Geometric morphometrics is the statistical analysis of landmark-based shape variation and its covariation with other variables. Over the past two decades, the gold standard of landmark data ...acquisition has been manual detection by a single observer. This approach has proven accurate and reliable in small-scale investigations. However, big data initiatives are increasingly common in biology and morphometrics. This requires fast, automated, and standardized data collection. We combine techniques from image registration, geometric morphometrics, and deep learning to automate and optimize anatomical landmark detection. We test our method on high-resolution, micro-computed tomography images of adult mouse skulls. To ensure generalizability, we use a morphologically diverse sample and implement fundamentally different deformable registration algorithms. Compared to landmarks derived from conventional image registration workflows, our optimized landmark data show up to a 39.1% reduction in average coordinate error and a 36.7% reduction in total distribution error. In addition, our landmark optimization produces estimates of the sample mean shape and variance–covariance structure that are statistically indistinguishable from expert manual estimates. For biological imaging datasets and morphometric research questions, our approach can eliminate the time and subjectivity of manual landmark detection whilst retaining the biological integrity of these expert annotations.
Characterising phenotypes often requires quantification of anatomical shape. Quantitative shape comparison (morphometrics) traditionally uses manually located landmarks and is limited by landmark ...number and operator accuracy. Here, we apply a landmark-free method to characterise the craniofacial skeletal phenotype of the Dp1Tyb mouse model of Down syndrome and a population of the Diversity Outbred (DO) mouse model, comparing it with a landmark-based approach. We identified cranial dysmorphologies in Dp1Tyb mice, especially smaller size and brachycephaly (front-back shortening), homologous to the human phenotype. Shape variation in the DO mice was partly attributable to allometry (size-dependent shape variation) and sexual dimorphism. The landmark-free method performed as well as, or better than, the landmark-based method but was less labour-intensive, required less user training and, uniquely, enabled fine mapping of local differences as planar expansion or shrinkage. Its higher resolution pinpointed reductions in interior mid-snout structures and occipital bones in both the models that were not otherwise apparent. We propose that this landmark-free pipeline could make morphometrics widely accessible beyond its traditional niches in zoology and palaeontology, especially in characterising developmental mutant phenotypes.
During the development of the face, tissues move, change shape, and fuse in tightly orchestrated patterns to create all the parts of a normal face. These shape changes are driven by factors such as ...cell signaling, migration, proliferation, and apoptosis. However, the contributions of each of these drivers to morphogenesis are poorly studied. Here, we explore differential cell proliferation as a driver of mouse facial morphogenesis. We quantify patterns in both the spatial distribution and orientation of proliferation in the developing face in 3D over a critical period of murine facial development (E9.5‐E11.5). We use immunostaining with light sheet microscopy (LSM) to capture total and proliferating nuclei. To compare proliferative density in facial tissues, we segment these images using a novel convolutional neural network. We then generate atlases of average proliferation at each half‐day age point within our range and use these to identify relationships between morphology and cell proliferation. We show that regions with more dense proliferation tend to undergo more intensive shape changes. We then simulate outgrowth of the maxillary process using a cell simulation engine, PhysiCell, to demonstrate that differential proliferation is necessary to maintain expected morphology in growing tissues. In addition to differential proliferation, localized orientation of cell division could also affect morphology. In plants and some animal tissues, including murine limb buds, preferentially oriented cell proliferation drives shape change by causing tissue elongation in specific directions. We explore the orientation and distribution of cell proliferation using LSM: we inject pregnant dams with a synthetic nucleotide, EdU, 5 minutes before harvest to mark the daughter cells of proliferative events occurring in the interim. We then compare the angles of the proliferative axes for each pair of daughters relative to the primary direction of tissue growth. Preliminary results suggest that cell proliferation in the maxillary and nasal processes is oriented preferentially towards the axis of tissue growth. These results suggest that both the distribution and orientation of cell proliferation play a role in murine facial morphogenesis. Understanding the mechanisms underlying morphogenesis is important to guide future research that could lead to earlier and more robust diagnosis and treatment of syndromes and facial abnormalities.
A variety of genetic mutations affect cell proliferation during organism development, leading to structural birth defects. However, the mechanisms by which these alterations influence the development ...of the face remain unclear. Cell proliferation and its relation to shape variation can be studied using Light-Sheet Microscopy (LSM) imaging across a range of developmental time points using mouse models. The aim of this work was to develop and evaluate accurate automatic methods based on convolutional neural networks (CNNs) for: (i) tissue segmentation (neural ectoderm and mesenchyme), (ii) cell segmentation in nuclear-stained images, and (iii) segmentation of proliferating cells in phospho-Histone H3 (pHH3)-stained LSM images of mouse embryos. For training and evaluation of the CNN models, 155 to 176 slices from 10 mouse embryo LSM images with corresponding manual segmentations were available depending on the segmentation task. Three U-net CNN models were trained optimizing their loss functions, among other hyper-parameters, depending on the segmentation task. The tissue segmentation achieved a macro-average F-score of 0.84, whereas the inter-observer value was 0.89. The cell segmentation achieved a Dice score of 0.57 and 0.56 for nuclear-stained and pHH3-stained images, respectively, whereas the corresponding inter-observer Dice scores were 0.39 and 0.45, respectively. The proposed pipeline using the U-net CNN architecture can accelerate LSM image analysis and together with the annotated datasets can serve as a reference for comparison of more advanced LSM image segmentation methods in future.
