Towards a computer aided diagnosis system dedicated to virtual microscopy based on stereology sampling and diffusion maps

An original strategy is presented, combining stereological sampling methods based on test grids and data reduction methods based on diffusion maps, in order to build a knowledge image database with no bias introduced by a subjective choice of exploration areas. The practical application of the exposed methodology concerns virtual slides of breast tumors.     

Mutagenesis in Muridae from regions with increased radiation level

Genetic monitoring of wild populations of mouse-like rodents has been carried out since 1992 in areas with the increased radioactivity in the south of the Bryansk region. The radioactive background ranged from 0.20 to 2.20 mkGy/h. Over 400 rodents of different species have been examined. Bank vole constituted more than a half of the examined animals. The genetic effect of radiation was estimated by the frequency of micronuclei in normochromatic erythrocytes of peripheral blood and by the frequency of abnormal sperm heads. In house mice the frequency of reciprocal translocations and the level of embryo mortality were studied in the progeny of the caught male mice mated with intact laboratory females. The data obtained demonstrate that the yield of induced genetic disorders by all tests was relatively low and only in a few cases the frequency of disturbances tended to increase with an increase in the level of radionuclide contamination in the examined areas.     

Merging regions across feature maps in texture segmentation

This paper describes segmentation phenomena of superimposed textures and the Linking phenomena. These phenomena provide us with information about the mechanism of a late stage of segmenting textures. The late stage takes place after the segmentation process forms regions in feature maps such that parameter values in one region are substantially different from those in the neighboring regions. At the stage, the segmentation process merges the regions across the feature maps to determine output regions by integrating information about how the regions occupy two-dimensional space. The segmentation process gets such information both from local areas and from areas far away from the local areas where it determines the output regions and boundaries.     

Interactive and automated application of virtual microscopy

Virtual microscopy can be applied in an interactive and an automated manner. Interactive application is performed in close association to conventional microscopy. It includes image standardization suitable to the performance of an individual pathologist such as image colorization, white color balance, or individual adjusted brightness. The steering commands have to include selection of wanted magnification, easy navigation, notification, and simple measurements (distances, areas). The display of the histological image should be adjusted to the physical limits of the human eye, which are determined by a view angle of approximately 35 seconds. A more sophisticated performance should include acoustic commands that replace the corresponding visual commands. Automated virtual microscopy includes so-called microscopy assistants which can be defined similar to the developed assistants in computer based editing systems (Microsoft Word, etc.). These include an automated image standardization and correction algorithms that excludes images of poor quality (for example uni-colored or out-of-focus images), an automated selection of the most appropriate field of view, an automated selection of the best magnification, and finally proposals of the most probable diagnosis. A quality control of the final diagnosis, and feedback to the laboratory determine the proposed system. The already developed tools of such a system are described in detail, as well as the results of first trials. In order to enhance the speed of such a system, and to allow further user-independent development a distributed implementation probably based upon Grid technology seems to be appropriate. The advantages of such a system as well as the present pathology environment and its expectations will be discussed in detail.     

Accuracy and perceptions of virtual microscopy compared with glass slide microscopy in cervical cytology

A. Evered and N. Dudding Accuracy and perceptions of virtual microscopy compared with glass slide microscopy in cervical cytology Objective: To evaluate virtual microscopy in terms of diagnostic performance and acceptability among practising cytologists. Methods: Twenty‐four experienced cytologists were recruited to examine 20 SurePath® cervical cytology slides by virtual microscopy. Diagnostic accuracy was compared with glass slide microscopy using an unbiased crossover experimental design. Responses were allocated a score of one for a correct identification of normal or abnormal (borderline/atypical changes in squamous or glandular cells or worse) and a score of zero for an incorrect response (a normal slide reported as abnormal or vice versa). Perceptions of virtual microscopy were assessed by questionnaire analysis. Results: Participants yielded a total of 285 responses for the virtual slide set and 300 for the glass slide set. The approximate time to screen a virtual slide was 18 minutes, compared with 8 minutes or less for a glass slide. Overall there was no significant difference between virtual microscopy and glass slide microscopy in terms of diagnostic accuracy (P = 0.22). Virtual microscopy under‐performed when images were captured over a narrow focal range (P = 0.01). Diagnostic accuracy of virtual microscopy equalled that of glass slide microscopy when participants were able to focus through the full thickness of the slide images (P = 0.07). The most common difficulties experienced by participants with virtual microscopy were freezing of the computer screen during image download, slow response of the computer during slide movement and, in some instances, ‘fuzzy’ images. Cytologists have a strong preference for glass slides over virtual microscopy despite the overall equal diagnostic performance of the two viewing modalities. Conclusions: Diagnostic accuracy of virtual microscopy can equal that of glass slide microscopy. However, without good computer network connections, wide focal range and software that permits effortless navigation across virtual slides, cytologists are unlikely to be convinced of the utility of this technology for cytology screening and diagnosis.     

