Stefan Klein

Erasmus Universiteit Rotterdam, Rotterdam, South Holland, Netherlands

Are you Stefan Klein?

Claim your profile

Publications (76)141.15 Total impact

  • [Show abstract] [Hide abstract]
    ABSTRACT: Algorithms for computer-aided diagnosis of dementia based on structural MRI have demonstrated high performance in the literature, but are difficult to compare as different data sets and methodology were used for evaluation. In addition, it is unclear how the algorithms would perform on previously unseen data, and thus, how they would perform in clinical practice when there is no real opportunity to adapt the algorithm to the data at hand. To address these comparability, generalizability and clinical applicability issues, we organized a grand challenge that aimed to objectively compare algorithms based on a clinically representative multi- center data set. Using clinical practice as starting point, the goal was to reproduce the clinical diagnosis. Therefore, we evaluated algorithms for multi-class classification of three diagnostic groups: patients with probable Alzheimer’s disease, patients with mild cognitive impairment and healthy controls. The diagnosis based on clinical criteria was used as reference standard, as it was the best available reference despite its known limitations. For evaluation, a previously unseen test set was used consisting of 354 T1-weighted MRI scans with the diagnoses blinded. Fifteen research teams participated with in total 29 algorithms. The algorithms were trained on a small training set (n=30) and optionally on data from other sources (e.g., the Alzheimer’s Disease Neuroimaging Initiative, the Australian Imaging Biomarkers and Lifestyle flagship study of aging). The best performing algorithm yielded an accuracy of 63.0% and an area under the receiver-operating-characteristic curve (AUC) of 78.8%. In general, the best performances were achieved using feature extraction based on voxel-based morphometry or a combination of features that included volume, cortical thickness, shape and intensity. The challenge is open for new submissions via the web-based framework:
    NeuroImage 03/2015; · 6.13 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Carotid plaque segmentation in B-mode ultrasound (BMUS) and contrast-enhanced ultrasound (CEUS) is crucial to the assessment of plaque morphology and composition, which are linked to plaque vulnerability. Segmentation in BMUS is challenging because of noise, artifacts and echo-lucent plaques. CEUS allows better delineation of the lumen but contains artifacts and lacks tissue information. We describe a method that exploits the combined information from simultaneously acquired BMUS and CEUS images. Our method consists of non-rigid motion estimation, vessel detection, lumen–intima segmentation and media–adventitia segmentation. The evaluation was performed in training (n = 20 carotids) and test (n = 28) data sets by comparison with manually obtained ground truth. The average root-mean-square errors in the training and test data sets were comparable for media–adventitia (411 ± 224 and 393 ± 239 μm) and for lumen–intima (362 ± 192 and 388 ± 200 μm), and were comparable to inter-observer variability. To the best of our knowledge, this is the first method to perform fully automatic carotid plaque segmentation using combined BMUS and CEUS.
    Ultrasound in Medicine & Biology. 02/2015;
  • Dirk H J Poot, Stefan Klein
    [Show abstract] [Hide abstract]
    ABSTRACT: In this work we present a framework for reliably detecting significant differences in quantitative magnetic resonance imaging (MRI) and evaluate it with diffusion tensor imaging (DTI) experiments. As part of this framework we propose a new spatially regularized maximum likelihood estimator that simultaneously estimates the quantitative parameters and the spatially -smoothly- varying noise level from the acquisitions. The noise level estimation method does not require repeated acquisitions. We show that the amount of regularization in this method can be set a-priori to achieve a desired coefficient of variation of the estimated noise level. The noise level estimate allows the construction of a Cram´er-Rao-lower-bound based test statistic that reliably assesses the significance of differences between voxels within a scan or across different scans. We show that the regularized noise level estimate improves upon existing methods and results in a substantially increased precision of the uncertainty estimates of the DTI parameters. It enables correct specification of the null distribution of the test statistic and with it the test statistic obtains the highest sensitivity and specificity. The source code of the estimation framework, test statistic and experiment scripts are made available to the community.
