An improved sparse representation model with structural information for Multicolour Fluorescence <Emphasis Type="Italic">In-Situ</Emphasis> Hybridization (M-FISH) image classification

Background

Multicolour Fluorescence In-Situ Hybridization (M-FISH) images are employed for detecting chromosomal abnormalities such as chromosomal translocations, deletions, duplication and inversions. This technique uses mixed colours of fluorochromes to paint the whole chromosomes for rapid detection of chromosome rearrangements. The M-FISH data sets used in our research are obtained from microscopic scanning of a metaphase cell labelled with five different fluorochromes and a DAPI staining. The reliability of the technique lies in accurate classification of chromosomes (24 classes for male and 23 classes for female) from M-FISH images. However, due to imaging noise, mis-alignment between multiple channels and many other imaging problems, there is always a classification error, leading to wrong detection of chromosomal abnormalities. Therefore, how to accurately classify different types of chromosomes from M-FISH images becomes a challenging problem.

Methods

This paper presents a novel sparse representation model considering structural information for the classification of M-FISH images. In our previous work a sparse representation based classification model was proposed. This model employed only individual pixel information for the classification. With the structural information of neighbouring pixels as well as the information of themselves simultaneously, the novel approach extended the previous one to the regional case. Based on Orthogonal Matching Pursuit (OMP), we developed simultaneous OMP algorithm (SOMP) to derive an efficient solution of the improved sparse representation model by incorporating the structural information.

Results

The p-value of two models shows that the newly proposed model incorporating the structural information is significantly superior to our previous one. In addition, we evaluated the effect of several parameters, such as sparsity level, neighbourhood size, and training sample size, on the of the classification accuracy.

Conclusions

The comparison with our previously used sparse model demonstrates that the improved sparse representation model is more effective than the previous one on the classification of the chromosome abnormalities.

     

Karyotyping of comparative genomic hybridization human metaphases by using support vector machines.

BACKGROUND: Comparative genomic hybridization (CGH) is a relatively new molecular cytogenetic method for detecting chromosomal imbalance. Karyotyping of human metaphases is an important step to assign each chromosome to one of 23 or 24 classes (22 autosomes and two sex chromosomes). Automatic karyotyping in CGH analysis is needed. However, conventional karyotyping approaches based on DAPI images require complex image enhancement procedures. METHODS: This paper proposes a simple feature extraction method, one that generates density profiles from original true color CGH images and uses normalized profiles as feature vectors without quantization. A classifier is developed by using support vector machine (SVM). It has good generalization ability and needs only limited training samples. RESULTS: Experiment results show that the feature extraction method of using color information in CGH images can improve greatly the classification success rate. The SVM classifier is able to acquire knowledge about human chromosomes from relatively few samples and has good generalization ability. A success rate of moe than 90% has been achieved and the time for training and testing is very short. CONCLUSIONS: The feature extraction method proposed here and the SVM-based classifier offer a promising computerized intelligent system for automatic karyotyping of CGH human chromosomes.     

多色-FISH技术的进展及其在分子生物医学上的应用

传统显带分析技术以每条染色体独特的显带带型为依据,提供染色体形态结构的基本信息,用于染色体核型的初步分析。然而有些染色体重排由于涉及的片断太小或具有相似的带型,用该方法难以探测或准确描绘。多元荧光原位杂交(M-FISH),光谱核型分析(SKY),FISH-显带分析技术是染色体特异的多色荧光原位杂交技术(mFISH)。它们能够探测出传统显带分析不能发现的染色体异常,提供更准确的核型。M-FISH和SKY均以组合标记的染色体涂染探针共杂交为基础,二者的不同在于观察仪器和分析方法上。它们可对中期染色体涂片进行快速准确分析,描绘复杂核型,确认标记染色体,主要用于恶性疾病的细胞遗传学诊断分析。FISH-显带分析技术以FISH技术为基础,能同时检测多条比染色体臂短的染色体亚区域。符合该定义的FISH-显带分析技术各有特点,其共同特点是都能产生DNA特异的染色体条带。这些条带有更多色彩,能提供更多信息。FISH-显带分析技术已经成功地被用于进化生物学、放射生物学以及核结构的研究,同时也被用于产前、产后以及肿瘤细胞遗传学诊断,是很有潜力的工具。     

