Search

Breast cancer, a high-incidence cancer among female, occupies a large incidence of total female patients with cancer. Pathological examination is the gold standard for breast cancer in clinic diagnosis. However, accuracy and efficient diagnosis is challengeable to pathologists for the complex of breast cancer and laborious work. Introducing computer-aid diagnosis (CAD) can relieve laborious work of pathologists and improve diagnosed accuracy for breast cancer. To promote development of CAD methods, we release a large-scale and hematoxylin-eosin (HE) staining dataset of breast cancer for medical image segmentation task, called the breast-cancer image segmentation 5000 (BIS5k). BIS5k contains 5929 images that are divided into training data (5000) and evaluated data (929). All images of BIS5k are collected from clinic cases which include patients with various age and cancer stages. All labels of images are annotated in pixel level for segmentation task and reviewed by pathological professors carefully. Furthermore, we construct a basic instance called breast-cancer segmentation network, BCSNet with a toolkit including comprehensive metrics to demonstrate the usage of BIS5k. Extensive experiments of BCSNet and compared methods provide that developing specific algorithm and constructing dataset are indispensable to promote CAD of pathological diagnosis for breast cancer. [ABSTRACT FROM AUTHOR]

INTRODUCTION: The rapid development of the space industry requires a deeper understanding of spaceflight's impact on the brain. MRI research reports brain volume changes following spaceflight in astronauts, potentially affecting cognition. Recently, we have demonstrated that this evidence of volumetric changes, as measured by typical T1-weighted sequences (e.g., magnetization-prepared rapid gradient echo sequence; MPRAGE), is error-prone due to the microgravity-related redistribution of cerebrospinal fluid in the brain. More modern neuroimaging methods, particularly dual-echo MPRAGE (DEMPRAGE) and magnetization-prepared rapid gradient echo sequence utilizing two inversion pulses (MP2RAGE), have been suggested to be resilient to this error. Here, we tested if these imaging modalities offered consistent segmentation performance improvements in some commonly employed neuroimaging software packages. METHODS: We conducted manual gray matter tissue segmentation in traditional T1w MRI images to utilize for comparison. Automated tissue segmentation was performed for traditional T1w imaging, as well as on DEMPRAGE and MP2RAGE images from the same subjects. Statistical analysis involved a comparison of total gray matter volumes for each modality, and the extent of tissue segmentation agreement was assessed using a test of similarity (Dice coefficient). RESULTS: Neither DEMPRAGE nor MP2RAGE exhibited consistent segmentation performance across all toolboxes tested. DISCUSSION: This research indicates that customized data collection and processing methods are necessary for reliable and valid structural MRI segmentation in astronauts, as current methods provide erroneous classification and hence inaccurate claims of neuroplastic brain changes in the astronaut population. [ABSTRACT FROM AUTHOR]

With the development of medical imaging technologies, breast cancer segmentation remains challenging, especially when considering multimodal imaging. Compared to a single-modality image, multimodal data provide additional information, contributing to better representation learning capabilities. This paper applies these advantages by presenting a deep learning network architecture for segmenting breast cancer with multimodal computed tomography (CT) images based on fusing U-Net architectures that can learn richer representations from multimodal data. The multipath fusion architecture introduces an additional fusion module across different paths, enabling the model to extract features from different modalities at each level of the encoding path. This approach enhances segmentation performance and produces more robust results compared to using a single modality. The study reports experiments conducted on multimodal CT images from 36 patients for training, validation, and testing purposes. The results demonstrate that the proposed model ouperforms the U-Net architecture when considering different combinations of input image modalities. Specifically, when combining two distinct CT modalities, the ZE and IoNW input combination yields the highest Dice score of 0.8546. [ABSTRACT FROM AUTHOR]

Multi-class segmentation of unlabelled living cells in time-lapse light microscopy images is challenging due to the temporal behaviour and changes in cell life cycles and the complexity of these images. The deep-learning-based methods achieved promising outcomes and remarkable success in single- and multi-class medical and microscopy image segmentation. The main objective of this study is to develop a hybrid deep-learning-based categorical segmentation and classification method for living HeLa cells in reflected light microscopy images. A symmetric simple U-Net and three asymmetric hybrid convolution neural networks—VGG19-U-Net, Inception-U-Net, and ResNet34-U-Net—were proposed and mutually compared to find the most suitable architecture for multi-class segmentation of our datasets. The inception module in the Inception-U-Net contained kernels with different sizes within the same layer to extract all feature descriptors. The series of residual blocks with the skip connections in each ResNet34-U-Net's level alleviated the gradient vanishing problem and improved the generalisation ability. The m-IoU scores of multi-class segmentation for our datasets reached 0.7062, 0.7178, 0.7907, and 0.8067 for the simple U-Net, VGG19-U-Net, Inception-U-Net, and ResNet34-U-Net, respectively. For each class and the mean value across all classes, the most accurate multi-class semantic segmentation was achieved using the ResNet34-U-Net architecture (evaluated as the m-IoU and Dice metrics). [ABSTRACT FROM AUTHOR]

Analysis of imaging mass cytometry (IMC) data and other low‐resolution multiplexed tissue imaging technologies is often confounded by poor single‐cell segmentation and suboptimal approaches for data visualization and exploration. This can lead to inaccurate identification of cell phenotypes, states, or spatial relationships compared to reference data from single‐cell suspension technologies. To this end we have developed the "OPTimized Imaging Mass cytometry AnaLysis (OPTIMAL)" framework to benchmark any approaches for cell segmentation, parameter transformation, batch effect correction, data visualization/clustering, and spatial neighborhood analysis. Using a panel of 27 metal‐tagged antibodies recognizing well‐characterized phenotypic and functional markers to stain the same Formalin‐Fixed Paraffin Embedded (FFPE) human tonsil sample tissue microarray over 12 temporally distinct batches we tested several cell segmentation models, a range of different arcsinh cofactor parameter transformation values, 5 different dimensionality reduction algorithms, and 2 clustering methods. Finally, we assessed the optimal approach for performing neighborhood analysis. We found that single‐cell segmentation was improved by the use of an Ilastik‐derived probability map but that issues with poor segmentation were only really evident after clustering and cell type/state identification and not always evident when using "classical" bivariate data display techniques. The optimal arcsinh cofactor for parameter transformation was 1 as it maximized the statistical separation between negative and positive signal distributions and a simple Z‐score normalization step after arcsinh transformation eliminated batch effects. Of the five different dimensionality reduction approaches tested, PacMap gave the best data structure with FLOWSOM clustering out‐performing phenograph in terms of cell type identification. We also found that neighborhood analysis was influenced by the method used for finding neighboring cells with a "disc" pixel expansion outperforming a "bounding box" approach combined with the need for filtering objects based on size and image‐edge location. Importantly, OPTIMAL can be used to assess and integrate with any existing approach to IMC data analysis and, as it creates.FCS files from the segmentation output and allows for single‐cell exploration to be conducted using a wide variety of accessible software and algorithms familiar to conventional flow cytometrists. [ABSTRACT FROM AUTHOR]

