Biomedical subjects

P Gader

Publications and source records attributed to P Gader.

8 recordsLinked to original sources

PubMed1999

Automatic karyotyping of metaphase cells with overlapping chromosomes.

Image analysis research aimed at analyzing images of banded, human chromosomes is described. An experimental automated karyotyping system (AKS) has been assembled. The goal of the system is to label the chromosomes from a metaphase image with minimal human intervention. The AKS is unique in that it is designed to automatically process cells containing overlapping chromosomes. It also has four novel algorithmic features: the use of cross section sequence graphs, an over-segmentation-based segmentation strategy, the use of the 'pale path' concept for banded chromosomes and the application of mathematical programming for cell-level chromosome classification in cells with overlapping chromosomes. System performance was analyzed using a variety of metaphase images exhibiting various levels of overlapping chromosomes.

Algorithms↗

PubMed1998

Data-driven homologue matching for chromosome identification.

Karyotyping involves the visualization and classification of chromosomes into standard classes. In "normal" human metaphase spreads, chromosomes occur in homologous pairs for the autosomal classes 1-22, and X chromosome for females. Many existing approaches for performing automated human chromosome image analysis presuppose cell normalcy, containing 46 chromosomes within a metaphase spread with two chromosomes per class. This is an acceptable assumption for routine automated chromosome image analysis. However, many genetic abnormalities are directly linked to structural or numerical aberrations of chromosomes within the metaphase spread. Thus, two chromosomes per class cannot be assumed for anomaly analysis. This paper presents the development of image analysis techniques which are extendible to detecting numerical aberrations evolving from structural abnormalities. Specifically, an approach to identifying "normal" chromosomes from selected class(es) within a metaphase spread is presented. Chromosome assignment to a specific class is initially based on neural networks, followed by banding pattern and centromeric index criteria checking, and concluding with homologue matching. Experimental results are presented comparing neural networks as the sole classifier to our homologue matcher for identifying class 17 within normal and abnormal metaphase spreads.

Chromosome Banding↗

PubMed1997

Homologue matching using the Choquet integral.

Automated Giemsa-banded chromosome image research has been largely restricted to classification schemes associated with isolated chromosomes within metaphase spreads. In normal human metaphase spreads, there are 46 chromosomes occurring in homologous pairs for the autosomal classes, 1-22, and X chromosome for females. For optimizing automated human chromosome image analysis, many existing techniques assume cell normalcy. With many genetic abnormalities directly linked to structural and numerical aberrations of chromosomes within the metaphase spread, the two chromosome per class assumption may not be appropriate for anomaly analysis. At the University of Missouri, a data-driven homologue matching approach has been developed to identify all normal chromosomes within a metaphase spread from a selected class. Chromosome assignment to a specific class is initially based on neural networks, followed by banding pattern and centromeric index criteria checking, and concluding with homologue matching utilizing a density profile-based classifier, a shape profile-based classifier, and a binary band profile-based classifier. Based on preliminary results for the profile-based classifiers assigning chromosome 17, the Choquet integral is presented as an extension to the homologue matching approach. Experimental results are presented comparing the extended homologue matching approach to the transportation algorithm for identifying chromosome 21 within normal metaphase spreads.

Chromosome Aberrations↗

PubMed1997

The effects of image manipulation on automated karyotyping.

Karyotyping is one of the standard tools for human genetic investigations. Karyotyping involves the classification and interpretation of chromosomes found in a metaphase spread. As part of the automated karyotyping process, image manipulation is required for appropriately positioning metaphase spread chromosomes in their corresponding karyotype. Image manipulation, specifically image rotation, reorganizes the grey-level information within chromosomes to facilitate analysis. Statistical tests are performed to compare features related to banding pattern and length between unmanipulated chromosomes and corresponding rotated chromosomes. Based on experimental results, reorganizing the grey-level information yields statistically different chromosome features. Depending on the purpose of chromosome image analysis, the interpretation process of karyotyping could be impaired with chromosome feature distortion due to image rotation.

Humans↗

PubMed1996

A centromere attribute integration approach to centromere identification.