Screening for subtelomeric rearrangements using genetic markers in 70 patients with unexplained mental retardation

Cryptic unbalanced rearrangements involving chromosome ends are a significant cause of idiopathic mental retardation. The most frequently used technique to screen for these subtle rearrangements is Multiprobe fluorescence in situ hybridization (FISH). As this is a labor-intensive technique, we used microsatellite genotyping to detect possible subtelomeric rearrangements in a study population. Out of the 70 patients we screened, three chromosomal rearrangements were detected: a deletion of marker D2S2986, a deletion of marker D7S594 and a deletion of marker D19S424. However, none of these aberrations appeared to be disease causing.     

Multiplex-fluorescence in situ hybridization for chromosome karyotyping

Multiplex-fluorescence in situ hybridization (M-FISH) was initially developed to stain human chromosomes--the 22 autosomes and X and Y sex chromosomes--with uniquely distinctive colors to facilitate karyotyping. The characteristic spectral signatures of all different combinations of fluorochromes are determined by multichannel image-analysis methods. Advantages of M-FISH include rapid analysis of metaphase spreads, even in complex cases with multiple chromosomal rearrangements, and identification of marker chromosomes. The M-FISH technology has been extended to other species, such as the mouse. Furthermore, in addition to painting probes, the method has been used with a variety of region-specific probes. M-FISH has even recently been used for 3D studies to analyze the distribution of human chromosomes in intact and preserved interphase nuclei. Hence, M-FISH has evolved into an essential tool for both clinical diagnostics and basic research. In this protocol, we describe how to use M-FISH to karyotype chromosomes, a procedure that takes approximately 14 d if new M-FISH probes have to be generated and 3 d if the M-FISH probes are ready to use.     

Multicolor banding technique, spectral color banding (SCAN): new development and applications

Conventional banding techniques can characterize chromosomal aberrations associated with tumors and congenital diseases with considerable precision. However, chromosomal aberrations that have been overlooked or are difficult to analyze even by skilled cytogeneticists were also often noted. Following the introduction of multicolor karyotyping such as spectral karyotyping (SKY) and multiplex-fluorescence in situ hybridization (M-FISH), it is possible to identify this kind of cryptic or complex aberration comprehensively by a single analysis. To date, multicolor karyotyping techniques have been established as useful tools for cytogenetic analysis. However, since this technique depends on whole chromosome painting probes, it involves limitations in that the origin of aberrant segments can be identified only in units of chromosomes. To overcome these limitations, we have recently developed spectral color banding (SCAN) as a new multicolor banding technique based on the SKY methodology. This new technique may be deemed as an ideal chromosome banding technique since it allows representation of a multicolor banding pattern matching the corresponding G-banding pattern. We applied this technique to the analysis of chromosomal aberrations in tumors that had not been fully characterized by G-banding or SKY and found it capable of (1) detecting intrachromosomal aberrations; (2) identifying the origin of aberrant segments in units of bands; and (3) precisely determining the breakpoints of complex rearrangements. We also demonstrated that SCAN is expected to allow cytogenetic analysis with a constant adequate resolution close to the 400-band level regardless of the degree of chromosome condensation. As compared to the conventional SKY analysis, SCAN has remarkably higher accuracy for a particular chromosome, allowing analysis in units of bands instead of in units of chromosomes and is hence promising as a means of cytogenetic analysis.     

