Developing Electron Microscopy Tools for Profiling Plasma Lipoproteins Using Methyl Cellulose Embedment, Machine Learning and Immunodetection of Apolipoprotein B and Apolipoprotein(a).

Plasma lipoproteins are important carriers of cholesterol and have been linked strongly to cardiovascular disease (CVD). Our study aimed to achieve fine-grained measurements of lipoprotein subpopulations such as low-density lipoprotein (LDL), lipoprotein(a) (Lp(a), or remnant lipoproteins (RLP) using electron microscopy combined with machine learning tools from microliter samples of human plasma. In the reported method, lipoproteins were absorbed onto electron microscopy (EM) support films from diluted plasma and embedded in thin films of methyl cellulose (MC) containing mixed metal stains, providing intense edge contrast. The results show that LPs have a continuous frequency distribution of sizes, extending from LDL (> 15 nm) to intermediate density lipoprotein (IDL) and very low-density lipoproteins (VLDL). Furthermore, mixed metal staining produces striking "positive" contrast of specific antibodies attached to lipoproteins providing quantitative data on apolipoprotein(a)-positive Lp(a) or apolipoprotein B (ApoB)-positive particles. To enable automatic particle characterization, we also demonstrated efficient segmentation of lipoprotein particles using deep learning software characterized by a Mask Region-based Convolutional Neural Networks (R-CNN) architecture with transfer learning. In future, EM and machine learning could be combined with microarray deposition and automated imaging for higher throughput quantitation of lipoproteins associated with CVD risk.

Comparing computer-generated and pathologist-generated tumour segmentations for immunohistochemical scoring of breast tissue microarrays.

BACKGROUND: Tissue microarrays (TMAs) have become a valuable resource for biomarker expression in translational research. Immunohistochemical (IHC) assessment of TMAs is the principal method for analysing large numbers of patient samples, but manual IHC assessment of TMAs remains a challenging and laborious task. With advances in image analysis, computer-generated analyses of TMAs have the potential to lessen the burden of expert pathologist review. METHODS: In current commercial software computerised oestrogen receptor (ER) scoring relies on tumour localisation in the form of hand-drawn annotations. In this study, tumour localisation for ER scoring was evaluated comparing computer-generated segmentation masks with those of two specialist breast pathologists. Automatically and manually obtained segmentation masks were used to obtain IHC scores for thirty-two ER-stained invasive breast cancer TMA samples using FDA-approved IHC scoring software. RESULTS: Although pixel-level comparisons showed lower agreement between automated and manual segmentation masks (κ=0.81) than between pathologists' masks (κ=0.91), this had little impact on computed IHC scores (Allred; =0.91, Quickscore; =0.92). CONCLUSIONS: The proposed automated system provides consistent measurements thus ensuring standardisation, and shows promise for increasing IHC analysis of nuclear staining in TMAs from large clinical trials.

Automatic Segmentation of Osteonal Microstructure in Human Cortical Bone Using Deep Learning: A Proof of Concept.

Cortical bone microstructure assessment in biological and forensic anthropology can assist with the estimation of age-at-death and animal-human differentiation, for example. Osteonal structures within cortical bone are the key feature under analysis, with osteon frequency and metric parameters providing crucial information for the assessment. Currently, the histomorphological assessment consists of a time-consuming manual process for which specific training is required. Our work investigates the feasibility of automatic analysis of human bone microstructure images through the application of deep learning. In this paper, we use a U-Net architecture to address the semantic segmentation of such images into three classes: intact osteons, fragmentary osteons, and background. Data augmentation was used to avoid overfitting. We evaluated our fully automatic approach using a sample of 99 microphotographs. The contours of intact and fragmentary osteons were traced manually to provide ground truth. The Dice coefficients were 0.73 for intact osteons, 0.38 for fragmented osteons, and 0.81 for background, giving an average of 0.64. The Dice coefficient of the binary classification osteon-background was 0.82. Although further refinement of the initial model and tests with larger datasets are needed, this study provides, to the best of our knowledge, the first proof of concept for the use of computer vision and deep learning for differentiating both intact and fragmentary osteons in human cortical bone. This approach has the potential to widen and facilitate the use of histomorphological assessment in the biological and forensic anthropology communities.

Recognition of immunogold markers in electron micrographs.