Purpose: There is no objective way to measure the amount of manipulation and retraction of neural tissue by the surgeon. Our goal is to develop metrics quantifying dynamic retraction and manipulation by instruments during neurosurgery. Methods: We trained a convolutional neural network (CNN) to analyze microscopic footage of neurosurgical procedures and thereby generate metrics evaluating the surgeon's dynamic retraction of brain tissue and, using an object tracking process, evaluate the surgeon's manipulation of the instruments themselves. U-Net image segmentation is used to output bounding polygons around cerebral parenchyma of interest, as well as the vascular structures and cranial nerves. A channel and spatial reliability tracker framework is used in conjunction with our CNN to track desired surgical instruments. Results: Our network achieved a state-of-the-art intersection over union ( 72.64 % ) for biological tissue segmentation. Multivariate statistical analysis was used to evaluate dynamic retraction, tissue handling, and instrument manipulation. Conclusion: Our model enables to evaluate dynamic retraction of soft tissue and manipulation of instruments during a surgical procedure, while accounting for movement of the operative microscope. This model can potentially provide the surgeon with objective feedback about the movement of instruments and its effect on brain tissue. [ABSTRACT FROM AUTHOR]

Background and purpose: In patients with epilepsies of structural origin, brain atrophy and pathological alterations of the tissue microstructure extending beyond the putative epileptogenic lesion have been reported. However, in patients without any evidence of epileptogenic lesions on diagnostic magnetic resonance imaging (MRI), impairment of the brain microstructure has been scarcely elucidated. Using multiparametric quantitative (q) magnetic resonance imaging MRI, we aimed to investigate diffuse impairment of the microstructural tissue integrity in MRI-negative focal epilepsy patients. Methods: 27 MRI-negative patients with focal epilepsy (mean age 33.1 ± 14.2 years) and 27 matched healthy control subjects underwent multiparametric qMRI including T1, T2, and PD mapping at 3 T. After tissue segmentation based on synthetic anatomies, mean qMRI parameter values were extracted from the cerebral cortex, the white matter (WM) and the deep gray matter (GM) and compared between patients and control subjects. Apart from calculating mean values for the qMRI parameters across the respective compartments, voxel-wise analyses were performed for each tissue class. Results: There were no significant differences for mean values of quantitative T1, T2, and PD obtained from the cortex, the WM and the deep GM between the groups. Furthermore, the voxel-wise analyses did not reveal any clusters indicating significant differences between patients and control subjects for the qMRI parameters in the respective compartments. Conclusions: Based on the employed methodology, no indication for an impairment of the cerebralmicrostructural tissue integrity inMRI-negative patients with focal epilepsy was found in this study. Further research will be necessary to identify relevant factors and mechanisms contributing to microstructural brain tissue damage in various subgroups of patients with epilepsy. [ABSTRACT FROM AUTHOR]

The prognostic ability of global white matter and gray matter metabolite ratios following pediatric traumatic brain injury (TBI) and their relationship to 12-month neuropsychological assessments of intelligence quotient (IQ), attention, and memory is presented. Three-dimensional proton magnetic resonance spectroscopic imaging (MRSI) in pediatric subjects with complicated mild (cMild), moderate, and severe TBI was acquired acutely (6–18 days) and 12 months post-injury and compared to age-matched typically developing adolescents. A global linear regression model, co-registering MRSI metabolite maps with 3D high-resolution magnetic resonance images, was used to identify longitudinal white matter and gray matter metabolite ratio changes. Acutely, gray matter NAA/Cr, white matter NAA/Cr, and white matter NAA/Cho ratios were significantly lower in TBI groups compared to controls. Gray matter NAA/Cho was reduced only in the severe TBI group. At 12 months, all metabolite ratios normalized to control levels in each of the TBI groups. Acute gray matter and white matter NAA ratios were significantly correlated to 12-month assessments of IQ, attention, and memory. These findings suggest that whole brain gray matter and white matter metabolite ratios reflect longitudinal changes in neuronal metabolism following TBI, which can be used to predict neuropsychological outcomes in pediatric subjects. [ABSTRACT FROM AUTHOR]

Brain is an amazing organ that controls all activities of a human. Any abnormality in the shape of anatomical regions of the brain needs to be detected as early as possible to reduce the mortality rate. It is also beneficial for treatment planning and therapy. The most crucial task is to isolate abnormal areas from normal tissue regions. To identify abnormalities in the earlier stage, various medical imaging modalities were used by medical practitioners as part of the diagnosis. Magnetic Resonance Imaging (MRI) is a non-invasive diagnostic tool used for analyzing the internal structures owing to its capability to provide images with high resolution and better contrast for soft tissues. This survey focuses on studies done in brain MRI. Manual segmentation of abnormal tissues is a time-consuming task, and the performance depends on the expert's efficiency. Hence automating tumor segmentation plays a vital role in medical imaging applications. This study aims to provide a comprehensive survey on recent works developed in brain tumor segmentation. In this paper, a systematic literature review is presented to the reader to understand three policies, namely classical scheme, machine learning strategy, and deep learning methodology meant for tumor segmentation. Our primary goal is to include classical methods like atlas-based strategy and statistical-based models employed for segmenting tumors from brain MRI. Few studies that utilized machine learning approaches for the segmentation and classification of brain structures are also discussed. After that, the study provides an overview of deep learning-based segmentation models for quantitative analysis of brain MRI. Deep learning plays a vital role in the automatic segmentation of brain tissues. Presently deep learning technique outshines traditional statistical methods and machine learning approaches. An effort is made to enclose the literature on patch-based and semantic-based tissue segmentation presented by researchers working in the discipline of medical imaging. The manuscript discusses the basic convolutional neural network architecture, Data Sets, and the existing deep learning techniques for tissue segmentation coupled with classification. This paper also attempts to summarize the current works in Convolutional Neural networks and Autoencoders that assist researchers in seeking future directions. Finally, this article is concluded with possible developments and open challenges in brain tumor segmentation. [ABSTRACT FROM AUTHOR]