    IEEE transactions on medical imaging. 12/2014;
  • [Show abstract] [Hide abstract]
    ABSTRACT: In standard B-mode ultrasound (BMUS), segmentation of the lumen of atherosclerotic carotid arteries and studying the lumen geometry over time are difficult owing to irregular lumen shapes, noise, artifacts, and echolucent plaques. Contrast enhanced ultrasound (CEUS) improves lumen visualization, but lumen segmentation remains challenging owing to varying intensities, CEUS-specific artifacts and lack of tissue visualization. To overcome these challenges, we propose a novel method using simultaneously acquired BMUS&CEUS image sequences. Initially, the method estimates nonrigid motion (NME) from the image sequences, using intensity-based image registration. The motion-compensated image sequence is then averaged to obtain a single 'epitome' image with improved signal-to-noise ratio. The lumen is segmented from the epitome image through an intensity joint-histogram classification and a graph-based segmentation. NME was validated by comparing displacements with manual annotations in eleven carotids. The average root-mean-squareerror (RMSE) was 112 73 μm. Segmentation results were validated against manual delineations in the epitome images of two different datasets, respectively containing eleven (RMSE 191 43 μm) and ten (RMSE 351 176 μm) carotids. From the deformation fields, we derived arterial distensibility with values comparable to the literature. The average errors in all experiments were in the inter-observer variability range. To the best of our knowledge, this is the first study exploiting combined BMUS&CEUS images for atherosclerotic carotid lumen segmentation.
    IEEE Transactions on Medical Imaging 11/2014; · 3.80 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Background To evaluate the influence of image registration on apparent diffusion coefficient (ADC) images obtained from abdominal free-breathing diffusion-weighted MR images (DW-MRIs).MethodsA comprehensive pipeline based on automatic three-dimensional nonrigid image registrations is developed to compensate for misalignments in DW-MRI datasets obtained from five healthy subjects scanned twice. Motion is corrected both within each image and between images in a time series. ADC distributions are compared with and without registration in two abdominal volumes of interest (VOIs). The effects of interpolations and Gaussian blurring as alternative strategies to reduce motion artifacts are also investigated.ResultsAmong the four considered scenarios (no processing, interpolation, blurring and registration), registration yields the best alignment scores. Median ADCs vary according to the chosen scenario: for the considered datasets, ADCs obtained without processing are 30% higher than with registration. Registration improves voxelwise reproducibility at least by a factor of 2 and decreases uncertainty (Fréchet-Cramér-Rao lower bound). Registration provides similar improvements in reproducibility and uncertainty as acquiring four times more data.Conclusion Patient motion during image acquisition leads to misaligned DW-MRIs and inaccurate ADCs, which can be addressed using automatic registration. J. Magn. Reson. Imaging 2014.
    Journal of Magnetic Resonance Imaging 11/2014; · 2.57 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: To assess (1) whether normal and degenerated menisci exhibit different T1GD on delayed gadolinium-enhanced MRI of the meniscus (dGEMRIM), (2) the reproducibility of dGEMRIM and (3) the correlation between meniscus and cartilage T1GD in knee osteoarthritis (OA) patients. In 17 OA patients who underwent dGEMRIM twice within 7 days, meniscus and cartilage T1GD was calculated. Meniscus pathology was evaluated on conventional MRI. T1GD in normal and degenerated menisci were compared using a Student's t-test. Reproducibility was assessed using ICCs. Pearson's correlation was calculated between meniscus and cartilage T1GD. A trend towards lower T1GD in degenerated menisci (mean: 402 ms; 95 % CI: 359-444 ms) compared to normal menisci (mean: 448 ms; 95 % CI: 423-473 ms) was observed (p = 0.05). Meniscus T1GD ICCs were 0.85-0.90. The correlation between meniscus and cartilage T1GD was moderate in the lateral (r = 0.52-0.75) and strong in the medial compartment (r = 0.78-0.94). Our results show that degenerated menisci have a clear trend towards lower T1GD compared to normal menisci. Since these results are highly reproducible, meniscus degeneration may be assessed within one delayed gadolinium-enhanced MRI simultaneously with cartilage. The strong correlation between meniscus and cartilage T1GD suggests concomitant degeneration in both tissues in OA, but also suggests that dGEMRIC may not be regarded entirely as sulphated glycosaminoglycan specific. • dGEMRIM T1 GD can possibly be used to assess meniscal degeneration; • dGEMRIM yields highly reproducible meniscal T1 GD in early stage osteoarthritic patients; • Concomitant degeneration of cartilage and meniscus tissue occurs in early stage osteoarthritis; • dGEMRIC cannot be regarded as entirely sulphated glycosaminoglycan specific.