Immunoelectron microscopy is used in cell biological research to study the spatial distribution of intracellular macromolecules at the ultrastructural level. Colloidal gold particles (immunogold markers) are commonly used to localise molecules of interest on ultrathin sections and can be visualised in transmission electron micrographs as dark spots. Quantitative analysis involves detection of the immunogold markers, and is often performed manually or interactively as part of a stereological estimation technique. The method presented in this paper automatically detects and counts immunogold markers, estimating the location, size and type of each marker. It was evaluated on single-labelled as well as double-labelled images showing markers of two different sizes. This is a first step towards automatic analysis of immunoelectron micrographs, enabling a rapid and more complete quantitative analysis than is currently practicable.

Automated motion estimation of root responses to sucrose in two Arabidopsis thaliana genotypes using confocal microscopy.

Root growth is a highly dynamic process influenced by genetic background and environment. This paper reports the development of R scripts that enable root growth kinematic analysis that complements a new motion analysis tool: PlantVis. Root growth of Arabidopsis thaliana expressing a plasma membrane targeted GFP (C24 and Columbia 35S:LTI6b-EGFP) was imaged using time-lapse confocal laser scanning microscopy. Displacement of individual pixels in the time-lapse sequences was estimated automatically by PlantVis, producing dense motion vector fields. R scripts were developed to extract kinematic growth parameters and report displacement to ± 0.1 pixel. In contrast to other currently available tools, Plantvis-R delivered root velocity profiles without interpolation or averaging across the root surface and also estimated the uncertainty associated with tracking each pixel. The PlantVis-R analysis tool has a range of potential applications in root physiology and gene expression studies, including linking motion to specific cell boundaries and analysis of curvature. The potential for quantifying genotype × environment interactions was examined by applying PlantVis-R in a kinematic analysis of root growth of C24 and Columbia, under contrasting carbon supply. Large genotype-dependent effects of sucrose were recorded. C24 exhibited negligible differences in elongation zone length and elongation rate but doubled the density of lateral roots in the presence of sucrose. Columbia, in contrast, increased its elongation zone length and doubled its elongation rate and the density of lateral roots.

High-Throughput, Time-Resolved Mechanical Phenotyping of Prostate Cancer Cells.

Worldwide, prostate cancer sits only behind lung cancer as the most commonly diagnosed form of the disease in men. Even the best diagnostic standards lack precision, presenting issues with false positives and unneeded surgical intervention for patients. This lack of clear cut early diagnostic tools is a significant problem. We present a microfluidic platform, the Time-Resolved Hydrodynamic Stretcher (TR-HS), which allows the investigation of the dynamic mechanical response of thousands of cells per second to a non-destructive stress. The TR-HS integrates high-speed imaging and computer vision to automatically detect and track single cells suspended in a fluid and enables cell classification based on their mechanical properties. We demonstrate the discrimination of healthy and cancerous prostate cell lines based on the whole-cell, time-resolved mechanical response to a hydrodynamic load. Additionally, we implement a finite element method (FEM) model to characterise the forces responsible for the cell deformation in our device. Finally, we report the classification of the two different cell groups based on their time-resolved roundness using a decision tree classifier. This approach introduces a modality for high-throughput assessments of cellular suspensions and may represent a viable application for the development of innovative diagnostic devices.

Automated assessment of polyethylene wear in cemented acetabular components using anteroposterior radiographs of total hip replacements.

Polyethylene wear in the acetabular components of hip prostheses is implicated in loosening and failure. Radiographic measurement of wear is used to identify patients at risk and to assess prosthesis designs. This paper focuses on analysis of prostheses with cemented acetabular cups from anteroposterior (AP) radiographs. The articular surface of the femoral head and the acetabular rim marker are modelled as spherical and circular respectively, resulting in elliptical image projections. Methods for automatically localising these structures in radiographs are presented using robust ellipse fitting and various error functions. Special attention is paid to the acetabular marker since this often projects as a highly eccentric ellipse. Robust fitting enables successful localisation in the presence of clutter without the need for user interaction. Finally, the use of these ellipses as reference structures for wear estimation is investigated and the effect of eccentricity errors is highlighted.

Structure Prediction for Gland Segmentation With Hand-Crafted and Deep Convolutional Features.