Summary: Background: Tissues such as the liver lobule, kidney nephron, and intestinal gland exhibit intricate patterns of zonated gene expression corresponding to distinct cell types and functions. To quantitatively understand zonation, it is important to measure cellular or genetic features as a function of position along a zonal axis. While it is possible to manually count, characterize, and locate features in relation to the zonal axis, it is labor-intensive and difficult to do manually while maintaining precision and accuracy. Methods: We addressed this challenge by developing a deep-learning-based quantification method called the “Tissue Positioning System” (TPS), which can automatically analyze zonation in the liver lobule as a model system. Findings: By using algorithms that identified vessels, classified vessels, and segmented zones based on the relative position along the portal vein to central vein axis, TPS was able to spatially quantify gene expression in mice with zone specific reporters. Interpretation: TPS could discern expression differences between zonal reporter strains, ages, and disease states. TPS could also reveal the zonal distribution of cells previously thought to be positioned randomly. The design principles of TPS could be generalized to other tissues to explore the biology of zonation. Funding: CPRIT (RP190208, RP220614, RP230330) and NIH (P30CA142543, R01AA028791, R01CA251928, R01DK1253961, R01GM140012, 1R01GM141519, 1R01DE030656, 1U01CA249245). The Pollack Foundation, Simmons Comprehensive Cancer Center Cancer & Obesity Translational Pilot Award, and the Emerging Leader Award from the Mark Foundation For Cancer Research (#21-003-ELA).

Background and purposeIn patients with epilepsies of structural origin, brain atrophy and pathological alterations of the tissue microstructure extending beyond the putative epileptogenic lesion have been reported. However, in patients without any evidence of epileptogenic lesions on diagnostic magnetic resonance imaging (MRI), impairment of the brain microstructure has been scarcely elucidated. Using multiparametric quantitative (q) magnetic resonance imaging MRI, we aimed to investigate diffuse impairment of the microstructural tissue integrity in MRI-negative focal epilepsy patients.Methods27 MRI-negative patients with focal epilepsy (mean age 33.1 ± 14.2 years) and 27 matched healthy control subjects underwent multiparametric qMRI including T1, T2, and PD mapping at 3 T. After tissue segmentation based on synthetic anatomies, mean qMRI parameter values were extracted from the cerebral cortex, the white matter (WM) and the deep gray matter (GM) and compared between patients and control subjects. Apart from calculating mean values for the qMRI parameters across the respective compartments, voxel-wise analyses were performed for each tissue class.ResultsThere were no significant differences for mean values of quantitative T1, T2, and PD obtained from the cortex, the WM and the deep GM between the groups. Furthermore, the voxel-wise analyses did not reveal any clusters indicating significant differences between patients and control subjects for the qMRI parameters in the respective compartments.ConclusionsBased on the employed methodology, no indication for an impairment of the cerebral microstructural tissue integrity in MRI-negative patients with focal epilepsy was found in this study. Further research will be necessary to identify relevant factors and mechanisms contributing to microstructural brain tissue damage in various subgroups of patients with epilepsy.

Multi-class segmentation of unlabelled living cells in time-lapse light microscopy images is challenging due to the temporal behaviour and changes in cell life cycles and the complexity of these images. The deep-learning-based methods achieved promising outcomes and remarkable success in single- and multi-class medical and microscopy image segmentation. The main objective of this study is to develop a hybrid deep-learning-based categorical segmentation and classification method for living HeLa cells in reflected light microscopy images. A symmetric simple U-Net and three asymmetric hybrid convolution neural networks—VGG19-U-Net, Inception-U-Net, and ResNet34-U-Net—were proposed and mutually compared to find the most suitable architecture for multi-class segmentation of our datasets. The inception module in the Inception-U-Net contained kernels with different sizes within the same layer to extract all feature descriptors. The series of residual blocks with the skip connections in each ResNet34-U-Net’s level alleviated the gradient vanishing problem and improved the generalisation ability. The m-IoU scores of multi-class segmentation for our datasets reached 0.7062, 0.7178, 0.7907, and 0.8067 for the simple U-Net, VGG19-U-Net, Inception-U-Net, and ResNet34-U-Net, respectively. For each class and the mean value across all classes, the most accurate multi-class semantic segmentation was achieved using the ResNet34-U-Net architecture (evaluated as the m-IoU and Dice metrics).

The early diagnosis of diabetes mellitus in normal people or maintaining stable blood sugar concentrations in diabetic patients requires frequent monitoring of the blood sugar levels. However, regular monitoring of the sugar levels is problematic owing to the pain and inconvenience associated with pricking the fingertip or using minimally invasive patches. In this study, we devise a noninvasive method to estimate the percentage of the in vivo glycated hemoglobin (HbA1c) values from Monte Carlo photon propagation simulations, based on models of the wrist using 3D magnetic resonance (MR) image data. The MR image slices are first segmented for several different tissue types, and the proposed Monte Carlo photon propagation system with complex composite tissue support is then used to derive several models for the fingertip and wrist sections with different wavelengths of light sources and photodetector arrangements. The Pearson r values for the estimated percent HbA1c values are 0.94 and 0.96 for the fingertip transmission- and reflection-type measurements, respectively. This is found to be the best among the related studies. Furthermore, a single-detector multiple-source arrangement resulted in a Pearson r value of 0.97 for the wrist. The Bland–Altman bias values were found to be −0.003 ± 0.36, 0.01 ± 0.25, and 0.01 ± 0.21, for the two fingertip and wrist models, respectively, which conform to the standards of the current state-of-the-art invasive point-of-care devices. The implementation of these algorithms will be a suitable alternative to the invasive state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Purpose: To provide a flexible, end-to-end platform for visually distinguishing diseased from undiseased tissue in a medical image, in particular pathology slides, and classifying diseased regions by subtype. Highly accurate results are obtained using small training datasets and reduced-scale source images that can be easily shared. Approach: An ensemble of lightweight convolutional neural networks (CNNs) is trained on different subsets of images derived from a relatively small number of annotated whole-slide histopathology images (WSIs). The WSIs are first reduced in scale in a manner that preserves anatomic features critical to analysis while also facilitating convenient handling and storage. The segmentation and subtyping tasks are performed sequentially on the reduced-scale images using the same basic workflow: generating and sifting tiles from the image, then classifying each tile with an ensemble of appropriately trained CNNs. For segmentation, the CNN predictions are combined using a function to favor a selected similarity metric, and a mask or map for a a candidate image is produced from tiles whose combined predictions exceed a decision boundary. For subtyping, the resulting mask is applied to the candidate image, and new tiles are derived from the unoccluded regions. These are classified by the subtyping CNNs to produce an overall subtype prediction. Results and conclusion: This approach was applied successfully to two very different datasets of large WSIs, one (PAIP2020) involving multiple subtypes of colorectal cancer and the other (CAMELYON16) single-type breast cancer metastases. Scored using standard similarity metrics, the segmentations outperformed more complex models typifying the state of the art. [ABSTRACT FROM AUTHOR]