    European Radiology 05/2014; · 4.34 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: To introduce a semiautomatic algorithm to perform the registration of free-hand B-Mode ultrasound (US) and magnetic resonance imaging (MRI) of the carotid artery. The authors' approach combines geometrical features and intensity information. The only user interaction consists of placing three seed points in US and MRI. First, the lumen centerlines are used as landmarks for point based registration. Subsequently, in a joint optimization the distance between centerlines and the dissimilarity of the image intensities is minimized. Evaluation is performed in left and right carotids from six healthy volunteers and five patients with atherosclerosis. For the validation, the authors measure the Dice similarity coefficient (DSC) and the mean surface distance (MSD) between carotid lumen segmentations in US and MRI after registration. The effect of several design parameters on the registration accuracy is investigated by an exhaustive search on a training set of five volunteers and three patients. The optimum configuration is validated on the remaining images of one volunteer and two patients. On the training set, the authors achieve an average DSC of 0.74 and a MSD of 0.66 mm on volunteer data. For the patient data, the authors obtain a DSC of 0.77 and a MSD of 0.69 mm. In the independent set composed of patient and volunteer data, the DSC is 0.69 and the MSD is 0.87 mm. The experiments with different design parameters show that nonrigid registration outperforms rigid registration, and that the combination of intensity and point information is superior to approaches that use intensity or points only. The proposed method achieves an accurate registration of US and MRI, and may thus enable multimodal analysis of the carotid plaque.
    Medical Physics 05/2014; 41(5):052904. · 3.01 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Atherosclerotic plaque composition can indicate plaque vulnerability. We segment atherosclerotic plaque components from the carotid artery on a combination of in vivo MRI and CT-angiography (CTA) data using supervised voxelwise classification. In contrast to previous studies the ground truth for training is directly obtained from 3D registration with histology for fibrous and lipid-rich necrotic tissue, and with [Formula: see text]CT for calcification. This registration does, however, not provide accurate voxelwise correspondence. We therefore evaluate three approaches that incorporate uncertainty in the ground truth used for training: I) soft labels are created by Gaussian blurring of the original binary histology segmentations to reduce weights at the boundaries between components, and are weighted by the estimated registration accuracy of the histology and in vivo imaging data (measured by overlap), II) samples are weighted by the local contour distance of the lumen and outer wall between histology and in vivo data, and III) 10% of each class is rejected by Gaussian outlier rejection. Classification was evaluated on the relative volumes (% of tissue type in the vessel wall) for calcified, fibrous and lipid-rich necrotic tissue, using linear discriminant (LDC) and support vector machine (SVM) classification. In addition, the combination of MRI and CTA data was compared to using only one imaging modality. Best results were obtained by LDC and outlier rejection: the volume error per vessel was 0.9[Formula: see text]1.0% for calcification, 12.7[Formula: see text]7.6% for fibrous and 12.1[Formula: see text]8.1% for necrotic tissue, with Spearman rank correlation coefficients of 0.91 (calcification), 0.80 (fibrous) and 0.81 (necrotic). While segmentation using only MRI features yielded low accuracy for calcification, and segmentation using only CTA features yielded low accuracy for necrotic tissue, the combination of features from MRI and CTA gave good results for all studied components.