Rapid pedobarographic image registration based on contour curvature and optimization

Image registration, the process of optimally aligning homologous structures in multiple images, has recently been demonstrated to support automated pixel-level analysis of pedobarographic images and, subsequently, to extract unique and biomechanically relevant information from plantar pressure data. Recent registration methods have focused on robustness, with slow but globally powerful algorithms. In this paper, we present an alternative registration approach that affords both speed and accuracy, with the goal of making pedobarographic image registration more practical for near-real-time laboratory and clinical applications. The current algorithm first extracts centroid-based curvature trajectories from pressure image contours, and then optimally matches these curvature profiles using optimization based on dynamic programming. Special cases of disconnected images (that occur in high-arched subjects, for example) are dealt with by introducing an artificial spatially linear bridge between adjacent image clusters. Two registration algorithms were developed: a ‘geometric’ algorithm, which exclusively matched geometry, and a ‘hybrid’ algorithm, which performed subsequent pseudo-optimization. After testing the two algorithms on 30 control image pairs considered in a previous study, we found that, when compared with previously published results, the hybrid algorithm improved overlap ratio (p=0.010), but both current algorithms had slightly higher mean-squared error, assumedly because they did not consider pixel intensity. Nonetheless, both algorithms greatly improved the computational efficiency (25±8 and 53±9 ms per image pair for geometric and hybrid registrations, respectively). These results imply that registration-based pixel-level pressure image analyses can, eventually, be implemented for practical clinical purposes.     

Molecular cytogenetic techniques in detecting subtle chromosomal imbalances

Diagnostic possibilities of CGH and M-FISH techniques for detection of submicroscopic chromosomal imbalancies were compared on the basis of two cases of t(X;Y) and one case of marker chromosome. In cases with t(X;Y), the sequences specific for chromosome Y were detected by PCR and CGH, but the localisation of these sequences on the short arm of chromosome X was confirmed by the FISH technique, employing two Yp-specific probes for SRY and TSPY genes. Significant differences between above cases were revealed in the size of Yp chromosome fragments translocated on chromosome X. An extra material of chromosome marker could not be identified by classical banding and FISH techniques and it was only CGH and M-FISH techniques that enabled detecting the chromosomal origin of the marker. The applied CGH technique enabled finding subtle chromosomal imbalancies in the presented cases with a resolution of approximately 3 Mbp.     

Automatic signal classification in fluorescence in situ hybridization images

BACKGROUND: Previous systems for dot (signal) counting in fluorescence in situ hybridization (FISH) images have relied on an auto-focusing method for obtaining a clearly defined image. Because signals are distributed in three dimensions within the nucleus and artifacts such as debris and background fluorescence can attract the focusing method, valid signals can be left unfocused or unseen. This leads to dot counting errors, which increase with the number of probes. METHODS: The approach described here dispenses with auto-focusing, and instead relies on a neural network (NN) classifier that discriminates between in and out-of-focus images taken at different focal planes of the same field of view. Discrimination is performed by the NN, which classifies signals of each image as valid data or artifacts (due to out of focusing). The image that contains no artifacts is the in-focus image selected for dot count proportion estimation. RESULTS: Using an NN classifier and a set of features to represent signals improves upon previous discrimination schemes that are based on nonadaptable decision boundaries and single-feature signal representation. Moreover, the classifier is not limited by the number of probes. Three classification strategies, two of them hierarchical, have been examined and found to achieve each between 83% and 87% accuracy on unseen data. Screening, while performing dot counting, of in and out-of-focus images based on signal classification suggests an accurate and efficient alternative to that obtained using an auto-focusing mechanism.     

Stylized chromosome images

Stylized chromosome images 1) serve as a format to test effects of preprocessing algorithms used in automated karyotyping; 2) enhance the ability of humans to perform quantitative analysis of chromosomal aberrations; 3) provide an alternative format for karyotype hard copies produced by automated systems. Stylized chromosomes are two-dimensional computer-generated images based on information extracted from one-dimensional width and density profiles. These profiles correspond to what cytogeneticists observe through the microscope as the shape and banding patterns of stained chromosomes. Stylized presentation sharpens chromosome band boundaries and perimeters, reduces "noise," and enhances gray level variations, which are difficult to distinguish by humans on photographic or computer generated karyotypes. Karyotyping accuracy using stylized images was used to detect difficult areas for automated chromosome identification. Landmark bands sufficient to classify chromosomes were identified; shapes of chromosomes reflected in width profiles were said to aid classification. A two-step automated karyotyping strategy proposed is: 1) classify chromosomes by landmarks, minimum information needed for identification; 2) subsequently employ the full banding pattern with maximum resolution to detect aberrations. Stylized images of abnormal chromosomes have potential for testing hypothesis regarding breakpoints and quantitative analysis, but improvements are needed in homologue normalization and definition of termini of chromosomes.     

