A processing strategy for automated Papanicolaou smear screening.

A multilayer processing strategy was developed for the automatic screening of conventionally prepared Papanicolaou smears. The processing stages include image segmentation, feature extraction, object classification and slide classification. Mathematical morphology functions were implemented in hardware with custom-built gate array processors for image segmentation. There were 68 features used for classifier training. In object classification we combined the evidential supports of a binary decision tree classifier and a multilayer perceptron classifier to achieve an integrated decision. In this feasibility study, 449 conventionally prepared cervical Papanicolaou smears were tested in a prototype research system between January and May 1991. The 95% confidence interval for the slide false-negative rate was 1-9%, and the 95% confidence interval for the slide sort rate was 45-55%. The estimated sort rate for clearly normal slides is within the range required for a cost-efficient screening system, and the estimated false-negative rate for premalignant and malignant smears is an improvement over published false-negative rates for human performance. Several performance improvement efforts are still under way. We expect that they will result in a vastly reduced slide false-negative rate.     

An automated rotisserie system for processing Western blots.

An apparatus to automate completely the processing of Western blots is described. The prototype is based on a popular rotisserie system design. The incubation chamber consists of an inner cylinder that rotates inside an outer cylinder (incubation chamber). The blot is contained in the inner cylinder. Two magnets are mounted at one end of the inner cylinder, and rotation of the inner cylinder is effected by two magnets mounted on a motor drive outside the incubation chamber. Movement of chemicals into and out of the incubation chamber is driven pneumatically, and the entire process is controlled by a computer. Processing a blot for chemiluminescent detection takes 7 h to complete without human intervention. The quality of the resulting image is comparable to or better than a blot using manual processing. In addition, the prototype is capable of re-collecting all three antisera for future use.     

The automated classification of mitotic phase for human chromosome spreads.

A system has been developed that automatically recognizes the mitotic phase of human chromosome spreads for karyotyping. Suitable spreads are classified into one of five subphases of mitosis. Classification is performed on the basis of summed chromosome length and most probable chromosome width. Classification requires 100-500 msec. A television camera scans the spread through microscope optics; computer and special purpose electronics process the video signals to generate run length histograms. The histograms are used to determine mitotic phase. Unbanded spreads, 133, were classified with a 4.5% error rate. One hundred banded spreads were classified with a 15% error rate.     

Image storage for automated crystallization imaging systems

To use crystallography for the determination of the three-dimensional structures of proteins, protein crystals need to be grown. Automated imaging systems are increasingly being used to monitor these crystallization experiments. These present problems of accessibility to the data, repeatability of any image analysis performed and the amount of storage required. Various image formats and techniques can be combined to provide effective solutions to high volume processing problems such as these, however lack of widespread support for the most effective algorithms, such as JPeg2000 which yielded a 64% improvement in file size over the bitmap, currently inhibits the immediate take up of this approach.     

Comparison of three different methods for automated classification of cervical cells.

Over 4600 exfoliated squamous cervical cells taken from appropriate Papanicolaou samples were classified as normal, mildly dysplastic, moderately dysplastic and severely dysplastic by an experienced cytopathologist. The slides were de-stained and subsequently re-stained with Feulgen Thionin-SO2 stain. Images of the nuclei were then captured, recorded and processed employing an image cytometry device. Automated classification of the cells was carried out using three different methods--discriminant function analysis, a decision tree classifier and a neutral network classifier. The discriminant function analysis method, which combined all dysplastic cells into an abnormal group, achieved a combined error rate of less than 0.4% for moderate and severe dysplastic cells, and less than 40% for mildly dysplastic cells. All three methods yielded comparable results, which approached those of human performance.     

An automated rotisserie system for processing Western blots

An apparatus to automate completely the processing of Western blots is described. The prototype is based on a popular rotisserie system design. The incubation chamber consists of an inner cylinder that rotates inside an outer cylinder (incubation chamber). The blot is contained in the inner cylinder. Two magnets are mounted at one end of the inner cylinder, and rotation of the inner cylinder is effected by two magnets mounted on a motor drive outside the incubation chamber. Movement of chemicals into and out of the incubation chamber is driven pneumatically, and the entire process is controlled by a computer. Processing a blot for chemiluminescent detection takes 7 h to complete without human intervention. The quality of the resulting image is comparable to or better than a blot using manual processing. In addition, the prototype is capable of re-collecting all three antisera for future use.     