    PLoS ONE 04/2014; 9(4):e94840. · 3.53 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Because hypoperfusion of brain tissue precedes atrophy in dementia, the detection of dementia may be advanced by the use of perfusion information. Such information can be obtained noninvasively with arterial spin labeling (ASL), a relatively new MR technique quantifying cerebral blood flow (CBF). Using ASL and structural MRI, we evaluated diagnostic classification in 32 prospectively included presenile early stage dementia patients and 32 healthy controls. Patients were suspected of Alzheimer's disease (AD) or frontotemporal dementia. Classification was based on CBF as perfusion marker, gray matter (GM) volume as atrophy marker, and their combination. These markers were each examined using six feature extraction methods: a voxel-wise method and a region of interest (ROI)-wise approach using five ROI-sets in the GM. These ROI-sets ranged in number from 72 brain regions to a single ROI for the entire supratentorial brain. Classification was performed with a linear support vector machine classifier. For validation of the classification method on the basis of GM features, a reference dataset from the AD Neuroimaging Initiative database was used consisting of AD patients and healthy controls. In our early stage dementia population, the voxelwise feature-extraction approach achieved more accurate results (area under the curve (AUC) range = 86 − 91%) than all other approaches (AUC = 57 − 84%). Used in isolation, CBF quantified with ASL was a good diagnostic marker for dementia. However, our findings indicated only little added diagnostic value when combining ASL with the structural MRI data (AUC = 91%), which did not significantly improve over accuracy of structural MRI atrophy marker by itself. Hum Brain Mapp, 2014. © 2014 Wiley Periodicals, Inc.
    Human Brain Mapping 04/2014; 35(9). · 6.92 Impact Factor
  • Wei Sun, Wiro J. Niessen, Stefan Klein
    [Show abstract] [Hide abstract]
    ABSTRACT: In nonrigid registration, deformations may take place on the coarse and fine scales. For the conventional B-splines based free-form deformation (FFD) registration, these coarse- and fine-scale deformations are all represented by basis functions of a single scale. Meanwhile, wavelets have been proposed as a signal representation suitable for multi-scale problems. Wavelet analysis leads to a unique decomposition of a signal into its coarse- and fine-scale components. Potentially, this could therefore be useful for image registration. In this work, we investigate whether a wavelet-based FFD model has advantages for nonrigid image registration. We use a B-splines based wavelet, as defined by Cai and Wang.1 This wavelet is expressed as a linear combination of B-spline basis functions. Derived from the original B-spline function, this wavelet is smooth, differentiable, and compactly supported. The basis functions of this wavelet are orthogonal across scales in Sobolev space. This wavelet was previously used for registration in computer vision, in 2D optical flow problems,2 but it was not compared with the conventional B-spline FFD in medical image registration problems. An advantage of choosing this B-splines based wavelet model is that the space of allowable deformation is exactly equivalent to that of the traditional B-spline. The wavelet transformation is essentially a (linear) reparameterization of the B-spline transformation model. Experiments on 10 CT lung and 18 T1-weighted MRI brain datasets show that wavelet based registration leads to smoother deformation fields than traditional B-splines based registration, while achieving better accuracy.