Fatty infiltration in pancreas leading to steatosis is a major risk factor in pancreas transplantation. Hematoxylin and eosin (H and E) is one of the common histological staining techniques that provides information on the tissue cytoarchitecture. Adipose (fat) cells accumulation in pancreas has been shown to impact beta cell survival, its endocrine function and pancreatic steatosis and can cause non-alcoholic fatty pancreas disease (NAFPD). The current automated tools (E.g. Adiposoft) available for fat analysis are suited for white fat tissue which is homogeneous and easier to segment unlike heterogeneous tissues such as pancreas where fat cells continue to play critical physiopathological functions. The currently, available pancreas segmentation tool focuses on endocrine islet segmentation based on cell nuclei detection for diagnosis of pancreatic cancer. In the current study, we present a fat quantifying tool, Fatquant, which identifies fat cells in heterogeneous H and E tissue sections with reference to diameter of fat cell. Using histological images from a public database, we observed an intersection over union of 0.797 to 0.962 and 0.675 to 0.937 for manual versus Fatquant analysis of pancreas and liver, respectively.

Purpose: To provide a flexible, end-to-end platform for visually distinguishing diseased from undiseased tissue in a medical image, in particular pathology slides, and classifying diseased regions by subtype. Highly accurate results are obtained using small training datasets and reduced-scale source images that can be easily shared. Approach: An ensemble of lightweight convolutional neural networks (CNNs) is trained on different subsets of images derived from a relatively small number of annotated whole-slide histopathology images (WSIs). The WSIs are first reduced in scale in a manner that preserves anatomic features critical to analysis while also facilitating convenient handling and storage. The segmentation and subtyping tasks are performed sequentially on the reduced-scale images using the same basic workflow: generating and sifting tiles from the image, then classifying each tile with an ensemble of appropriately trained CNNs. For segmentation, the CNN predictions are combined using a function to favor a selected similarity metric, and a mask or map for a a candidate image is produced from tiles whose combined predictions exceed a decision boundary. For subtyping, the resulting mask is applied to the candidate image, and new tiles are derived from the unoccluded regions. These are classified by the subtyping CNNs to produce an overall subtype prediction. Results and conclusion: This approach was applied successfully to two very different datasets of large WSIs, one (PAIP2020) involving multiple subtypes of colorectal cancer and the other (CAMELYON16) single-type breast cancer metastases. Scored using standard similarity metrics, the segmentations outperformed more complex models typifying the state of the art.

Tissue segmentation is the mainstay of pathological examination, whereas the manual delineation is unduly burdensome. To assist this time-consuming and subjective manual step, researchers have devised methods to automatically segment structures in pathological images. Recently, automated machine and deep learning based methods dominate tissue segmentation research studies. However, most machine and deep learning based approaches are supervised and developed using a large number of training samples, in which the pixel-wise annotations are expensive and sometimes can be impossible to obtain. This paper introduces a novel unsupervised learning paradigm by integrating an end-to-end deep mixture model with a constrained indicator to acquire accurate semantic tissue segmentation. This constraint aims to centralise the components of deep mixture models during the calculation of the optimisation function. In so doing, the redundant or empty class issues, which are common in current unsupervised learning methods, can be greatly reduced. By validation on both public and in-house datasets, the proposed deep constrained Gaussian network achieves significantly (Wilcoxon signed-rank test) better performance (with the average Dice scores of 0.737 and 0.735, respectively) on tissue segmentation with improved stability and robustness, compared to other existing unsupervised segmentation approaches. Furthermore, the proposed method presents a similar performance (p-value >0.05) compared to the fully supervised U-Net. [ABSTRACT FROM AUTHOR]

Tissue segmentation is an essential task in computational pathology. However, relevant datasets for such a pixel-level classification task are hard to obtain due to the difficulty of annotation, bringing obstacles for training a deep learning-based segmentation model. Recently, contrastive learning has provided a feasible solution for mitigating the heavy reliance of deep learning models on annotation. Nevertheless, applying contrastive loss to the most abstract image representations, existing contrastive learning frameworks focus on global features, therefore, are less capable of encoding finer-grained features (e.g., pixel-level discrimination) for the tissue segmentation task. Enlightened by domain knowledge, we design three contrastive learning tasks with multi-granularity views (from global to local) for encoding necessary features into representations without accessing annotations. Specifically, we construct: (1) an image-level task to capture the difference between tissue components, i.e., encoding the component discrimination; (2) a superpixel-level task to learn discriminative representations of local regions with different tissue components, i.e., encoding the prototype discrimination; (3) a pixel-level task to encourage similar representations of different tissue components within a local region, i.e., encoding the spatial smoothness. Through our global-to-local pre-training strategy, the learned representations can reasonably capture the domain-specific and fine-grained patterns, making them easily transferable to various tissue segmentation tasks in histopathological images. We conduct extensive experiments on two tissue segmentation datasets, while considering two real-world scenarios with limited or sparse annotations. The experimental results demonstrate that our framework is superior to existing contrastive learning methods and can be easily combined with weakly supervised and semi-supervised segmentation methods. [ABSTRACT FROM AUTHOR]

