Knowledge distillation with ensembles of convolutional neural networks for medical image segmentation.

Ensembles of convolutional neural networks (CNNs) often outperform a single CNN in medical image segmentation tasks, but inference is computationally more expensive and makes ensembles unattractive for some applications. We compared the performance of differently constructed ensembles with the performance of CNNs derived from these ensembles using knowledge distillation, a technique for reducing the footprint of large models such as ensembles. We investigated two different types of ensembles, namely, diverse ensembles of networks with three different architectures and two different loss-functions, and uniform ensembles of networks with the same architecture but initialized with different random seeds.

A Simulation Toolkit for Testing the Sensitivity and Accuracy of Corticometry Pipelines.

In recent years, the replicability of neuroimaging findings has become an important concern to the research community. Neuroimaging pipelines consist of myriad numerical procedures, which can have a cumulative effect on the accuracy of findings. To address this problem, we propose a method for simulating artificial lesions in the brain in order to estimate the sensitivity and specificity of lesion detection, using different automated corticometry pipelines.

LAMA: automated image analysis for the developmental phenotyping of mouse embryos.

Advanced 3D imaging modalities, such as micro-computed tomography (micro-CT), have been incorporated into the high-throughput embryo pipeline of the International Mouse Phenotyping Consortium (IMPC). This project generates large volumes of raw data that cannot be immediately exploited without significant resources of personnel and expertise. Thus, rapid automated annotation is crucial to ensure that 3D imaging data can be integrated with other multi-dimensional phenotyping data.

Magnetic resonance imaging reveals specific anatomical changes in the brain of Agat- and Gamt-mice attributed to creatine depletion and guanidinoacetate alteration.

Arginine:glycine amidinotransferase- and guanidinoacetate methyltransferase deficiency are severe neurodevelopmental disorders. It is not known whether mouse models of disease express a neuroanatomical phenotype. High-resolution magnetic resonance imaging (MRI) with advanced image analysis was performed in perfused, fixed mouse brains encapsulated with the skull from male, 10-12 week old Agat and B6J.

Progressive neuroanatomical changes caused by Grin1 loss-of-function mutation.

NMDA receptor dysfunction is central to the encephalopathies caused by missense mutations in the NMDA receptor subunit genes. Missense variants of GRIN1, GRIN2A, and GRIN2B cause similar syndromes with varying severity of intellectual impairment, autism, epilepsy, and motor dysfunction. To gain insight into possible biomarkers of NMDAR hypofunction, we asked whether a loss-of-function variant in the Grin1 gene would cause structural changes in the brain that could be detected by MRI.

The effect of automated landmark identification on morphometric analyses.

Morphometric analysis of anatomical landmarks allows researchers to identify specific morphological differences between natural populations or experimental groups, but manually identifying landmarks is time-consuming. We compare manually and automatically generated adult mouse skull landmarks and subsequent morphometric analyses to elucidate how switching from manual to automated landmarking will impact morphometric analysis results for large mouse (Mus musculus) samples (n = 1205) that represent a wide range of 'normal' phenotypic variation (62 genotypes). Other studies have suggested that the use of automated landmarking methods is feasible, but this study is the first to compare the utility of current automated approaches to manual landmarking for a large dataset that allows the quantification of intra- and inter-strain variation.

MRI to Assess Neurological Function.

This article describes a detailed set of protocols for mouse brain imaging using MRI. We focus primarily on measuring changes in neuroanatomy, and provide both instructions for mouse preparation and details on image acquisition, image processing, and statistics. Practical details as well as theoretical considerations are provided.

Structural covariance of brain region volumes is associated with both structural connectivity and transcriptomic similarity.

An organizational pattern seen in the brain, termed structural covariance, is the statistical association of pairs of brain regions in their anatomical properties. These associations, measured across a population as covariances or correlations usually in cortical thickness or volume, are thought to reflect genetic and environmental underpinnings. Here, we examine the biological basis of structural volume covariance in the mouse brain.

P63 expression plays a role in developmental rate, embryo size, and local morphogenesis.

Background: The p63 gene is integral to the development of many body parts including limb, palate, teeth, and urogenital tract. Loss of p63 expression may alter developmental rate, which is crucial to normal morphogenesis. To validate a novel, unbiased embryo phenotyping software tool, we tested whether delayed development contributes to the pathological phenotype of a p63 mouse mutant (p63 ).

Neurobiological Mechanisms of Chemotherapy-induced Cognitive Impairment in a Transgenic Model of Breast Cancer.

Animal studies have reinforced clinical reports of cognitive impairment in cancer survivors following chemotherapy but, until now, all pre-clinical research in this area has been conducted on normal rodents. The present study investigated the effects of chemotherapy on cognition and underlying biological mechanisms in the FVB/N-Tg (MMTV-neu) 202 Mul/J mouse, a well-characterized transgenic model of breast cancer that has similarities to the tumorigenesis which occurs in humans. Tumor-bearing and control mice received three weekly injections of a combination of methotrexate + 5-fluorouracil, or an equal volume of saline.

Variability of brain anatomy for three common mouse strains.

The way in which brain structures express different morphologies is not fully understood. Here we investigate variability in brain anatomy using ex vivo MRI of three common laboratory mouse strains: in two inbred strains (C57BL/6 and 129S6) and one outbred strain (CD-1). We use Generalised Procrustes Analysis (GPA) to estimate modes of anatomical variability.

4D atlas of the mouse embryo for precise morphological staging.

After more than a century of research, the mouse remains the gold-standard model system, for it recapitulates human development and disease and is quickly and highly tractable to genetic manipulations. Fundamental to the power and success of using a mouse model is the ability to stage embryonic mouse development accurately. Past staging systems were limited by the technologies of the day, such that only surface features, visible with a light microscope, could be recognized and used to define stages.