Scanning electron microscopy of scales from different body regions of three lizard species

The Oberhautchen of scales from the dorsal, parietal, and ventral regions of Sceloporus occidentalis (Iguanidae), Gerrhonotus multicarinatus (Anguinidae), and Anniella pulchra (Anniellidae) were examined with a scanning electron microscope. At low magnification, all scales of S. occidentalis exhibit well-defined outlines of cells belonging to the Oberhautchen layer and the previously overlying clear layer. The dorsal and parietal cells of this species exhibit a minutely dentate Oberhautchen that forms tooth-like spinules 0.2 to 0.5 μ long and arranged in irregular rows. Minute pits 0.1 to 0.3 μ in diameter characterize the Oberhautchen of a ventral scale. Cell outlines are not evident on the scales of G. multicarinatus. The Oberhautchen of dorsal and parietal scales of this species is prominently laminated. Laminae are less prominent on scales of the lateral fold, and no intrinsic surface structure is evident on a ventral scale. In contrast, the fossorial anguinomorph Anniella pulchra exhibits Oberhautchen surfaces with practically no intrinsic microornamentation. However, what appear to be outlines of Oberhautchen cells are visible on the dorsal and ventral scales. These observations suggest that modifications of Oberhautchen microornamentation may have evolved to reduce friction with the substrate or other scales. The lack of pronounced microornamentation of the Oberhautchen on some body scales may indicate that a complex interdigitation between clear layer and Oberhautchen cells is not essential to the sloughing process.     

Microarray-based maps of copy-number variant regions in European and sub-Saharan populations

The genetic basis of phenotypic variation can be partially explained by the presence of copy-number variations (CNVs). Currently available methods for CNV assessment include high-density single-nucleotide polymorphism (SNP) microarrays that have become an indispensable tool in genome-wide association studies (GWAS). However, insufficient concordance rates between different CNV assessment methods call for cautious interpretation of results from CNV-based genetic association studies. Here we provide a cross-population, microarray-based map of copy-number variant regions (CNVRs) to enable reliable interpretation of CNV association findings. We used the Affymetrix Genome-Wide Human SNP Array 6.0 to scan the genomes of 1167 individuals from two ethnically distinct populations (Europe, N=717; Rwanda, N=450). Three different CNV-finding algorithms were tested and compared for sensitivity, specificity, and feasibility. Two algorithms were subsequently used to construct CNVR maps, which were also validated by processing subsamples with additional microarray platforms (Illumina 1M-Duo BeadChip, Nimblegen 385K aCGH array) and by comparing our data with publicly available information. Both algorithms detected a total of 42669 CNVs, 74% of which clustered in 385 CNVRs of a cross-population map. These CNVRs overlap with 862 annotated genes and account for approximately 3.3% of the haploid human genome.We created comprehensive cross-populational CNVR-maps. They represent an extendable framework that can leverage the detection of common CNVs and additionally assist in interpreting CNV-based association studies.     

Development of multigene expression signature maps at the protein level from digitized immunohistochemistry slides

Molecular classification of diseases based on multigene expression signatures is increasingly used for diagnosis, prognosis, and prediction of response to therapy. Immunohistochemistry (IHC) is an optimal method for validating expression signatures obtained using high-throughput genomics techniques since IHC allows a pathologist to examine gene expression at the protein level within the context of histologically interpretable tissue sections. Additionally, validated IHC assays may be readily implemented as clinical tests since IHC is performed on routinely processed clinical tissue samples. However, methods have not been available for automated n-gene expression profiling at the protein level using IHC data. We have developed methods to compute expression level maps (signature maps) of multiple genes from IHC data digitized on a commercial whole slide imaging system. Areas of cancer for these expression level maps are defined by a pathologist on adjacent, co-registered H&E slides, allowing assessment of IHC statistics and heterogeneity within the diseased tissue. This novel way of representing multiple IHC assays as signature maps will allow the development of n-gene expression profiling databases in three dimensions throughout virtual whole organ reconstructions.     

Virtual nanoscopy: Generation of ultra-large high resolution electron microscopy maps

A key obstacle in uncovering the orchestration between molecular and cellular events is the vastly different length scales on which they occur. We describe here a methodology for ultrastructurally mapping regions of cells and tissue as large as 1 mm(2) at nanometer resolution. Our approach employs standard transmission electron microscopy, rapid automated data collection, and stitching to create large virtual slides. It greatly facilitates correlative light-electron microscopy studies to relate structure and function and provides a genuine representation of ultrastructural events. The method is scalable as illustrated by slides up to 281 gigapixels in size. Here, we applied virtual nanoscopy in a correlative light-electron microscopy study to address the role of the endothelial glycocalyx in protein leakage over the glomerular filtration barrier, in an immunogold labeling study of internalization of oncolytic reovirus in human dendritic cells, in a cryo-electron microscopy study of intact vitrified mouse embryonic cells, and in an ultrastructural mapping of a complete zebrafish embryo slice.     