Accurate and reproducible tissue identification is essential for understanding structural and functional changes that may occur naturally with aging, or because of a chronic disease, or in response to intervention therapies. Peripheral quantitative computed tomography (pQCT) is regularly employed for body composition studies, especially for the structural and material properties of the bone. Furthermore, pQCT acquisition requires low radiation dose and the scanner is compact and portable. However, pQCT scans have limited spatial resolution and moderate SNR. pQCT image quality is frequently degraded by involuntary subject movement during image acquisition. These limitations may often compromise the accuracy of tissue quantification, and emphasize the need for automated and robust quantification methods. We propose a tissue identification and quantification methodology that addresses image quality limitations and artifacts, with increased interest in subject movement. We introduce a multi-atlas image segmentation (MAIS) framework for semantic segmentation of hard and soft tissues in pQCT scans at multiple levels of the lower leg. We describe the stages of statistical atlas generation, deformable registration and multi-tissue classifier fusion. We evaluated the performance of our methodology using multiple deformable registration approaches against reference tissue masks. We also evaluated the performance of conventional model-based segmentation against the same reference data to facilitate comparisons. We studied the effect of subject movement on tissue segmentation quality. We also applied the top performing method to a larger out-of-sample dataset and report the quantification results. The results show that multi-atlas image segmentation with diffeomorphic deformation and probabilistic label fusion produces very good quality over all tissues, even for scans with significant quality degradation. The application of our technique to the larger dataset reveals trends of agerelated body composition changes that are consistent with the literature. Because of its robustness to subject motion artifacts, our MAIS methodology enables analysis of larger number of scans than conventional state-of-the-art methods. Automated analysis of both soft and hard tissues in pQCT is another contribution of this work. [ABSTRACT FROM AUTHOR]

Introduction: The aim of this study is to applied CT scan-based tissue segmentation to measure visceral adipose tissue (VAT) and subcutaneous adipose tissue (SAT) volumes. Then, the ratio of both fat tissue volumes was calculated to get two more parameters: visceral fat volume ratio (VFVR) and subcutaneous fat volume ratio (SFVR). After that, the relationship between these factors and urolithiasis recurrence was examined using correlation analysis. Other parameters, which are the patient’s age and gender, were also tested for correlation analysis with urolithiasis recurrence. Finally, logistic regression analysis was performed to find the association between urolithiasis recurrence and the parameters (age, gender, VAT volume, SAT volume, VFVR and SFVR). Methods: This study was a retrospective cross-sectional study design using the images collected from CT Urology cases in the year 2019. The patients selected have a history of stone removal in 2014. The application used for CT tissue segmentation is 3D Slicer. Results: Urolithiasis recurrence shows medium and high degree of positive correlation with total fat volume, VAT volume, and VFVR (correlation coefficient, cc = 0.254, p = 0.023), (cc = 0.390, p< 0.001) and (cc = 0.688, p< 0.001), respectively and high degree of negative correlation with SFVR (cc = -0.688, p< 0.001). However, using logistic regression analysis, only VAT volume was significantly associated with urolithiasis recurrence (OR 1.11, 95% CI 1.01-1.22, p= 0.03), while the total fat volume, VFVR, and SFVR are not significant. Conclusion: CT scan-based tissue segmentation has a huge impact on fat volume quantification. This study confirms that VAT volume was strongly correlated with urolithiasis recurrence, indicating that VAT volume plays a more important role than SAT volume, total fat volume, VFVR and SFVR in the production of urinary stone. Thus, VAT volume can be further considered as a new independent risk factor for urolithiasis. [ABSTRACT FROM AUTHOR]

Accurate and reproducible tissue identification is essential for understanding structural and functional changes that may occur naturally with aging, or because of a chronic disease, or in response to intervention therapies. Peripheral quantitative computed tomography (pQCT) is regularly employed for body composition studies, especially for the structural and material properties of the bone. Furthermore, pQCT acquisition requires low radiation dose and the scanner is compact and portable. However, pQCT scans have limited spatial resolution and moderate SNR. pQCT image quality is frequently degraded by involuntary subject movement during image acquisition. These limitations may often compromise the accuracy of tissue quantification, and emphasize the need for automated and robust quantification methods. We propose a tissue identification and quantification methodology that addresses image quality limitations and artifacts, with increased interest in subject movement. We introduce a multi-atlas image segmentation (MAIS) framework for semantic segmentation of hard and soft tissues in pQCT scans at multiple levels of the lower leg. We describe the stages of statistical atlas generation, deformable registration and multi-tissue classifier fusion. We evaluated the performance of our methodology using multiple deformable registration approaches against reference tissue masks. We also evaluated the performance of conventional model-based segmentation against the same reference data to facilitate comparisons. We studied the effect of subject movement on tissue segmentation quality. We also applied the top performing method to a larger out-of-sample dataset and report the quantification results. The results show that multi-atlas image segmentation with diffeomorphic deformation and probabilistic label fusion produces very good quality over all tissues, even for scans with significant quality degradation. The application of our technique to the larger dataset reveals trends of age-related body composition changes that are consistent with the literature. Because of its robustness to subject motion artifacts, our MAIS methodology enables analysis of larger number of scans than conventional state-of-the-art methods. Automated analysis of both soft and hard tissues in pQCT is another contribution of this work.

Changes in levels of the inhibitory neurotransmitter γ‐aminobutyric acid (GABA) may underlie aging‐related changes in brain function. GABA and co‐edited macromolecules (GABA+) can be measured with MEGA‐PRESS magnetic resonance spectroscopy (MRS). The current study investigated how changes in the aging brain impact the interpretation of GABA+ measures in bilateral auditory cortices of healthy young and older adults. Structural changes during aging appeared as decreasing proportion of grey matter in the MRS volume of interest and corresponding increase in cerebrospinal fluid. GABA+ referenced to H2O without tissue correction declined in aging. This decline persisted after correcting for tissue differences in MR‐visible H2O and relaxation times but vanished after considering the different abundance of GABA+ in grey and white matter. However, GABA+ referenced to creatine and N‐acetyl aspartate (NAA), which showed no dependence on tissue composition, decreased in aging. All GABA+ measures showed hemispheric asymmetry in young but not older adults. The study also considered aging‐related effects on tissue segmentation and the impact of co‐edited macromolecules. Tissue segmentation differed significantly between commonly used algorithms, but aging‐related effects on tissue‐corrected GABA+ were consistent across methods. Auditory cortex macromolecule concentration did not change with age, indicating that a decline in GABA caused the decrease in the compound GABA+ measure. Most likely, the macromolecule contribution to GABA+ leads to underestimating an aging‐related decrease in GABA. Overall, considering multiple GABA+ measures using different reference signals strengthened the support for an aging‐related decline in auditory cortex GABA levels. [ABSTRACT FROM AUTHOR]

