Amalia Cifor

Smoothness-guided 3-D reconstruction of 2-D histological images

This paper tackles two problems: (1) the reconstruction of 3-D volumes from 2-D post-mortem slices (e.g., histology, autoradiography, immunohistochemistry) in the absence of external reference, and (2) the quantitative evaluation of the 3-D reconstruction. We note that the quality of a reconstructed volume is usually assessed by considering the smoothness of some reconstructed structures of interest (e.g., the gray-white matter surfaces in brain images). Here we propose to use smoothness as a means to drive the reconstruction process itself. From a pair-wise rigid reconstruction of the 2-D slices, we first extract the boundaries of structures of interest. Those are then smoothed with a min-max curvature flow confined to the 2-D planes in which the slices lie. Finally, for each slice, we estimate a linear or flexible transformation from the sparse displacement field computed from the flow, which we apply to the original 2-D slices to obtain a smooth volume. In addition, we present a co-occurrence matrix-based technique to quantify the smoothness of reconstructed volumes. We discuss and validate the application of both our reconstruction approach and the smoothness measure on synthetic examples as well as real histological data.

The 2014 liver ultrasound tracking benchmark

Abstract The Challenge on Liver Ultrasound Tracking (CLUST) was held in conjunction with the MICCAI 2014 conference to enable direct comparison of tracking methods for this application. This paper reports the outcome of this challenge, including setup, methods, results and experiences. The database included 54 2D and 3D sequences of the liver of healthy volunteers and tumor patients under free breathing. Participants had to provide the tracking results of 90% of the data (test set) for pre-defined point-landmarks (healthy volunteers) or for tumor segmentations (patient data). In this paper we compare the best six methods which participated in the challenge. Quantitative evaluation was performed by the organizers with respect to manual annotations. Results of all methods showed a mean tracking error ranging between 1.4 mm and 2.1 mm for 2D points, and between 2.6 mm and 4.6 mm for 3D points. Fusing all automatic results by considering the median tracking results, improved the mean error to 1.2 mm (2D) and 2.5 mm (3D). For all methods, the performance is still not comparable to human inter-rater variability, with a mean tracking error of 0.5–0.6 mm (2D) and 1.2–1.8 mm (3D). The segmentation task was fulfilled only by one participant, resulting in a Dice coefficient ranging from 76.7% to 92.3%. The CLUST database continues to be available and the online leader-board will be updated as an ongoing challenge.

Registration of 3D fetal neurosonography and MRI

Graphical abstract Highlights • A novel method for affine registration of fetal neurosonography and brain MRI proposed.• Conversion of fetal MRI into pseudo US image described.• All data were successfully aligned using robust block-matching approach.• Average of 27 US volumes revealed near-complete anatomy of the fetal brain.

Smooth 3-D Reconstruction for 2-D Histological Images

We present an image driven approach to the reconstruction of 3-D volumes from stacks of 2-D post-mortem sections (histology, cryoimaging, autoradiography or immunohistochemistry) in the absence of any external information. We note that a desirable quality of the reconstructed volume is the smoothness of its notable structures (e.g. the gray/white matter surfaces in brain images). Here we propose to use smoothness as a means to drive the reconstruction process itself. From an initial rigid pair-wise reconstruction of the input 2-D sections, we extract the boundaries of structures of interest. Those are then evolved under a mean curvature flow modified to constrain the flow within 2-D planes. Sparse displacement fields are then computed, independently for each slice, from the resulting flow. A variety of transformations, from globally rigid to arbitrarily flexible ones, can then be estimated from those fields and applied to the individual input 2-D sections to form a smooth volume. We detail our method and discuss preliminary results on both real histological data and synthetic examples.

Registration of 3d fetal brain US and MRI

We propose a novel method for registration of 3D fetal brain ultrasound and a reconstructed magnetic resonance fetal brain volumes. The reconstructed MR volume is first segmented using a probabilistic atlas and an ultrasound-like image volume is simulated from the segmentation of the MR image. This ultrasound-like image volume is then affinely aligned with real ultrasound volumes of 27 fetal brains using a robust block-matching approach which can deal with intensity artefacts and missing features in ultrasound images. We show that this approach results in good overlap of four small structures. The average of the co-aligned US images shows good correlation with anatomy of the fetal brain as seen in the MR reconstruction.

Hybrid feature-based Log-Demons registration for tumour tracking in 2-D liver ultrasound images

Traditional intensity-based registration methods are often insufficient for tumour tracking in time-series ultrasound, where the low signal-to-noise ratio significantly degrades the quality of the output images, and topological changes may occur as the anatomical structures slide in and out of the focus plane. To overcome these issues, we propose a hybrid feature-based Log-Demons registration method. The novelty of our approach lies in estimating a hybrid update deformation field from demons forces that carry voxel-based local information and regional spatial correspondences yielded by a block-matching scheme within the diffeomorphic Log-Demons framework. Instead of relying on intensities alone to drive the registration, we use multichannel Log-Demons, with channels representing features like intensity, local phase and phase congruency. Results on clinical data show that our method successfully registers various patient-specific cases, where the tumours are of variable visibility, and in the presence of shadows and topological changes.

Automated Colorectal Tumour Segmentation in DCE-MRI Using Supervoxel Neighbourhood Contrast Characteristics

Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) is a powerful protocol for assessing tumour progression from changes in tissue contrast enhancement. Manual colorectal tumour delineation is a challenging and time consuming task due to the complex enhancement patterns in the 4D sequence. There is a need for a consistent approach to colorectal tumour segmentation in DCE-MRI and we propose a novel method based on detection of the tumour from signal enhancement characteristics of homogeneous tumour subregions and their neighbourhoods. Our method successfully detected 20 of 23 cases with a mean Dice score of 0.68 +/- 0.15 compared to expert annotations, which is not significantly different from expert inter-rater variability of 0.73 +/- 0.13 and 0.77 +/- 0.10. In comparison, a standard DCE-MRI tumour segmentation technique, fuzzy c-means, obtained a Dice score of 0.28 +/- 0.17.

Rigid Registration of Untracked Freehand 2D Ultrasound Sweeps to 3D CT of Liver Tumours

We present a rigid registration framework for freehand 2D ultrasound sweeps to 3D CT of liver tumours. The method registers the 2D sweeps in a group-wise manner, without the need for prior 3D ultrasound compounding or probe tracking during acquisition. We first introduce a specific acquisition model to keep the dimension of this problem reasonable. Only seven parameters are indeed required to register the images. These are estimated using simulated annealing optimization of a robust modality-independent similarity measure. The framework contrasts the current methods that rely on tracking devices and phantom calibration, which are often difficult to use routinely in clinical practice. Our results on both synthetic and real data show that the method is well-suited for such ultrasound-CT registration of liver tumours.

A Semi-automated Toolkit for Analysis of Liver Cancer Treatment Response Using Perfusion CT

Delineation of hepatic tumours is challenging in CT due to limited inherent tissue contrast, leading to significant intra-/inter-observer variability. Perfusion CT (pCT) allows quantitative assessment of enhancement patterns in normal and abnormal liver. This study aims to develop a semi-automated perfusion analysis toolkit that classifies hepatic tissue based on perfusion-derived parameters. pCT data from patients with hepatic metastases were used in this study. Tumour motion was minimized through image registration; perfusion parameters were derived and then employed in the training of a machine learning algorithm used to classify hepatic tissue. This method was found to deliver promising results for 10 data sets, with recorded sensitivity and specificity of the tissue classification in the ranges of 0.92–0.99 and 0.98–0.99 respectively. This semi-automated method could be used to analyze response over the treatment course, as it is not based on intensity values.