Reference intervals (RIs) of carotid intima media thickness (CIMT) from large healthy population are still lacking in Latin America. The aim of this study was to determine CIMT RIs in a cohort of ...1012 healthy subjects from Argentina. We evaluated if RIs for males and females and for left and right carotids were necessary. Second, mean and standard deviation (SD) age-related equations were obtained for left, right, and average (left + right)/2) CIMT using parametric regression methods based on fractional polynomials, in order to obtain age-specific percentiles curves. Age-specific percentile curves were obtained. Males showed higher A-CIMT (0.577±0.003 mm versus 0.566±0.004 mm, P=0.039) in comparison with females. For males, the equations were as follows: A-CIMT mean = 0.42 + 8.14×10-5⁎Age2; A-CIMT SD = 5.9 × 10−2 + 1.09×10-5⁎Age2. For females, they were as follows: A-CIMT mean = 0.40 + 8.20×10-5⁎Age2; A-CIMT SD = 4.67 × 10−2 + 1.63×10-5⁎Age2. Our study provides the largest database concerning RIs of CIMT in healthy people in Argentina. Specific RIs and percentiles of CIMT for children, adolescents, and adults are now available according to age and gender, for right and left common carotid arteries.
Placental abnormalities have been sporadically implicated as a source of developmental heart defects. Yet it remains unknown how often the placenta is at the root of congenital heart defects (CHDs), ...and what the cellular mechanisms are that underpin this connection. Here, we selected three mouse mutant lines, Atp11a, Smg9 and Ssr2, that presented with placental and heart defects in a recent phenotyping screen, resulting in embryonic lethality. To dissect phenotype causality, we generated embryo- and trophoblast-specific conditional knockouts for each of these lines. This was facilitated by the establishment of a new transgenic mouse, Sox2-Flp, that enables the efficient generation of trophoblast-specific conditional knockouts. We demonstrate a strictly trophoblast-driven cause of the CHD and embryonic lethality in one of the three lines (Atp11a) and a significant contribution of the placenta to the embryonic phenotypes in another line (Smg9). Importantly, our data reveal defects in the maternal blood-facing syncytiotrophoblast layer as a shared pathology in placentally induced CHD models. This study highlights the placenta as a significant source of developmental heart disorders, insights that will transform our understanding of the vast number of unexplained congenital heart defects.
Canonical Wnt signaling plays multiple roles critical to normal craniofacial development while its dysregulation is known to be involved in structural birth defects of the face. However, when and how ...Wnt signaling influences phenotypic variation, including those associated with disease, remains unclear. One potential mechanism is via Wnt signaling's role in the patterning of an early facial signaling center, the frontonasal ectodermal zone (FEZ), and its subsequent regulation of early facial morphogenesis. For example, Wnt signaling may directly alter the shape and/or magnitude of expression of the
(
) domain in the FEZ. To test this idea, we used a replication-competent avian sarcoma retrovirus (RCAS) encoding
to modulate its expression in the facial mesenchyme. We then quantified and compared ontogenetic changes in treated to untreated embryos in the three-dimensional (3D) shape of both the
expression domain of the FEZ, and the morphology of the facial primordia and brain using iodine-contrast microcomputed tomography imaging and 3D geometric morphometrics (3DGM). We found that increased
expression in early stages of head development produces correlated variation in shape between both structural and signaling levels of analysis. In addition, altered
activation disrupted the integration between the forebrain and other neural tube derivatives. These results show that activation of Wnt signaling influences facial shape through its impact on the forebrain and
expression in the FEZ, and highlights the close relationship between morphogenesis of the forebrain and midface.
•Fully-automatic segmentation method for LI and MA in IVUS images•It matches and improves state-of-the-art fully-automatic segmentation methods.•Modular open-source implementation that can be ...extended and improved is provided.
Intravascular ultrasound (IVUS) provides axial grey-scale images of blood vessels. The large number of images require automatic analysis, specifically to identify the lumen and outer vessel wall. However, the high amount of noise, the presence of artifacts and anatomical structures, such as bifurcations, calcifications and fibrotic plaques, usually hinder the proper automatic segmentation of the vessel wall.
Lumen, media, adventitia and surrounding tissues are automatically detected using Support Vector Machines (SVMs). The classification performance of the SVMs vary according to the kind of structure present within each region of the image. Random Forest (RF) is used to detect different morphological structures and to modify the initial layer classification depending on the detected structure. The resulting classification maps are fed into a segmentation method based on deformable contours to detect lumen-intima (LI) and media-adventitia (MA) interfaces.
The modifications in the layer classifications according to the presence of structures proved to be effective improving LI and MA segmentations. The proposed method reaches a Jaccard Measure (JM) of 0.88 ± 0.08 for LI segmentation, compared with 0.88 ± 0.05 of a semiautomatic method. When looking at MA, our method reaches a JM of 0.84 ± 0.09, and outperforms previous automatic methods in terms of HD, with 0.51mm ± 0.30.
A simple modification to the arterial layer classification produces results that match and improve state-of-the-art fully-automatic segmentation methods for LI and MA in 20MHz IVUS images. For LI segmentation, the proposed automatic method performs accurately as semi-automatic methods. For MA segmentation, our method matched the quality of state-of-the-art automatic methods described in the literature. Furthermore, our implementation is modular and open-source, allowing for future extensions and improvements.
In an increasingly data-driven world, artificial intelligence is expected to be a key tool for converting big data into tangible benefits and the healthcare domain is no exception to this. Machine ...learning aims to identify complex patterns in multi-dimensional data and use these uncovered patterns to classify new unseen cases or make data-driven predictions. In recent years, deep neural networks have shown to be capable of producing results that considerably exceed those of conventional machine learning methods for various classification and regression tasks. In this paper, we provide an accessible tutorial of the most important supervised machine learning concepts and methods, including deep learning, which are potentially the most relevant for the medical domain. We aim to take some of the mystery out of machine learning and depict how machine learning models can be useful for medical applications. Finally, this tutorial provides a few practical suggestions for how to properly design a machine learning model for a generic medical problem.