Decision tree based information integration for automated protein classification

We propose a novel technique for automatically generating the SCOP classification of a protein structure with high accuracy. We achieve accurate classification by combining the decisions of multiple methods using the consensus of a committee (or an ensemble) classifier. Our technique, based on decision trees, is rooted in machine learning which shows that by judicially employing component classifiers, an ensemble classifier can be constructed to outperform its components. We use two sequence- and three structure-comparison tools as component classifiers. Given a protein structure and using the joint hypothesis, we first determine if the protein belongs to an existing category (family, superfamily, fold) in the SCOP hierarchy. For the proteins that are predicted as members of the existing categories, we compute their family-, superfamily-, and fold-level classifications using the consensus classifier. We show that we can significantly improve the classification accuracy compared to the individual component classifiers. In particular, we achieve error rates that are 3-12 times less than the individual classifiers' error rates at the family level, 1.5-4.5 times less at the superfamily level, and 1.1-2.4 times less at the fold level.     

An integrated automated system for crackles extraction and classification

This paper presents an integrated automated system for crackles recognition. This system comprises three serial modules with following functions: (1) separation of crackles from vesicular sounds using a wavelet packet filter (WPST–NST); (2) detection of crackles by fractal dimension (FD); (3) classification of crackles based on Gaussian mixture models (GMM). The WPST–NST filter incorporates a multi-resolution decomposition of the original respiratory signal and an entropy-based best basis selection of the coefficients. Two thresholds are defined, in time and frequency domains respectively, to separate the crackles from the respiratory sounds. Then, a denoising filter is applied to the discontinuous output of WPST–NST and a crackle-peak-detector (CPD) localizes the individual crackles by means of their fractal dimension. After that, three feature parameters, including the Gaussian bandwidth (GBW), the peak frequency (PF) and the maximal deflection width (MDW), of the crackles are extracted. Finally, crackles are classified into fine crackles (FC) and coarse crackles (CC) using Gaussian mixture models.     

Spectral bandwidth in automated leukocyte classification

Investigators have repeatedly pointed out the importance of spectral information in the automated classification of white blood cells. In general, monochromatic images recorded through two or three color filters are used to extract this information. Although it has generally been thought that the use of narrow band filters provides "cleaner" color information than is obtainable through wide band filters, the choice has not been fully investigated and the question is far from being settled. The use of wide band filters has the clear practical advantage of increased light levels at the detector, resulting in higher signal-to-noise ratio with less demand on light source design. In order to investigate this issue, a series of 681 leukocytes of the most frequently occurring types were digitized by the use of both narrow (10 nm) and wide (90 nm) band filters. Parameters were extracted independently from both sets of images. These parameters were then used to develop a classifier for each set of images. The choice of features and classifier results indicate that there are no major performance differences between the two types of filters.     

Proteomic applications of automated GPCR classification

The G-protein coupled receptor (GPCR) superfamily fulfils various metabolic functions and interacts with a diverse range of ligands. There is a lack of sequence similarity between the six classes that comprise the GPCR superfamily. Moreover, most novel GPCRs found have low sequence similarity to other family members which makes it difficult to infer properties from related receptors. Many different approaches have been taken towards developing efficient and accurate methods for GPCR classification, ranging from motif-based systems to machine learning as well as a variety of alignment-free techniques based on the physiochemical properties of their amino acid sequences. This review describes the inherent difficulties in developing a GPCR classification algorithm and includes techniques previously employed in this area.     

Fast automated cell phenotype image classification

Background  

The genomic revolution has led to rapid growth in sequencing of genes and proteins, and attention is now turning to the function of the encoded proteins. In this respect, microscope imaging of a protein's sub-cellular localisation is proving invaluable, and recent advances in automated fluorescent microscopy allow protein localisations to be imaged in high throughput. Hence there is a need for large scale automated computational techniques to efficiently quantify, distinguish and classify sub-cellular images. While image statistics have proved highly successful in distinguishing localisation, commonly used measures suffer from being relatively slow to compute, and often require cells to be individually selected from experimental images, thus limiting both throughput and the range of potential applications. Here we introduce threshold adjacency statistics, the essence which is to threshold the image and to count the number of above threshold pixels with a given number of above threshold pixels adjacent. These novel measures are shown to distinguish and classify images of distinct sub-cellular localization with high speed and accuracy without image cropping.     

Artificial intelligence in automated classification of rat vaginal smear cells.

