Image segmentation and registration are said to be challenging when applied to dynamic contrast enhanced MRI sequences (DCE-MRI). The contrast agent causes rapid changes in intensity in the region of interest and elsewhere, which can lead to false positive predictions for segmentation tasks and confound the image registration similarity metric. While it is widely assumed that contrast changes increase the difficulty of these tasks, to our knowledge no work has quantified these effects. In this paper we examine the effect of training with different ratios of contrast enhanced (CE) data on two popular tasks: segmentation with nnU-Net and Mask R-CNN and registration using VoxelMorph and VTN. We experimented further by strategically using the available datasets through pretraining and fine tuning with different splits of data. We found that to create a generalisable model, pretraining with CE data and fine tuning with non-CE data gave the best result. This interesting find could be expanded to other deep learning based image processing tasks with DCE-MRI and provide significant improvements to the models’ performance.
At the core of many neuro-imaging studies, atlas-based brain parcellations are used for example to study normal brain evolution across the lifespan. These atlases rely on the assumption that the same anatomical features are present on all subjects to be studied and that these features are stable enough to allow meaningful comparisons between different brain surfaces and structures These methods, however, often fail when applied to fetal MRI data, due to the lack of consistent anatomical features present across gestation. This paper presents a novel surface-based fetal cortical parcellation framework which attempts to circumvent the lack of consistent anatomical features by proposing a brain parcellation scheme that is based solely on learned geometrical features. A mesh signature incorporating both extrinsic and intrinsic geometrical features is proposed and used in a clustering scheme to define a parcellation of the fetal brain. This parcellation is then learned using a Random Forest (RF) based learning approach and then further refined in an alpha-expansion graph-cut scheme. Based on the votes obtained by the RF inference procedure, a probability map is computed and used as a data term in the graph-cut procedure. The smoothness term is defined by learning a transition matrix based on the dihedral angles of the faces. Qualitative and quantitative results on a cohort of both healthy and high-risk fetuses are presented. Both visual and quantitative assessments show good results demonstrating a reliable method for fetal brain data and the possibility of obtaining a parcellation of the fetal cortical surfaces using only geometrical features.
KEYWORDS: Image segmentation, Tissues, Magnetic resonance imaging, 3D metrology, 3D image processing, 3D modeling, Liver, Kidney, 3D applications, Visualization
This paper presents a novel shape-guided multi-region variational region growing framework for extract- ing simultaneously thoracic and abdominal organs on 3D infants whole body MRI. Due to the inherent low quality of these data, classical segmentation methods tend to fail at the multi-segmentation task. To compensate for the low resolution and the lack of contrast and to enable the simultaneous segmentation of multiple organs, we introduce a segmentation framework on a graph of supervoxels that combines supervoxels intensity distribution weighted by gradient vector ow value and a shape prior per tissue. The intensity-based homogeneity criteria and the shape prior, encoded using Legendre moments, are added as energy terms in the functional to be optimized. The intensity-based energy is computed using both local (voxel value) and global (neighboring regions mean values, adjacent voxels values and distance to the neighboring regions) criteria. Inter-region con ict resolution is handled using a weighted Voronoi decomposition method, the weights being determined using tissues densities. The energy terms of the global energy equation are weighted using an information on growth direction and on gradient vector flow value. This allows us to either guide the segmentation toward the image natural edges if it is consistent with image and shape prior terms, or enforce the shape prior term otherwise. Results on 3D infants MRI data are presented and compared to a set of manual segmentations. Both visual comparison and quantitative measurements show good results.
Ultrasound images appearance is characterized by speckle, shadows, signal dropout and low contrast which make
them really difficult to process and leads to a very poor signal to noise ratio. Therefore, for main imaging
applications, a denoising step is necessary to apply successfully medical imaging algorithms on such images.
However, due to speckle statistics, denoising and enhancing edges on these images without inducing additional
blurring is a real challenging problem on which usual filters often fail. To deal with such problems, a large number
of papers are working on B-mode images considering that the noise is purely multiplicative. Making such an
assertion could be misleading, because of internal pre-processing such as log compression which are done in the
ultrasound device. To address those questions, we designed a novel filtering method based on 1D Radiofrequency
signal. Indeed, since B-mode images are initially composed of 1D signals and since the log compression made by ultrasound devices modifies noise statistics, we decided to filter directly the 1D Radiofrequency signal envelope before log compression and image reconstitution, in order to conserve as much information as possible. A bi-orthogonal wavelet transform is applied to the log transform of each signal and an adaptive 1D split and merge like algorithm is used to denoise wavelet coefficients. Experiments were carried out on synthetic data sets simulated with Field II simulator and results show that our filter outperforms classical speckle filtering methods like Lee, non-linear means or SRAD filters.
Segmentation of ultrasound kidney images represents a challenge due to low quality data. Speckle, shadows, signal
dropout and low contrast make segmentation a harsh task. In addition, kidney ultrasound imaging presents a great
variability concerning the organ's shape on the image. This characteristic makes learning methods hard to use. The aim
of this study is to develop a real time kidney ultrasound image segmentation method usable during surgical operations
such as punctures. To deal with real time constraints, we decided to focus on region based methods and particularly split
and merge algorithm. In this prospective study, the selection of the interesting area in the initial image is made by the
physician, drawing a coarse bounding box around the organ. A pre-processing phase is first performed to correct image's
artefacts. This phase is composed of three major steps. First, an image specification is made between the image to
segment and a reference one. Then, a Haar wavelet filtering method is applied on the resulting image and finally an
anisotropic diffusion filter is applied to smooth the result. Then, a split and merge algorithm is applied on the resulting
image. Both split and merge criteria are based on regions statistics. Our method has been successfully applied on a set of
22 clinical images coming from 10 different patients and presenting different points of view regarding kidney's shape.
We obtained very good results, for an average computational time of 8.5 seconds per image.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.