New concepts to improve resolution and sensitivity of molecular cytogenetic diagnostics by multicolor fluorescence in situ hybridization

BACKGROUND: Routine application of multicolor fluorescence in situ hybridization (M-FISH) technology for molecular cytogenetic diagnostics has been hampered by several technical limitations. First, when using chromosome-specific painting probes, there is a limit in cytogenetic resolution of approximately 2-3 Mb, which can mask hidden structural abnormalities that have a significant clinical effect. Second, using whole chromosome painting probes, intrachromosomal rearrangements cannot be detected and the exact localization of breakpoints is often not possible. METHODS: We suggest the use of multiplex-labeled region or locus- specific probes in combination with an optimal probe design to improve the sensitivity and resolution of the M-FISH technology. To allow the application of this assay in routine diagnostics, we developed a multipurpose image analysis system. RESULTS: goldFISH was applied to the study of cryptic translocations in mental retardation patients and to the study of high-resolution breakpoint mapping in non-small cell lung cancer patients. For an individual with mental retardation, who had an apparently normal karyotype by G-banding, we detected an unbalanced translocation involving chromosomes 2 and 7. CONCLUSIONS: In combination with optimally designed probe kits, goldFISH overcomes most of the present limitations of the M-FISH technology and results in virtually 100% reliability for detecting interchromosomal and intrachromosomal rearrangements.     

A Hybrid Color Space for Skin Detection Using Genetic Algorithm Heuristic Search and Principal Component Analysis Technique

Color is one of the most prominent features of an image and used in many skin and face detection applications. Color space transformation is widely used by researchers to improve face and skin detection performance. Despite the substantial research efforts in this area, choosing a proper color space in terms of skin and face classification performance which can address issues like illumination variations, various camera characteristics and diversity in skin color tones has remained an open issue. This research proposes a new three-dimensional hybrid color space termed SKN by employing the Genetic Algorithm heuristic and Principal Component Analysis to find the optimal representation of human skin color in over seventeen existing color spaces. Genetic Algorithm heuristic is used to find the optimal color component combination setup in terms of skin detection accuracy while the Principal Component Analysis projects the optimal Genetic Algorithm solution to a less complex dimension. Pixel wise skin detection was used to evaluate the performance of the proposed color space. We have employed four classifiers including Random Forest, Naïve Bayes, Support Vector Machine and Multilayer Perceptron in order to generate the human skin color predictive model. The proposed color space was compared to some existing color spaces and shows superior results in terms of pixel-wise skin detection accuracy. Experimental results show that by using Random Forest classifier, the proposed SKN color space obtained an average F-score and True Positive Rate of 0.953 and False Positive Rate of 0.0482 which outperformed the existing color spaces in terms of pixel wise skin detection accuracy. The results also indicate that among the classifiers used in this study, Random Forest is the most suitable classifier for pixel wise skin detection applications.     

Neural network-based segmentation and classification system for automated grading of histologic sections of bladder carcinoma

OBJECTIVE: To develop an image analysis system for automated nuclear segmentation and classification of histologic bladder sections employing quantitative nuclear features. STUDY DESIGN: Ninety-two cases were classified into three classes by experienced pathologists according to the WHO grading system: 18 cases as grade 1, 45 as grade 2, and 29 as grade 3. Nuclear segmentation was performed by means of an artificial neural network (ANN)-based pixel classification algorithm, and each case was represented by 36 nuclei features. Automated grading of bladder tumor histologic sections was performed by an ANN classifier implemented in a two-stage hierarchic tree. RESULTS: On average, 95% of the nuclei were correctly detected. At the first stage of the hierarchic tree, classifier performance in discriminating between cases of grade 1 and 2 and cases of grade 3 was 89%. At the second stage, 79% of grade 1 cases were correctly distinguished from grade 2 cases. CONCLUSION: The proposed image analysis system provides the means to reduce subjectivity in grading bladder tumors and may contribute to more accurate diagnosis and prognosis since it relies on nuclear features, the value of which has been confirmed.     