We present a novel method to segment instances of glandular structures from colon histopathology images. We use a structure learning approach which represents local spatial configurations of class labels, capturing structural information normally ignored by sliding-window methods. This allows us to reveal different spatial structures of pixel labels (e.g., locations between adjacent glands, or far from glands), and to identify correctly neighboring glandular structures as separate instances. Exemplars of label structures are obtained via clustering and used to train support vector machine classifiers. The label structures predicted are then combined and post-processed to obtain segmentation maps. We combine hand-crafted, multi-scale image features with features computed by a deep convolutional network trained to map images to segmentation maps. We evaluate the proposed method on the public domain GlaS data set, which allows extensive comparisons with recent, alternative methods. Using the GlaS contest protocol, our method achieves the overall best performance.

Learning active shape models for bifurcating contours.

Statistical shape models are often learned from examples based on landmark correspondences between annotated examples. A method is proposed for learning such models from contours with inconsistent bifurcations and loops. Automatic segmentation of tibial and femoral contours in knee X-ray images is investigated as a step towards reliable, quantitative radiographic analysis of osteoarthritis for diagnosis and assessment of progression. Results are presented using various features, the Mahalanobis distance, distance weighted K-nearest neighbours, and two relevance vector machine-based methods as quality of fit measure.

Discriminating dysplasia: Optical tomographic texture analysis of colorectal polyps.

Optical projection tomography enables 3-D imaging of colorectal polyps at resolutions of 5-10 µm. This paper investigates the ability of image analysis based on 3-D texture features to discriminate diagnostic levels of dysplastic change from such images, specifically, low-grade dysplasia, high-grade dysplasia and invasive cancer. We build a patch-based recognition system and evaluate both multi-class classification and ordinal regression formulations on a 90 polyp dataset. 3-D texture representations computed with a hand-crafted feature extractor, random projection, and unsupervised image filter learning are compared using a bag-of-words framework. We measure performance in terms of error rates, F-measures, and ROC surfaces. Results demonstrate that randomly projected features are effective. Discrimination was improved by carefully manipulating various important aspects of the system, including class balancing, output calibration and approximation of non-linear kernels.

Classification and immunohistochemical scoring of breast tissue microarray spots.

Tissue microarrays (TMAs) facilitate the survey of very large numbers of tumors. However, the manual assessment of stained TMA sections constitutes a bottleneck in the pathologist's work flow. This paper presents a computational pipeline for automatically classifying and scoring breast cancer TMA spots that have been subjected to nuclear immunostaining. Spots are classified based on a bag of visual words approach. Immunohistochemical scoring is performed by computing spot features reflecting the proportion of epithelial nuclei that are stained and the strength of that staining. These are then mapped onto an ordinal scale used by pathologists. Multilayer perceptron classifiers are compared with latent topic models and support vector machines for spot classification, and with Gaussian process ordinal regression and linear models for scoring. Intraobserver variation is also reported. The use of posterior entropy to identify uncertain cases is demonstrated. Evaluation is performed using TMA images stained for progesterone receptor.

Immunohistochemical analysis of breast tissue microarray images using contextual classifiers.

BACKGROUND: Tissue microarrays (TMAs) are an important tool in translational research for examining multiple cancers for molecular and protein markers. Automatic immunohistochemical (IHC) scoring of breast TMA images remains a challenging problem. METHODS: A two-stage approach that involves localization of regions of invasive and in-situ carcinoma followed by ordinal IHC scoring of nuclei in these regions is proposed. The localization stage classifies locations on a grid as tumor or non-tumor based on local image features. These classifications are then refined using an auto-context algorithm called spin-context. Spin-context uses a series of classifiers to integrate image feature information with spatial context information in the form of estimated class probabilities. This is achieved in a rotationally-invariant manner. The second stage estimates ordinal IHC scores in terms of the strength of staining and the proportion of nuclei stained. These estimates take the form of posterior probabilities, enabling images with uncertain scores to be referred for pathologist review. RESULTS: The method was validated against manual pathologist scoring on two nuclear markers, progesterone receptor (PR) and estrogen receptor (ER). Errors for PR data were consistently lower than those achieved with ER data. Scoring was in terms of estimated proportion of cells that were positively stained (scored on an ordinal scale of 0-6) and perceived strength of staining (scored on an ordinal scale of 0-3). Average absolute differences between predicted scores and pathologist-assigned scores were 0.74 for proportion of cells and 0.35 for strength of staining (PR). CONCLUSIONS: The use of context information via spin-context improved the precision and recall of tumor localization. The combination of the spin-context localization method with the automated scoring method resulted in reduced IHC scoring errors.