    SPIE Medical Imaging; 03/2014
  • [Show abstract] [Hide abstract]
    ABSTRACT: Patients with carotid atherosclerotic plaques carry an increased risk of cardiovascular events such as stroke. Ultrasound has been employed as a standard for diagnosis of carotid atherosclerosis. To assess atherosclerosis, the intima contour of the carotid artery lumen should be accurately outlined. For this purpose, we use simultaneously acquired side-by-side longitudinal contrast enhanced ultrasound (CEUS) and B-mode ultrasound (BMUS) images and exploit the information in the two imaging modalities for accurate lumen segmentation. First, nonrigid motion compensation is performed on both BMUS and CEUS image sequences, followed by averaging over the 150 time frames to produce an image with improved signal-to-noise ratio (SNR). After that, we segment the lumen from these images using a novel method based on dynamic programming which uses the joint histogram of the CEUS and BMUS pair of images to distinguish between background, lumen, tissue and artifacts. Finally, the obtained lumen contour in the improved-SNR mean image is transformed back to each time frame of the original image sequence. Validation was done by comparing manual lumen segmentations of two independent observers with automated lumen segmentations in the improved-SNR images of 9 carotid arteries from 7 patients. The root mean square error between the two observers was 0.17±0.10mm and between automated and average of manual segmentation of two observers was 0.19±0.06mm. In conclusion, we present a robust and accurate carotid lumen segmentation method which overcomes the complexity of anatomical structures, noise in the lumen, artifacts and echolucent plaques by exploiting the information in this combined imaging modality.
    SPIE Medical Imaging 2014, San Diego; 02/2014
  • Wei Sun, Wiro J Niessen, Stefan Klein
    [Show abstract] [Hide abstract]
    ABSTRACT: In traditional free-form deformation (FFD) based registration, a B-spline basis function is commonly utilized to build the transformation model. As the B-spline order increases, the corresponding B-spline function becomes smoother. However, the higher-order B-spline has a larger support region, which means higher computational cost. For a given D-dimensional nth-order B-spline, an mth-order B-spline where (m < or = n) has (m +1/n + 1)D times lower computational complexity. Generally, the third-order B-spline is regarded as keeping a good balance between smoothness and computation time. A lower-order function is seldom used to construct the deformation field for registration since it is less smooth. In this research, we investigated whether lower-order B-spline functions can be utilized for efficient registration, by using a novel stochastic perturbation technique in combination with a postponed smoothing technique to higher B-spline order. Experiments were performed with 3D lung and brain scans, demonstrating that the lower-order B-spline FFD in combination with the proposed perturbation and postponed smoothing techniques even results in better accuracy and smoothness than the traditional third-order B-spline registration, while substantially reducing computational costs.
    01/2014; 17(Pt 1):194-201.
  • Source
  • Mach Learn Med Imag; 01/2014
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Viscosupplementation with hyaluronic acid (HA) of osteoarthritic (OA) knee joints has a well-established positive effect on clinical symptoms. This effect, however, is only temporary and the working mechanism of HA injections is not clear. It was suggested that HA might have disease modifying properties because of its beneficial effect on cartilage sulphated glycosaminoglycan (sGAG) content. Delayed gadolinium-enhanced MRI of cartilage (dGEMRIC) is a highly reproducible, non-invasive surrogate measure for sGAG content and hence composition of cartilage. The aim of this study was to assess whether improvement in cartilage structural composition is detected using dGEMRIC 14 weeks after 3 weekly injections with HA in patients with early-stage knee OA. In 20 early-stage knee OA patients (KLG I-II), 3D dGEMRIC at 3T was acquired before and 14 weeks after 3 weekly injections with HA. To evaluate patient symptoms, the knee injury and osteoarthritis outcome score (KOOS) and a numeric rating scale (NRS) for pain were recorded. To evaluate cartilage composition, six cartilage regions in the knee were analyzed on dGEMRIC. Outcomes of dGEMRIC, KOOS and NRS before and after HA were compared using paired t-testing. Since we performed multiple t-tests, we applied a Bonferroni-Holm correction to determine statistical significance for these analyses. All KOOS subscales ('pain', 'symptoms', 'daily activities', 'sports' and 'quality of life') and the NRS pain improved significantly 14 weeks after Viscosupplementation with HA. Outcomes of dGEMRIC did not change significantly after HA compared to baseline in any of the cartilage regions analyzed in the knee. Our results confirm previous findings reported in the literature, showing persisting improvement in symptomatic outcome measures in early-stage knee OA patients 14 weeks after Viscosupplementation. Outcomes of dGEMRIC, however, did not change after Viscosupplementation, indicating no change in cartilage structural composition as an explanation for the improvement of clinical symptoms.