Microscopic examination of vaginal smears has been used routinely to determine the stage of the estrous cycle of female rats in reproductive research. The stage of the estrous cycle is based on relative counts of nucleated epithelial cells, cornified epithelial cells and leukocytes. The purpose of this project was to explore automation of vaginal smear analysis using image processing and artificial intelligence techniques. A fully connected back-propagation neural network was used to locate all potential objects in a digitized scene. A unique algorithm was then employed to center a subsequent sampling box to collect pixel intensity values from the red and green components of each image. A final neural network was used in the classification of cell type. Neural networks were used because of their ability to generalize among input patterns and to tolerate extraneous noise due to variations in staining artifacts and aberrant illumination of the microscope field. This preliminary cell diagnosing system not only provides the basis for the fully automated system but also provides a method by which many other cytologic image processing problems can be automated.     

Image cytometry in automated cervical screening

Severe restrictions with regard to false negative rates have played a major role in the development of the LEYden Television Analysis System (LEYTAS). The present paper describes a test with a continuous series of 1500 cervical samples illustrating the accuracy of LEYTAS in a fully automated screening procedure using cell selection transformations and artefact rejection procedures. Specimen classification with a cut-off at greater than 0.3% alarms (= percentage of automatically selected objects per epithelial cells) and greater than 10 alarms, results in a false negative rate (FNR) of 0.3% (1 case out of 321 cases with severe dysplasia or more serious lesions), a false positive rate (FPR) of 13% (663 negative cases) and a rejection rate of 2.7%. Besides a machine classification, LEYTAS offers a second, machine-interaction classification of those preparations which have been declared positive by the machine. Machine-interaction involves visual evaluation of the stored images of the detected objects (alarms) and reduces the FPR from 13 to 8%. Statistical tests further demonstrate the significance of the screening results. Presently the main drawback for routine use of automated screening with LEYTAS seems to be the time consuming preparation procedure, since instrumentation has now been updated to a new, fast and user-friendly version of LEYTAS.     

Segmentation of color images from serous cytology for automated cell classification

OBJECTIVE: To design an automated system for the classification of cells based on analysis of serous cytology, with the aim of segmenting both cytoplasm and nucleus using color information from the images as the main characteristic of the cells. STUDY DESIGN: The segmentation strategy uses color information coupled with mathematical morphology tools, such as watersheds. Cytoplasm and nuclei of all diagnostic cells are retained; erythrocytes and debris are eliminated. Special techniques are used for the separation of clustered cells. RESULTS: A large set of cells was assessed by experts to score the segmentation success rate. All cells were segmented whatever their spatial configurations. The average success rate was 92.5% for nuclei and 91.1% for cytoplasm. CONCLUSION: This color information-based segmentation of images of serous cells is accurate and provides a useful tool. This segmentation strategy will improve the automated classification of cells.     

Insulin receptor processing as a function of erythrocyte age. A kinetic model for down-regulation

The effects of cell aging on insulin binding and on insulin receptor processing in human erythrocytes were studied. Erythrocytes were found to exponentially lose equal proportions of both high and low affinity receptors as a function of age. The affinities of remaining surface receptors did not change significantly. The maximum extent of insulin receptor down-regulation that could be induced decreased linearly with age over the range studied. Together with dose-response and time course studies, these age-related changes in insulin binding and receptor down-regulation were used to develop a kinetic model in which receptor internalization is a function of surface receptor concentration. The ability of the model to predict the behavior of a heterogeneous population suggests that changes in receptor processing with age may be attributed to changes in the surface receptor concentration.     

Building an automated classification of DNA-binding protein domains

Intensive growth in 3D structure data on DNA-protein complexes as reflected in the Protein Data Bank (PDB) demands new approaches to the annotation and characterization of these data and will lead to a new understanding of critical biological processes involving these data. These data and those from other protein structure classifications will become increasingly important for the modeling of complete proteomes. We propose a fully automated classification of DNA-binding protein domains based on existing 3D-structures from the PDB. The classification, by domain, relies on the Protein Domain Parser (PDP) and the Combinatorial Extension (CE) algorithm for structural alignment. The approach involves the analysis of 3D-interaction patterns in DNA-protein interfaces, assignment of structural domains interacting with DNA, clustering of domains based on structural similarity and DNA-interacting patterns. Comparison with existing resources on describing structural and functional classifications of DNA-binding proteins was used to validate and improve the approach proposed here. In the course of our study we defined a set of criteria and heuristics allowing us to automatically build a biologically meaningful classification and define classes of functionally related protein domains. It was shown that taking into consideration interactions between protein domains and DNA considerably improves the classification accuracy. Our approach provides a high-throughput and up-to-date annotation of DNA-binding protein families which can be found at http://spdc.sdsc.edu.