A supervised visual model for finding regions of interest in basal cell carcinoma images

Background

Expectation maximizing (EM) is one of the common approaches for image segmentation.

Methods

an improvement of the EM algorithm is proposed and its effectiveness for MRI brain image segmentation is investigated. In order to improve EM performance, the proposed algorithms incorporates neighbourhood information into the clustering process. At first, average image is obtained as neighbourhood information and then it is incorporated in clustering process. Also, as an option, user-interaction is used to improve segmentation results. Simulated and real MR volumes are used to compare the efficiency of the proposed improvement with the existing neighbourhood based extension for EM and FCM.

Results

the findings show that the proposed algorithm produces higher similarity index.

Conclusions

experiments demonstrate the effectiveness of the proposed algorithm in compare to other existing algorithms on various noise levels.     

Flexible fitting of atomic structures into electron microscopy maps using molecular dynamics

A novel method to flexibly fit atomic structures into electron microscopy (EM) maps using molecular dynamics simulations is presented. The simulations incorporate the EM data as an external potential added to the molecular dynamics force field, allowing all internal features present in the EM map to be used in the fitting process, while the model remains fully flexible and stereochemically correct. The molecular dynamics flexible fitting (MDFF) method is validated for available crystal structures of protein and RNA in different conformations; measures to assess and monitor the fitting process are introduced. The MDFF method is then used to obtain high-resolution structures of the E. coli ribosome in different functional states imaged by cryo-EM.     

Cryoelectron microscopy of low density lipoprotein in vitreous ice.

In this report, images of low density lipoprotein (LDL) in vitreous ice at approximately 30 A resolution are presented. These images show that LDL is a quasi-spherical particle, approximately 220-240 A in diameter, with a region of low density (lipid) surrounded by a ring (in projection) of high density believed to represent apolipoprotein B-100. This ring is seen to be composed of four or five (depending on view) large regions of high density material that may represent protein superdomains. Analysis of LDL images obtained at slightly higher magnification reveals that areas of somewhat lower density connect these regions, in some cases crossing the projectional interiors of the LDL particles. Preliminary image analysis of LDL covalently labeled at Cys3734 and Cys4190 with 1.4-nm Nanogold clusters demonstrates that this methodology will provide an important site-specific marker in studies designed to map the organization of apoB at the surface of LDL.     

The positive impact of team-based virtual microscopy on student learning in physiology and histology

Team-based virtual microscopy and on-line learning were used to transform the first-year Physiology/Histology course at The Johns Hopkins School of Medicine into a student-centered learning environment. Prior to each laboratory session, students were required to view prelaboratory virtual lectures and examine digital slides that had been enhanced with annotations and 2-min microlectures. The laboratory classroom was then used for team-based learning exercises including student presentations and small-group discussions designed to integrate histology and physiology. The results of quantitative assessments indicated an 8- to 14-point increase over the identical final exams given over the past 5 yr. Means (+/-SD) of percent correct answers on the final exam were found to be 75.2% (11.1%), 72.5% (12.6%), 70.5% (12.6%), 73.6% (11.3%), 73.1% (12.2%), and 84.1% (9.1%) for years 2001-2006, respectively. The mean test scores for all other years were statistically lower compared with 2006, as determined by the Bonferroni post hoc multiple-comparison test (P < 0.001 for all years).     

Comparative fine maps of bovine toll-like receptor 4 and toll-like receptor 2 regions

Abstract Toll-like receptors are cell-surface receptors that activate innate and adaptive immune responses. We have used a 5000-rad, whole-genome radiation hybrid panel to map Toll-like receptor 4 (TLR4) to the distal end of bovine Chromosome (Chr) 8, and Toll-like receptor 2 (TLR2) to the proximal end of bovine Chr 17. To facilitate comparative mapping and contig construction, we have also used 5000- and 12,000-rad, whole-genome radiation hybrid panels to produce fine maps of the regions surrounding these genes in cattle. These fine maps triple the number of available markers in the TLR4 region and more than double the number of available markers in the TLR2 region. Comparative analyses show gene order conservation between the bovine Chr 8 region and human Chr 9, and between the bovine Chr 17 region and human Chr 4. In addition, the bovine Chr 8 region refines an evolutionary chromosomal breakpoint from a 10-megabase region to a 2.5-megabase region, and the bovine Chr 17 map suggests a new evolutionary chromosomal breakpoint.