Learning from partially annotated OPT images by contextual relevance ranking.

Annotations delineating regions of interest can provide valuable information for training medical image classification and segmentation methods. However the process of obtaining annotations is tedious and time-consuming, especially for high-resolution volumetric images. In this paper we present a novel learning framework to reduce the requirement of manual annotations while achieving competitive classification performance. The approach is evaluated on a dataset with 59 3D optical projection tomography images of colorectal polyps. The results show that the proposed method can robustly infer patterns from partially annotated images with low computational cost.

RERBEE: robust efficient registration via bifurcations and elongated elements applied to retinal fluorescein angiogram sequences.

We present RERBEE (robust efficient registration via bifurcations and elongated elements), a novel feature-based registration algorithm able to correct local deformations in high-resolution ultra-wide field-of-view (UWFV) fluorescein angiogram (FA) sequences of the retina. The algorithm is able to cope with peripheral blurring, severe occlusions, presence of retinal pathologies and the change of image content due to the perfusion of the fluorescein dye in time. We have used the computational power of a graphics processor to increase the performance of the most computationally expensive parts of the algorithm by a factor of over × 1300, enabling the algorithm to register a pair of 3900 × 3072 UWFV FA images in 5-10 min instead of the 5-7 h required using only the CPU. We demonstrate accurate results on real data with 267 image pairs from a total of 277 (96.4%) graded as correctly registered by a clinician and 10 (3.6%) graded as correctly registered with minor errors but usable for clinical purposes. Quantitative comparison with state-of-the-art intensity-based and feature-based registration methods using synthetic data is also reported. We also show some potential usage of a correctly aligned sequence for vein/artery discrimination and automatic lesion detection.

Root responses to soil physical conditions; growth dynamics from field to cell.

Root growth in the field is often slowed by a combination of soil physical stresses, including mechanical impedance, water stress, and oxygen deficiency. The stresses operating may vary continually, depending on the location of the root in the soil profile, the prevailing soil water conditions, and the degree to which the soil has been compacted. The dynamics of root growth responses are considered in this paper, together with the cellular responses that underlie them. Certain root responses facilitate elongation in hard soil, for example, increased sloughing of border cells and exudation from the root cap decreases friction; and thickening of the root relieves stress in front of the root apex and decreases buckling. Whole root systems may also grow preferentially in loose versus dense soil, but this response depends on genotype and the spatial arrangement of loose and compact soil with respect to the main root axes. Decreased root elongation is often accompanied by a decrease in both cell flux and axial cell extension, and recent computer-based models are increasing our understanding of these processes. In the case of mechanical impedance, large changes in cell shape occur, giving rise to shorter fatter cells. There is still uncertainty about many aspects of this response, including the changes in cell walls that control axial versus radial extension, and the degree to which the epidermis, cortex, and stele control root elongation. Optical flow techniques enable tracking of root surfaces with time to yield estimates of two-dimensional velocity fields. It is demonstrated that these techniques can be applied successfully to time-lapse sequences of confocal microscope images of living roots, in order to determine velocity fields and strain rates of groups of cells. In combination with new molecular approaches this provides a promising way of investigating and modelling the mechanisms controlling growth perturbations in response to environmental stresses.

Laparoscopic distal pancreatectomy with splenic preservation for serous cystadenoma: a case report and literature review.

A minimally invasive approach can be beneficial in a spleen-preserving distal pancreatectomy. This article reports a 71-year-old woman who presented to her internist with hypertension and persistent hypokalemia. A computed tomography scan to rule out a functional adrenal mass incidentally revealed a 4 cm x 3 cm x 2 cm serous cystadenoma of the distal pancreas and normal adrenal glands. The patient was referred to the general surgery service for resection of the distal pancreatic lesion. A laparoscopic spleen-preserving distal pancreatectomy was performed. The lesion was completely excised, and the pathology revealed serous cystadenoma with focal fibrosis and atrophic acini. The postoperative advantages of this approach were the early return of bowel function, minimal narcotic requirements, and early resumption of normal activities. This case illustrates the advantages of minimally invasive surgery in the performance of a spleen-preserving distal pancreatectomy.