Classification accuracy in multiple color fluorescence imaging microscopy

BACKGROUND: The discriminatory power and imaging efficiency of different multicolor FISH (M-FISH) analysis systems are key factors in obtaining accurate and reproducible classification results. In a recent paper, Garini et al. put forth an analytical technique to quantify the discriminatory power ("S/N ratio") and imaging efficiency ('excitation efficiency') of multicolor fluorescent karyotyping systems. METHODS: A parametric model of multicolor fluorescence microscopy, based on the Beer-Lambert law, is analyzed and reduced to a simple expression for S/N ratio. Parameters for individual system configurations are then plugged into the model for comparison purposes. RESULTS: We found that several invalid assumptions, which are used to reduce the complex mathematics of the Beer-Lambert law to a simple S/N ratio, result in some completely misleading conclusions about classification accuracy. The authors omit the most significant noise source, and consider only one highly abstract and unrepresentative situation. Unwisely chosen parameters used in the examples lead to predictions that are not consistent with actual results. CONCLUSIONS: The earlier paper presents an inaccurate view of the M-FISH situation. In this short communication, we point out several inaccurate assumptions in the mathematical development of Garini et al. and the poor choices of parameters in their examples. We show results obtained with different imaging systems that indicate that reliable and comparable results are obtained if the metaphase samples are well-hybridized. We also conclude that so-called biochemical noise, not photon noise, is the primary factor that limits pixel classification accuracy, given reasonable exposure times. Copyright Wiley-Liss, Inc.     

Comparative M-FISH and CGH analyses in sensitive and drug-resistant human T-cell acute leukemia cell lines

Cell lines of human T-cell acute lymphoblastic leukemias (T-ALL) have gained high interest for study of mechanisms of cytostatic drug resistance. However, they should also be suited to examine the validity and reliability of molecular cytogenetic techniques in detecting genomic alterations in neoplastic cells. Therefore, comparative genomic hybridization (CGH) and 24-color-fluorescence-in-situ-hybridization (M-FISH) were applied to eight sublines of CCRF-CEM leukemia cells selected in vitro for drug resistance and to their drug-sensitive parental counterparts. All cell lines were characterized by altered chromosome numbers and by a variety of chromosomal structural aberrations as shown by M-FISH. The great majority of anomalies detected by this technique were confirmed by CGH. Interestingly, a considerable number of the rearrangements found were imbalanced. Amplifications of 5q13 in the six methotrexate-resistant cell lines, a del(9)(p21pter) in all lines examined, and a gain of chromosome 20 in 9 of the 10 lines examined were readily detected by both techniques. The same held true for losses of chromosomes 17 and 18 in the near tetraploid cell lines which could also be confirmed by CGH. Some imbalances of genomic material detected by CGH were, however, not observed by means of M-FISH, possibly due to the limited extension of the corresponding chromosomal segment involved or the small subpopulation of cells affected. On the other hand, reciprocal translocations, balanced isochromosomes, and small deletions remained mainly undetected by CGH. A comparison of chromosomal alterations in drug-resistant and parental cell lines showed not only amplifications of chromosomal segments harboring well-known drug resistance genes, e.g., the dihydrofolate reductase gene, but also chromosomal changes which may involve novel genes associated with drug resistance. Thus, the present study has clearly unveiled the strengths and weaknesses of both techniques which can excellently complement each other. Their combination allowed a distinct improvement of the definition of the complex karyotypes of drug-resistant cell lines.     

