are isometry invariants of the object. Employing the Laplace-Beltrami spectra (not the spectra of the mesh Laplacian) as fingerprints of surfaces and solids is a novel approach in the field of geometric modeling and computer graphics. Those spectra can be calculated for any representation of the geometric object (e.g. NURBS or any parametrized or implicitly represented surface or even for polyhedra). Since the spectrum is an isometry invariant of the respective object this fingerprint is also independent of the spatial position. Additionally the eigenvalues can be normalized so that scaling factors for the geometric object can be obtained easily. Therefore checking if two objects are isometric needs no prior alignment (registration / localization) of the objects, but only a comparison of their spectra. With the help of such fingerprints it is possible to support copyright protection, database retrieval and quality assessment of digital data representing surfaces and solids. hide

International Patent Application (PCT/DE06/00857). F.-E. Wolter, M. Reuter and N. Peinecke. Fingerprints - Verfahren zur Charaketerisierung von Objekten, Nov. 2005. details

This patent describes a method to identify and compare shape of solids, surfaces and images by using the spectrum of the Laplace-Beltrami operator of the given object as a fingerprint. Please see our corresponding publications for a detailed description of the properties and the numerical computation of these spectra for the different kinds of objects. Possible applications like "Copyright protection", "Database retrieval", "Shape Matching" and "Quality Assessment" are described, making use of the fact that the spectrum is an isometry invariant and therefore completely independent of the objects representation and spatial position. Furthermore it is shown how the spectra can be compared independently of the objects size, if desired. hide

Geometric Modeling of Complex Shapes and Engineering Artifacts. F.-E. Wolter, N. Peinecke and M. Reuter. Encyclopedia of Computational Mechanics, vol 1, chap 16, pp. 475-495,2004. [BibTex] details

This work gives an overview on representation and processing of geometric models. Volume based representations (e.g. voxel rep.) and boundary representations (planar polygons, parametric surfaces defined by a map from 2D-space to 3D-space, especially spline surfaces and trimmed surfaces, multiresolutionally represented surfaces, e.g. wavelet-based surfaces and surfaces obtained by subdivision schemes) as well as explicit or implicit mathematical object representations are described and compared. The rather new method of "Medial Modeling", where an object is described by its medial axis and an associated radius function, is also presented. This medial modeling concept developed at the Welfenlab yields a very intuitive user interface useful for solid modeling, and also gives as a by-product a natural meshing of the solid for FEM computations. Finally additional attributes that can be attached to an object, i.e. attributes of physical origin or logical attributes, are discussed. hide

Diploma Thesis: Spectrum of the Laplace-Beltrami Operator for Surfaces. M. Reuter. (german) - 2001. details

Subject of this thesis is the numerical computation of the eigenvalues of the Laplace-Beltrami operator employing the Finite-Element-Method (FEM) with quadratic and cubic form functions on triangular elements. The Laplace-Beltrami operator is the natural extension of the Laplacian for Riemannian manifolds (such as curved surfaces). First an overview on the necessary background on differential equations, eigenvalue problems, FEM (variational formulation, Galerkin method) and the form functions of higher degree is given. The numbering of the vertices using the Cuthill algorithm, a mesh refinement method and meshing techniques are described. A similarity invariant for triangles is introduced and used as a measure for the quality of a triangle. The structure of the "Laplace" software project as a powerful tool to compute Laplace-Beltrami spectra on parametrized surfaces is explained. Numerical computations of the spectra of several surfaces in R² and R³ are presented and the dependency of the eigenvalues on deformations of the surface and on the triangulation is analyzed. hide

Junior Thesis: Fractal Dimension of Grayscale Images. M. Reuter. (german) - 1999. [PDF] details

Nature is rich in highly irregular structures such as trees, clouds, flashes or coast lines. They can absolutely not be described by simple geometric objects but rather embody a new level of geometry. With the help of a family of scale invariant "fractals" these irregular structures can be described. The fractal dimension measures the degree of irregularity that stays constant on any resolution of the object. The fractal dimension presents a good measure of the roughness of objects or textures. This concept can be transfered to gray scale images to get a measure of the roughness of the depicted texture. This work presents and compares different techniques to compute the fractal dimension for discrete data to classify gray scale images concerning their roughness. For this purpose the program "FDim" is introduced and described. hide

Seminar Paper: Knot Theorie - Polynom Invariants. M. Reuter. (german). [PDF] details

This paper gives an overview on one of the most successful methods to detect the dissimilarity of knots. A polynomial is assigned to every knot. Since these polynomials can be computed from any knot projection and since they always stay the same for the same knot they are called knot invariants. If two such polynomials differ, it is sure that they cannot describe the same knot. First the Jones polynomial is derived together with the prove that it is a knot invariant (using the Reidemeister moves). Afterwards a closer look is taken at alternating knots and their invariants. The final section describes the Alexander and the HOMFLY polynomials and their properties. hide
Martin Reuter - MIT - Cambridge, MA, USA - EMail:
Last modified 14.01.2022 -- [xhtml] -- [css] -- [welcome]