    PLoS ONE 11/2013; 8(11):e79785. · 3.53 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: To evaluate the reproducibility and sensitivity of the modified CINE inversion recovery (mCINE-IR) acquisition on rats for measuring the myocardial T1 at 7 Tesla. The recently published mCINE-IR acquisition on humans was applied on rats for the first time, enabling the possibility of translational studies with an identical sequence. Simulations were used to study signal evolution and heart rate dependency. Gadolinium phantoms, a heart specimen and a healthy rat were used to study reproducibility. Two cryo-infarcted rats were scanned to measure late gadolinium enhancement (LGE). In the phantom reproducibility studies the T1 measurements had a maximum coefficient of variation (COV) of 1.3%. For the in vivo reproducibility the COV was below 5% in the anterior cardiac segments. In simulations with phantoms and specimens, a heart rate dependency of approximately 0.5 ms/bpm was present. The T1 maps of the cryo-infarcted rats showed a clear lowering of T1 in de LGE region. The results show that mCINE-IR is highly reproducible and that the sensitivity allows detecting T1 changes in the rat myocardium.J. Magn. Reson. Imaging 2013. © 2013 Wiley Periodicals, Inc.
    Journal of Magnetic Resonance Imaging 10/2013; · 2.57 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Purpose: There is increasing evidence that epicardial fat (i.e., adipose tissue contained within the pericardium) plays an important role in the development of cardiovascular disease. Obtaining the epicardial fat volume from routinely performed non-enhanced cardiac CT scans is therefore of clinical interest. The purpose of this work is to investigate the feasibility of automatic pericardium segmentation and subsequent quantification of epicardial fat on non-enhanced cardiac CT scans.Methods: Imaging data of 98 randomly selected subjects belonging to a larger cohort of subjects who underwent a cardiac CT scan at our medical center were retrieved. The data were acquired on two different scanners. Automatic multi-atlas based method for segmenting the pericardium and calculating the epicardial fat volume has been developed. The performance of the method was assessed by (1) comparing the automatically segmented pericardium to a manually annotated reference standard, (2) comparing the automatically obtained epicardial fat volumes to those obtained manually, and (3) comparing the accuracy of the automatic results to the inter-observer variability.Results: Automatic segmentation of the pericardium was achieved with a Dice similarity index of 89.1 ± 2.6% with respect to Observer 1 and 89.2 ± 1.9% with respect to Observer 2. The correlation between the automatic method and the manual observers with respect to the epicardial fat volume computed as the Pearson's correlation coefficient (R) was 0.91 (P < 0.001) for both observers. The inter-observer study resulted in a Dice similarity index of 89.0 ± 2.4% for segmenting the pericardium and a Pearson's correlation coefficient of 0.92 (P < 0.001) for computation of the epicardial fat volume.Conclusions: The authors developed a fully automatic method that is capable of segmenting the pericardium and quantifying epicardial fat on non-enhanced cardiac CT scans. The authors demonstrated the feasibility of using this method to replace manual annotations by showing that the automatic method performs as good as manual annotation on a large dataset.