Automated and nonautomated approaches to chromosome classification involves assessing several chromosome attributes. The centromere is an important attribute which provides insight to other features such as chromosome orientation and the banding pattern sequence. Improving the ability to identify the centromere will enhance feature determination and analysis. Techniques to identify the centromere attempt to isolate specific centromere attributes. The centromere can be characterized as possessing the following properties: 1) usually the narrowest region in the chromosome image, 2) usually located in a region containing extreme concavities along the chromosome contour, and 3) usually located in a region of uniform dark grey-level. A centromere attribute integration approach for automated centromere identification has been developed which has a correct identification rate of 93.5% on a diversified data set. This approach determines and evaluates centromere candidates based on quantified centromere attributes. Centromere attribute integration incorporates other commonly used techniques for centromere identification. Some of the techniques integrated into the experimental algorithm include evaluating chromosome curvature, analyzing the shape profile, and inspecting the width profile.

Centromere↗

PubMed1996

Multiple neural network response variability as a predictor of neural network accuracy for chromosome recognition.

Human chromosome classification requires all chromosome appearing in a microphotograph of a dividing human cell to be classified within the known normal or abnormal 24 chromosome types. In recent years, research has focused on the use of neural networks for classification of normal chromosomes. Experimental work in this area led us to question whether learning variability, resulting when multiple neural networks are trained to solve the same problem, could be used as a predictor of classification performance. The Copenhagen chromosome data bank, consisting of 30-component feature vectors from 8106 chromosomes isolated from 180 cells, was divided into a training and a test subsets. Back propagation neural networks with 30 input nodes, 1 to 100 nodes in the hidden layer, and 24 output nodes were trained with the same learning parameters. After training, each neural network was tested. The neural network yielding the best classification was labeled as the optimal neural network. An error variability score was calculated for each test chromosome. This score was a function of all (100) neural network outputs obtained for that chromosome. The error variability scores ranged from 0.16 to 1.31 with a mean value of 0.41 and a SD of 0.12. There was significant difference (p < 0.0001) between the variability scores from chromosomes classified correctly (mean = 0.4, SD = 0.1, n = 3804) and incorrectly (mean = 0.62, SD = 0.19, n = 241) by the optimal neural network. When the variability score was used as a threshold to decide whether or not to accept the output of the optimal neural network, a peak classification rate of 98.93% was observed for chromosomes with an error variability score < 0.35. Results indicate that the error variability of multiple neural network responses can be used as a confidence indicator for a optimal neural network.

Humans↗

PubMed1996

Centromere attribute integration based chromosome polarity assignment.

Automated karyotyping involves evaluating quantified chromosome attributes for proper classification. Chromosome attributes derived from the banding pattern require the correct chromosome polarity for correct banding sequence interpretation. Chromosome polarity is defined in terms of determining the short and long arms of the chromosome using the centromere as the reference point for measuring the chromosome length on both sides of the centromere. In addition to banding sequence interpretation, polarity is used in the chromosome orientation for chromosome repositioning from the metaphase spread to the karyotype. Automated polarity determination is often not performed for classifying chromosomes in the metaphase spread image. Polarity may be determined user interactively, by the system, or not at all. In order to reduce the computational complexity of evaluating banding sequence features using both chromosome ends as reference points, there is a need to improve chromosome polarity determination in automated karyotyping. A centromere attribute integration approach has been developed at the University of Missouri-Columbia which performs correct chromosome polarity assessment at a rate comparable to other studies of 96.1% on a diversified data set.

Algorithms↗

PubMed1995

Automated chromosome classification limitations due to image processing.

Automated chromosome classification from metaphase spreads has been a difficult research problem over the past 30 years. Numerous techniques have been implemented to address the research problem. Image processing techniques support many methods utilized for automated and/or semiautomated karyotyping. Some current systems correctly classify individual chromosomes at high rates but lack the capability to properly classify chromosomes for entire cells consistently. Most systems attain high classification rates ignoring overlapping metaphase chromosomes for testing purposes. Chromosome classification depends on identifying features common to each chromosome class or number. Some of the common features incorporated into various chromosome classification systems include length, centromere location, banding pattern, and width. To complicate classification, chromosome features tend to vary not only between people but from cell to cell for the same person. Additionally, chromosomes found in metaphase spreads may have any orientation and virtually any degree of overlap with other chromosomes present. Besides the inherent barriers impeding chromosome classification, many systems perform image processing operations to the chromosomes for feature determination. Image processing operations, such as image rotation, that manipulate the chromosome grey-level information may distort the feature calculations utilized in karyotyping. Consequently, feature stability becomes an important issue for improving karyotyping capability.

Chromosomes, Human↗