In this paper, we propose a novel squeeze M-SegNet (SM-SegNet) architecture featuring a fire module to perform accurate as well as fast segmentation of the brain on magnetic resonance imaging (MRI) scans. The proposed model utilizes uniform input patches, combined-connections, long skip connections, and squeeze–expand convolutional layers from the fire module to segment brain MRI data. The proposed SM-SegNet architecture involves a multi-scale deep network on the encoder side and deep supervision on the decoder side, which uses combined-connections (skip connections and pooling indices) from the encoder to the decoder layer. The multi-scale side input layers support the deep network layers' extraction of discriminative feature information, and the decoder side provides deep supervision to reduce the gradient problem. By using combined-connections, extracted features can be transferred from the encoder to the decoder resulting in recovering spatial information, which makes the model converge faster. Long skip connections were used to stabilize the gradient updates in the network. Owing to the adoption of the fire module, the proposed model was significantly faster to train and offered a more efficient memory usage with 83% fewer parameters than previously developed methods, owing to the adoption of the fire module. The proposed method was evaluated using the open-access series of imaging studies (OASIS) and the internet brain segmentation registry (IBSR) datasets. The experimental results demonstrate that the proposed SM-SegNet architecture achieves segmentation accuracies of 95% for cerebrospinal fluid, 95% for gray matter, and 96% for white matter, which outperforms the existing methods in both subjective and objective metrics in brain MRI segmentation. [ABSTRACT FROM AUTHOR]

The prognostic ability of global white matter and gray matter metabolite ratios following pediatric traumatic brain injury (TBI) and their relationship to 12-month neuropsychological assessments of intelligence quotient (IQ), attention, and memory is presented. Three-dimensional proton magnetic resonance spectroscopic imaging (MRSI) in pediatric subjects with complicated mild (cMild), moderate, and severe TBI was acquired acutely (6–18 days) and 12 months post-injury and compared to age-matched typically developing adolescents. A global linear regression model, co-registering MRSI metabolite maps with 3D high-resolution magnetic resonance images, was used to identify longitudinal white matter and gray matter metabolite ratio changes. Acutely, gray matter NAA/Cr, white matter NAA/Cr, and white matter NAA/Cho ratios were significantly lower in TBI groups compared to controls. Gray matter NAA/Cho was reduced only in the severe TBI group. At 12 months, all metabolite ratios normalized to control levels in each of the TBI groups. Acute gray matter and white matter NAA ratios were significantly correlated to 12-month assessments of IQ, attention, and memory. These findings suggest that whole brain gray matter and white matter metabolite ratios reflect longitudinal changes in neuronal metabolism following TBI, which can be used to predict neuropsychological outcomes in pediatric subjects.

Transcranial magnetic stimulation (TMS) is used to treat a range of brain disorders by inducing an electric field (E-field) in the brain. However, the precise neural effects of TMS are not well understood. Nonhuman primates (NHPs) are used to model the impact of TMS on neural activity, but a systematic method of quantifying the induced E-field in the cortex of NHPs has not been developed. The pipeline uses statistical parametric mapping (SPM) to automatically segment a structural MRI image of a rhesus macaque into five tissue compartments. Manual corrections are necessary around implants. The segmented tissues are tessellated into 3D meshes used in finite element method (FEM) software to compute the TMS induced E-field in the brain. The gray matter can be further segmented into cortical laminae using a volume preserving method for defining layers. Models of three NHPs were generated with TMS coils placed over the precentral gyrus. Two coil configurations, active and sham, were simulated and compared. The results demonstrated a large difference in E-fields at the target. Additionally, the simulations were calculated using two different E-field solvers and were found to not significantly differ. Current methods segment NHP tissues manually or use automated methods for only the brain tissue. Existing methods also do not stratify the gray matter into layers. The pipeline calculates the induced E-field in NHP models by TMS and can be used to plan implant surgeries and determine approximate E-field values around neuron recording sites. • Whole-head segmentation of NHPs is possible using statistical parametric mapping. • The gray matter segmentation can be further stratified into cortical layers. • E-fields induced by TMS can be simulated using free and commercial FEM packages. [ABSTRACT FROM AUTHOR]

Segmentation of thigh tissues (muscle, fat, inter-muscular adipose tissue (IMAT), bone, and bone marrow) from magnetic resonance imaging (MRI) scans is useful for clinical and research investigations in various conditions such as aging, diabetes mellitus, obesity, metabolic syndrome, and their associated comorbidities. Towards a fully automated, robust, and precise quantification of thigh tissues, herein we designed a novel semi-supervised segmentation algorithm based on deep network architectures. Built upon Tiramisu segmentation engine, our proposed deep networks use variational and specially designed targeted dropouts for faster and robust convergence, and utilize multi-contrast MRI scans as input data. In our experiments, we have used 150 scans from 50 distinct subjects from the Baltimore Longitudinal Study of Aging (BLSA). The proposed system made use of both labeled and unlabeled data with high efficacy for training, and outperformed the current state-of-the-art methods. In particular, dice scores of 97.52%, 94.61%, 80.14%, 95.93%, and 96.83% are achieved for muscle, fat, IMAT, bone, and bone marrow segmentation, respectively. Our results indicate that the proposed system can be useful for clinical research studies where volumetric and distributional tissue quantification is pivotal and labeling is a significant issue. To the best of our knowledge, the proposed system is the first attempt at multi-tissue segmentation using a single end-to-end semi-supervised deep learning framework for multi-contrast thigh MRI scans. [ABSTRACT FROM AUTHOR]

Confocal imaging is a well-established method for investigating plant phenotypes on the tissue and organ level. However, many differences are difficult to assess by visual inspection and researchers rely extensively on ad hoc manual quantification techniques and qualitative assessment. Here we present a method for quantitatively phenotyping large samples of plant tissue morphologies using triangulated isosurfaces. We successfully demonstrate the applicability of the approach using confocal imaging of aerial organs in Arabidopsis thaliana. Automatic identification of flower primordia using the surface curvature as an indication of outgrowth allows for high-throughput quantification of divergence angles and further analysis of individual flowers. We demonstrate the throughput of our method by quantifying geometric features of 1065 flower primordia from 172 plants, comparing auxin transport mutants to wild type. Additionally, we find that a paraboloid provides a simple geometric parameterisation of the shoot inflorescence domain with few parameters. We utilise parameterisation methods to provide a computational comparison of the shoot apex defined by a fluorescent reporter of the central zone marker gene CLAVATA3 with the apex defined by the paraboloid. Finally, we analyse the impact of mutations which alter mechanical properties on inflorescence dome curvature and compare the results with auxin transport mutants. Our results suggest that region-specific expression domains of genes regulating cell wall biosynthesis and local auxin transport can be important in maintaining the wildtype tissue shape. Altogether, our results indicate a general approach to parameterise and quantify plant development in 3D, which is applicable also in cases where data resolution is limited, and cell segmentation not possible. This enables researchers to address fundamental questions of plant development by quantitative phenotyping with high throughput, consistency and reproducibility. [ABSTRACT FROM AUTHOR]