    Medical Physics 09/2013; 40(9):091910. · 3.01 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Multiresolution strategies are commonly used in the nonrigid registration to avoid local minima in the optimization space. Generally, a step-by-step hierarchical approach is adopted, in which the registration starts on a level with reduced complexity (downsampled images, global transformations), then continuing to levels with increased complexity, until the finest level is reached. In this work we propose two alternative multiresolution strategies for both the data model and transformation model, in which different resolution levels are considered simultaneously instead of subsequently. By combining the different strategies for data and transformation, we systematically define 3 3 multiresolution schemes, including both existing and novel methods. Experiments on 10 pairs of CT lung datasets showed that the best performing strategy resulted in a reduction of the upper quartile of the mean target registration error from 2mm to 1.5 mm, compared with the conventionally hierarchical multiresolution method, while achieving smoother deformations. Experiments with intersubject registration of 18 3D T1-weighted MRI brain scans confirmed that simultaneous multiresolution strategies produce more accurate registration results (median of mean overlap increased from 0.55 to 0.57) and smoother deformation fields than the traditionally hierarchical method. Evaluation of robustness indicated that the largest differences in accuracy between methods are observed for structures with a relatively large initial misalignment.
    IEEE Transactions on Image Processing 08/2013; · 3.11 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Background / Purpose: Because hypoperfusion of brain tissue precedes atrophy in dementia, the detection of dementia may be advanced by the use of perfusion information. Such information can be obtained with arterial spin labeling (ASL), a relatively new MR technique quantifying cerebral blood flow (CBF). To evaluate several classification methods for diagnosing early-stage dementia in prospectively included presenile patients and healthy controls, we examined a number of feature-extraction methods for CBF, grey matter (GM) volumes, and their combination. Main conclusion: CBF quantified with ASL was a good diagnostic marker of early dementia by itself. We found no significant added value of CBF over structural MRI as dementia marker.
    19th Annual Meeting of the Organization for Human Brain Mapping (OHBM) 2013; 07/2013
  • [Show abstract] [Hide abstract]
    ABSTRACT: Anatomical alignment in neuroimaging studies is of such importance that considerable effort is put into improving the registration used to establish spatial correspondence. Tract-based spatial statistics (TBSS) is a popular method for comparing diffusion characteristics across subjects. TBSS establishes spatial correspondence using a combination of nonlinear registration and a "skeleton projection" that may break topological consistency of the transformed brain images. We therefore investigated feasibility of replacing the two-stage registration-projection procedure in TBSS with a single, regularized, high-dimensional registration. To optimize registration parameters and to evaluate registration performance in diffusion MRI, we designed an evaluation framework that uses native space probabilistic tractography for 23 white matter tracts, and quantifies tract similarity across subjects in standard space. We optimized parameters for two registration algorithms on two diffusion datasets of different quality. We investigated reproducibility of the evaluation framework, and of the optimized registration algorithms. Next, we compared registration performance of the regularized registration methods and TBSS. Finally, feasibility and effect of incorporating the improved registration in TBSS were evaluated in an example study. The evaluation framework was highly reproducible for both algorithms (R(2) 0.993; 0.931). The optimal registration parameters depended on the quality of the dataset in a graded and predictable manner. At optimal parameters, both algorithms outperformed the registration of TBSS, showing feasibility of adopting such approaches in TBSS. This was further confirmed in the example experiment.
    NeuroImage 03/2013; · 6.13 Impact Factor

Publication Stats

1k Citations
141.15 Total Impact Points


  • 2011–2014
    • Erasmus Universiteit Rotterdam
      • Department of Medical Informatics
      Rotterdam, South Holland, Netherlands
  • 2009–2013
    • Erasmus MC
      • • Department of Radiology
      • • Department of Medical Informatics
      • • Department of Cardiology
      Rotterdam, South Holland, Netherlands
    • Universiteit Utrecht
      • Image Sciences Institute
      Utrecht, Utrecht, Netherlands
  • 2011–2012
    • Delft University Of Technology
      • • Faculty of Applied Sciences (AS)
      • • Department of Imaging Science and Technology
      Delft, South Holland, Netherlands
  • 2006–2011
    • University Medical Center Utrecht
      • Image Sciences Institute
      Utrecht, Provincie Utrecht, Netherlands