Theory and implementation of an electronic, automated measurement system for images obtained from immunohistochemically stained slides

OBJECTIVE: To develop and implement an Internet-based, automated image measurement system for immunohistochemically stained slides including fluorescence images in online and off-line modes. STUDY DESIGN: An image analyzing system was developed that automatically measures digitized images obtained from immunohistochemically stained slides. It is divided into a common server platform and a specific image quantification system based upon DIAS (University of Jena). After registration, the user fills in an input data form and attaches images to be measured. The server periodically transfers the data to the measurement system. The measurement works on dynamic thresholding and active sampling of objects visualized by fluorescence and conventional chromogens. It includes stereologic algorithms, object quantification, syntactic structure analysis and quality assurance. RESULTS: The system has been tested for diaminobenzidene, alkaline phosphatase and fluorescence images (FITC, etc.). The reproducibility and stability of the system are > 98%. The series of successfully measured images comprises > 1,000 images in total in the online and off-line modes. CONCLUSION: An Internet-based automated image measurement system has been developed that offers worldwide access to the major requests for quantification of immunohistochemically stained slides-tissue array analysis, nuclear stains (MIB, hormones), membrane stains (CerbB2), vascularization and fluorescence in situ hybridization.     

Karyotyping of comparative genomic hybridization human metaphases using kernel nearest-neighbor algorithm

BACKGROUND: Comparative genomic hybridization (CGH) is a relatively new molecular cytogenetic method that detects chromosomal imbalances. Automatic karyotyping is an important step in CGH analysis because the precise position of the chromosome abnormality must be located and manual karyotyping is tedious and time-consuming. In the past, computer-aided karyotyping was done by using the 4',6-diamidino-2-phenylindole, dihydrochloride (DAPI)-inverse images, which required complex image enhancement procedures. METHODS: An innovative method, kernel nearest-neighbor (K-NN) algorithm, is proposed to accomplish automatic karyotyping. The algorithm is an application of the "kernel approach," which offers an alternative solution to linear learning machines by mapping data into a high dimensional feature space. By implicitly calculating Euclidean or Mahalanobis distance in a high dimensional image feature space, two kinds of K-NN algorithms are obtained. New feature extraction methods concerning multicolor information in CGH images are used for the first time. RESULTS: Experiment results show that the feature extraction method of using multicolor information in CGH images improves greatly the classification success rate. A high success rate of about 91.5% has been achieved, which shows that the K-NN classifier efficiently accomplishes automatic chromosome classification from relatively few samples. CONCLUSIONS: The feature extraction method proposed here and K-NN classifiers offer a promising computerized intelligent system for automatic karyotyping of CGH human chromosomes.     

Development of a multistage classifier for a monitoring system of cell activity based on imaging of chromosomal dynamics.

BACKGROUND: Cell-based assays utilizing digital image cytometry yield multivariate sets of information measuring the efficacy of medicines/chemicals. The use of a HeLa cell line that expresses a GFP-Histone-H1 fusion protein further enhances the performance of these systems, avoiding the use of dyes that may have detrimental influence on cells. Aside from the mitotic index, the distribution of the cell-cycle phases during mitosis can be used as measures of drug/treatment efficacy. Quantification of these parameters, however, requires skill and is time consuming. The purpose of this research was therefore to create a classifier to be incorporated into a system that can automatically identify the cell-cycle phases in a given image. METHODS: Features based on the shape and texture of the chromosomal regions in images of live HeLa cells were measured and analyzed. Linear discriminant functions were calculated for the eight cell-cycle phases: interphase, prophase, prometaphase, metaphase, early anaphase, anaphase, telophase and cytokinesis. RESULTS: The multistage linear discriminant classifier developed had an average classification efficiency of 87.30%. CONCLUSION: We demonstrated the possibility of creating a classifier to discriminate between cell-cycle phases using shape and texture features of chromosomal regions. The classifier can be fused to an algorithm for image segmentation, forming a system to automatically and rapidly measure the aforementioned parameters. The results can then be collated to constitute an assay assessing the effects of a drug or treatment on mammalian cells.     

Correlation analysis of two-dimensional gel electrophoretic protein patterns and biological variables

Background  

Two-dimensional gel electrophoresis (2DE) is a powerful technique to examine post-translational modifications of complexly modulated proteins. Currently, spot detection is a necessary step to assess relations between spots and biological variables. This often proves time consuming and difficult when working with non-perfect gels. We developed an analysis technique to measure correlation between 2DE images and biological variables on a pixel by pixel basis. After image alignment and normalization, the biological parameters and pixel values are replaced by their specific rank. These rank adjusted images and parameters are then put into a standard linear Pearson correlation and further tested for significance and variance.