Confocal imaging is a well-established method for investigating plant phenotypes on the tissue and organ level. However, many differences are difficult to assess by visual inspection and researchers rely extensively on ad hoc manual quantification techniques and qualitative assessment. Here we present a method for quantitatively phenotyping large samples of plant tissue morphologies using triangulated isosurfaces. We successfully demonstrate the applicability of the approach using confocal imaging of aerial organs in Arabidopsis thaliana. Automatic identification of flower primordia using the surface curvature as an indication of outgrowth allows for high-throughput quantification of divergence angles and further analysis of individual flowers. We demonstrate the throughput of our method by quantifying geometric features of 1065 flower primordia from 172 plants, comparing auxin transport mutants to wild type. Additionally, we find that a paraboloid provides a simple geometric parameterisation of the shoot inflorescence domain with few parameters. We utilise parameterisation methods to provide a computational comparison of the shoot apex defined by a fluorescent reporter of the central zone marker gene CLAVATA3 with the apex defined by the paraboloid. Finally, we analyse the impact of mutations which alter mechanical properties on inflorescence dome curvature and compare the results with auxin transport mutants. Our results suggest that region-specific expression domains of genes regulating cell wall biosynthesis and local auxin transport can be important in maintaining the wildtype tissue shape. Altogether, our results indicate a general approach to parameterise and quantify plant development in 3D, which is applicable also in cases where data resolution is limited, and cell segmentation not possible. This enables researchers to address fundamental questions of plant development by quantitative phenotyping with high throughput, consistency and reproducibility.

Objective: Brain atrophy has the potential to become a biomarker for severity of radiation-induced side-effects. Particularly brain tumour patients can show great MRI signal changes over time caused by e.g. oedema, tumour progress or necrosis. The goal of this study was to investigate if such changes affect the segmentation accuracy of normal appearing brain and thus influence longitudinal volumetric measurements. Materials and methods: T1-weighted MR images of 52 glioblastoma patients with unilateral tumours acquired before and three months after the end of radio(chemo)therapy were analysed. GM and WM volumes in the contralateral hemisphere were compared between segmenting the whole brain (full) and the contralateral hemisphere only (cl) with SPM and FSL. Relative GM and WM volumes were compared using paired t tests and correlated with the corresponding mean dose in GM and WM, respectively. Results: Mean GM atrophy was significantly higher for full segmentation compared to cl segmentation when using SPM (mean ± std: ΔVGM,full = − 3.1% ± 3.7%, ΔVGM,cl = − 1.6% ± 2.7%; p < 0.001, d = 0.62). GM atrophy was significantly correlated with the mean GM dose with the SPM cl segmentation (r = − 0.4, p = 0.004), FSL full segmentation (r = − 0.4, p = 0.004) and FSL cl segmentation (r = -0.35, p = 0.012) but not with the SPM full segmentation (r = − 0.23, p = 0.1). Conclusions: For accurate normal tissue volume measurements in brain tumour patients using SPM, abnormal tissue needs to be masked prior to segmentation, however, this is not necessary when using FSL. [ABSTRACT FROM AUTHOR]

Accurate segmentation of brain tissues, such as gray matter (GM), white matter (WM), and cerebrospinal fluid (CSF), in magnetic resonance imaging (MRI) images, is helpful for the diagnosis of neurological disorders, such as schizophrenia, Alzheimer's disease, and dementia. Studies on MRI-based brain segmentation have received significant attention in recent years based on the non-invasive imaging and good soft-tissue contrast provided by MRI. A number of studies have used conventional machine learning strategies, as well as convolutional neural network approaches. In this paper, we propose a patch-wise M-net architecture for the automatic segmentation of brain MRI images. In the proposed brain segmentation method, slices from a brain MRI scan are divided into non-overlapping patches, which are then fed into an M-net model with corresponding ground-truth patches to train the network, which is composed of two encoder-decoder processes. Dilated convolutional kernels with different sizes are used in the encoder and decoder modules to derive abundant semantic features from brain MRI scans. The proposed patch-wise M-net overcomes the drawbacks of conventional methods and provides greater retention of fine details. The proposed M-net model was trained and tested on the open-access series of imaging studies dataset. The performance was measured quantitatively using the Dice similarity coefficient. Experimental results demonstrate that the proposed method achieves average segmentation accuracies of 94.81% for CSF, 95.44% for GM, and 96.33% for WM, meaning it outperforms state-of-the-art methods.

The early diagnosis of diabetes mellitus in normal people or maintaining stable blood sugar concentrations in diabetic patients requires frequent monitoring of the blood sugar levels. However, regular monitoring of the sugar levels is problematic owing to the pain and inconvenience associated with pricking the fingertip or using minimally invasive patches. In this study, we devise a noninvasive method to estimate the percentage of the in vivo glycated hemoglobin (HbA1c) values from Monte Carlo photon propagation simulations, based on models of the wrist using 3D magnetic resonance (MR) image data. The MR image slices are first segmented for several different tissue types, and the proposed Monte Carlo photon propagation system with complex composite tissue support is then used to derive several models for the fingertip and wrist sections with different wavelengths of light sources and photodetector arrangements. The Pearson r values for the estimated percent HbA1c values are 0.94 and 0.96 for the fingertip transmission- and reflection-type measurements, respectively. This is found to be the best among the related studies. Furthermore, a single-detector multiple-source arrangement resulted in a Pearson r value of 0.97 for the wrist. The Bland–Altman bias values were found to be −0.003 ± 0.36, 0.01 ± 0.25, and 0.01 ± 0.21, for the two fingertip and wrist models, respectively, which conform to the standards of the current state-of-the-art invasive point-of-care devices. The implementation of these algorithms will be a suitable alternative to the invasive state-of-the-art methods.