Pydpiper: a flexible toolkit for constructing novel registration pipelines.

Using neuroimaging technologies to elucidate the relationship between genotype and phenotype and brain and behavior will be a key contribution to biomedical research in the twenty-first century. Among the many methods for analyzing neuroimaging data, image registration deserves particular attention due to its wide range of applications. Finding strategies to register together many images and analyze the differences between them can be a challenge, particularly given that different experimental designs require different registration strategies.

Whole-brain mapping of behaviourally induced neural activation in mice.

The ability to visualize behaviourally evoked neural activity patterns across the rodent brain is essential for understanding the distributed brain networks mediating particular behaviours. However, current imaging methods are limited in their spatial resolution and/or ability to obtain brain-wide coverage of functional activity. Here, we describe a new automated method for obtaining cellular-level, whole-brain maps of behaviourally induced neural activity in the mouse.

Mapping registration sensitivity in MR mouse brain images.

Nonlinear registration algorithms provide a way to estimate structural (brain) differences based on magnetic resonance images. Their ability to align images of different individuals and across modalities has been well-researched, but the bounds of their sensitivity with respect to the recovery of salient morphological differences between groups are unclear. Here we develop a novel approach to simulate deformations on MR brain images to evaluate the ability of two registration algorithms to extract structural differences corresponding to biologically plausible atrophy and expansion.

Effects of a ketogenic diet during pregnancy on embryonic growth in the mouse.

Background: The increasing use of the ketogenic diet (KD), particularly by women of child-bearing age, raises a question about its suitability during gestation. To date, no studies have thoroughly investigated the direct implications of a gestational ketogenic diet on embryonic development.

Methods: To fill this knowledge gap we imaged CD-1 mouse embryos whose mothers were fed either a Standard Diet (SD) or a KD 30 days prior to, as well as during gestation.

Proximal cerebral arteries develop myogenic responsiveness in heart failure via tumor necrosis factor-α-dependent activation of sphingosine-1-phosphate signaling.

Background: Heart failure is associated with neurological deficits, including cognitive dysfunction. However, the molecular mechanisms underlying reduced cerebral blood flow in the early stages of heart failure, particularly when blood pressure is minimally affected, are not known.

Methods And Results: Using a myocardial infarction model in mice, we demonstrate a tumor necrosis factor-α (TNFα)-dependent enhancement of posterior cerebral artery tone that reduces cerebral blood flow before any overt changes in brain structure and function.

Performing label-fusion-based segmentation using multiple automatically generated templates.

Classically, model-based segmentation procedures match magnetic resonance imaging (MRI) volumes to an expertly labeled atlas using nonlinear registration. The accuracy of these techniques are limited due to atlas biases, misregistration, and resampling error. Multi-atlas-based approaches are used as a remedy and involve matching each subject to a number of manually labeled templates.

Cerebellar abnormalities in purine nucleoside phosphorylase deficient mice.

Inherited defects in purine nucleoside phosphorylase (PNP) cause severe T cell immunodeficiency and progressive neurological dysfunction, yet little is known about the effects of PNP deficiency on the brain. PNP-KO mice display metabolic and immune anomalies similar to those observed in patients. Our objectives were to characterize brain abnormalities in PNP-KO mice and determine whether restoring PNP activity prevents these abnormalities.

Suppressor of fused controls mid-hindbrain patterning and cerebellar morphogenesis via GLI3 repressor.

Sonic Hedgehog and its GLI transcriptional effectors control foliation complexity during cerebellar morphogenesis by promoting granule cell precursor proliferation. Here, we reveal a novel contribution of Sonic Hedgehog-GLI signaling to cerebellar patterning and cell differentiation by generating mice with targeted deletion of Suppressor of Fused (SuFu), a regulator of Sonic Hedgehog signaling, in the mid-hindbrain. Postnatal SuFu-deficient mice exhibit impaired motor coordination and severe cerebellar mispatterning.

Abnormalities in brain structure and behavior in GSK-3alpha mutant mice.

Background: Glycogen synthase kinase-3 (GSK-3) is a widely expressed and highly conserved serine/threonine protein kinase encoded by two genes that generate two related proteins: GSK-3alpha and GSK-3beta. Mice lacking a functional GSK-3alpha gene were engineered in our laboratory; they are viable and display insulin sensitivity. In this study, we have characterized brain functions of GSK-3alpha KO mice by using a well-established battery of behavioral tests together with neurochemical and neuroanatomical analysis.

Time course and nature of brain atrophy in the MRL mouse model of central nervous system lupus.

Objective: Similar to patients with systemic lupus erythematosus, autoimmune MRL/lpr mice spontaneously develop behavioral deficits and pathologic changes in the brain. Given that the disease-associated brain atrophy in this model is not well understood, the present study was undertaken to determine the time course of morphometric changes in major brain structures of autoimmune MRL/lpr mice.

Methods: Computerized planimetry and high-resolution magnetic resonance imaging (MRI) were used to compare the areas and volumes of brain structures in cohorts of mice that differ in severity of lupus-like disease.

Regionally reduced brain volume, altered serotonin neurochemistry, and abnormal behavior in mice null for the circadian rhythm output gene Magel2.

Magel2 belongs to the MAGE/necdin family of proteins, which have roles in cell cycle, differentiation, and apoptosis. The Magel2 gene is expressed in various brain regions, most notably the hypothalamus. Mice with a targeted deletion of Magel2 display hypoactivity, blunted circadian rhythm, decreased fertility, and increased adiposity.