Within developing tissues, cell proliferation, cell motility and other cell behaviors vary spatially, and this variability gives a complexity to the morphogenesis. Recently, novel formalisms have been developed to quantify tissue deformation and underlying cellular processes. A major challenge for the study of morphogenesis now is to objectively define tissue sub-regions exhibiting different dynamics. Here, we propose a method to automatically divide a tissue into regions where the local deformation rate is homogeneous. This was achieved by several steps including image segmentation, clustering and region boundary smoothing. We illustrate the use of the pipeline using a large dataset obtained during the metamorphosis of the Drosophila pupal notum. We also adapt it to determine regions in which the time evolution of the local deformation rate is homogeneous. Finally, we generalize its use to find homogeneous regions for cellular processes such as cell division, cell rearrangement, or cell size and shape changes. We also illustrate it on wing blade morphogenesis. This pipeline will contribute substantially to the analysis of complex tissue shaping, and the biochemical and biomechanical regulations driving tissue morphogenesis. [ABSTRACT FROM AUTHOR]

Tissue segregation plays a crucial role in the measurement of breast density in breast magnetic resonance (MR) images. This paper proposes a mathematical analysis of the new distribution mixture model for the intensity spectrum of breast MR images using Gamma and Gaussian distribution for fibro-glandular and adipose tissues, respectively. The thorough regression analysis and mapping presented in this paper clearly indicate that the distribution of Gamma is best suited to the spectrum of fibro-glandular tissue intensities relative to the standard Gaussian distribution. Moreover, Gamma distribution can represent both symmetric and non-symmetric (skewed) intensity distributions in a more efficient way, leading to a more accurate segmentation of fibro-glandular and adipose tissues. The efficiency of the segmentation is quantified by measuring the standard performance appraisal steps : Dice similarity coefficient, Jaccard index and dissimilarity index. The whole mathematical analysis is performed on a data set of 200 patients with 160 axial slices per subject with various breast sizes and densities. The Gamma Gaussian mixture model (GaGMM's) assessment metrics indicate an improvement of 39.4 %, 46.8 % and 54.9 %, respectively, in relation to the Gaussian mixture model. [ABSTRACT FROM AUTHOR]

Background: Current registration methods for diffusion-MRI (dMRI) data mostly focus on white matter (WM) areas. Recently, dMRI has been employed for the characterization of gray matter (GM) microstructure, emphasizing the need for registration methods that consider all tissue types.Purpose: To develop a dMRI registration method based on GM, WM, and cerebrospinal fluid (CSF) tissue probability maps (TPMs).Study Type: Retrospective longitudinal study.Population: Thirty-two healthy participants were scanned twice (legacy data), divided into a training-set (n = 16) and a test-set (n = 16), and 35 randomly-selected participants from the Human Connectome Project.Field Strength/sequence: 3.0T, diffusion-weighted spin-echo echo-planar sequence; T1-weighted spoiled gradient-recalled echo (SPGR) sequence.Assessment: A joint segmentation-registration approach was implemented: Diffusion tensor imaging (DTI) maps were classified into TPMs using machine-learning approaches. The resulting GM, WM, and CSF probability maps were employed as features for image alignment. Validation was performed on the test dataset and the HCP dataset. Registration performance was compared with current mainstream registration tools.Statistical Tests: Classifiers used for segmentation were evaluated using leave-one-out cross-validation and scored using Dice-index. Registration success was evaluated by voxel-wise variance, normalized cross-correlation of registered DTI maps, intra- and inter-subject similarity of the registered TPMs, and region-based intra-subject similarity using an anatomical atlas. One-way ANOVAs were performed to compare between our method and other registration tools.Results: The proposed method outperformed mainstream registration tools as indicated by lower voxel-wise variance of registered DTI maps (SD decrease of 10%) and higher similarity between registered TPMs within and across participants, for all tissue types (Dice increase of 0.1-0.2; P < 0.05).Data Conclusion: A joint segmentation-registration approach based on diffusion-driven TPMs provides a more accurate registration of dMRI data, outperforming other registration tools. Our method offers a "translation" of diffusion data into structural information in the form of TPMs, allowing to directly align diffusion and structural images.Level Of Evidence: 1 Technical Efficacy Stage: 1. [ABSTRACT FROM AUTHOR]

Abstract Background Accurate measurement and reconstruction of orbital soft tissue is important to diagnosis and treatment of orbital diseases. This study applied an interactive graph cut method to orbital soft tissue precise segmentation and calculation in computerized tomography (CT) images, and to estimate its application in orbital reconstruction. Methods The interactive graph cut method was introduced to segment extraocular muscle and intraorbital fat in CT images. Intra- and inter-observer variability of tissue volume measured by graph cut segmentation was validated. Accuracy and reliability of the method was accessed by comparing with manual delineation and commercial medical image software. Intraorbital structure of 10 patients after enucleation surgery was reconstructed based on graph cut segmentation and soft tissue volume were compared within two different surgical techniques. Results Both muscle and fat tissue segmentation results of graph cut method showed good consistency with ground truth in phantom data. There were no significant differences in muscle calculations between observers or segmental methods (p > 0.05). Graph cut results of fat tissue had coincidental variable trend with ground truth which could identify 0.1cm3 variation. The mean performance time of graph cut segmentation was significantly shorter than manual delineation and commercial software (p

Accurate brain tissue segmentation in magnetic resonance imaging (MRI) has attracted the attention of medical doctors and researchers since variations in tissue volume and shape permit diagnosing and monitoring neurological diseases. Several proposals have been designed throughout the years comprising conventional machine learning strategies as well as convolutional neural networks (CNNs) approaches. In particular, in this paper, we analyze a sub-group of deep learning methods producing dense predictions. This branch, referred in the literature as fully CNN (FCNN), is of interest as these architectures can process an input volume in less time than CNNs. Our study focuses on understanding the architectural strengths and weaknesses of literature-like approaches. We implement eight FCNN architectures inspired by robust state-of-the-art methods on brain segmentation related tasks and use them within a standard pipeline. We evaluate them using the IBSR18, MICCAI2012, and iSeg2017 datasets as they contain infant and adult data and exhibit different voxel spacing, image quality, number of scans, and available imaging modalities. The discussion is driven in four directions: comparison between 2D and 3D approaches, the relevance of multiple imaging sequences, the effect of patch size, and the impact of patch overlap as a sampling strategy for training and testing models. Besides the aforementioned analysis, we show that the methods under evaluation can yield top performance on the three data collections. A public version is accessible to download from our research website to encourage other researchers to explore the evaluation framework.