Automated analysis and classification of infected macrophages using bright-field amplitude contrast data

This article presents a methodology for acquisition and analysis of bright-field amplitude contrast image data in high-throughput screening (HTS) for the measurement of cell density, cell viability, and classification of individual cells into phenotypic classes. We present a robust image analysis pipeline, where the original data are subjected to image standardization, image enhancement, and segmentation by region growing. This work develops new imaging and analysis techniques for cell analysis in HTS and successfully addresses a particular need for direct measurement of cell density and other features without using dyes.     

Nuclear morphology in preneoplastic lesions of rat liver

The nuclear morphology of preneoplastic and unaltered hepatocytes in diethylnitrosamine-treated rats was investigated. Two-micrometer-thick sections of methacrylate-embedded liver were scanned with a TV camera and evaluated in a computer using multivariate analysis methods. The preneoplastic cell populations (islands) were distinguished from unaltered hepatocytes by histochemical demonstration of glycogen storage in specimens from starved animals. After the hemalaun-stained liver sections were scanned randomly, the sections were stained for glycogen, and the previously registered cells were identified visually using a scanning stage for relocation. This objective identification of unaltered and preneoplastic hepatocytes formed the basis for the selection of a training set for feature evaluation and supervised classification. Image analysis for quantitative nuclear morphology was applied to the hemalaun-stained cells. The results showed that condensed chromatin was reduced and nuclear area was increased in the nuclei of glycogen-storing cells. Differences in the nuclear structure were also found. A multivariate analysis including seven features gave a correct classification result of about 82%.     

Analysis of rare high-DNA cell populations in serous effusions using continuous-motion imaging

The rare high-DNA cell sub-populations in a series of serous effusion specimens were analysed to determine whether such measurements could provide a basis for the improved diagnosis of malignancy. Monolayer specimens stained with gallocyanin chrome-alum were scanned with the CERVIFIP continuous-motion image analyser to locate and measure the highest-DNA cells in the sample. Two types of features were obtained for the detected sub-populations; firstly, 'percentile ploidy' values which characterise the ploidy levels above which specified proportions of the total cells are found; and secondly 'percentage abnormal' values which characterise the proportion of the cells diagnosed as malignant during examination by a cytopathologist. The classification accuracy for one or both of these features was then obtained by comparison with the clinical outcome of each patient. The results gave a classification error of 9/44 (20%) using the 0.01% percentile ploidy alone, 6/44 (14%) using the 75% percentage abnormal feature alone, but only 2/44 (5%) from a box discriminant using both features. It was therefore concluded that the analysis of the high-DNA cell population could be of value in the diagnosis of malignancy in serous effusion specimens.     

Comparison of Papanicolaou staining and Feulgen staining for automated prescreening

Feulgen staining is considered to be a quantitative DNA-specific cytochemical procedure. The applicability of this staining in high-resolution cytometry was tested in comparison with a regressive Papanicolaou staining. Papanicolaou-stained or Feulgen-stained intermediate and carcinoma cells selected by a cytologist were examined with a Zeiss scanning microscope photometer at 546 and 560 nm, respectively. After cell image segmentation and feature extraction, a statistical data evaluation was carried out by computer. Cell distributions with respect to four selected nuclear features demonstrated the influence of the staining procedure on cell feature measurements. The discriminatory power of the classification system as related to both staining procedures was studied using discriminant analysis. Using only nuclear features, a 7.3% improvement of the overall correct classification rate (from 85.0% to 92.3%) was achieved using Feulgen staining. The misclassification rate was simultaneously reduced by 50%. Using cytoplasmic as well as nuclear features, a 98% rate of correct classification was achieved.     

Robust algorithm for brain magnetic resonance image (MRI) classification based on GARCH variances series

In this paper, a robust algorithm for disease type determination in brain magnetic resonance image (MRI) is presented. The proposed method classifies MRI into normal or one of the seven different diseases. At first two-level two-dimensional discrete wavelet transform (2D DWT) of input image is calculated. Our analysis show that the wavelet coefficients of detail sub-bands can be modeled by generalized autoregressive conditional heteroscedasticity (GARCH) statistical model. The parameters of GARCH model are considered as the primary feature vector. After feature vector normalization, principal component analysis (PCA) and linear discriminant analysis (LDA) are used to extract the proper features and remove the redundancy from the primary feature vector. Finally, the extracted features are applied to the K-nearest neighbor (KNN) and support vector machine (SVM) classifiers separately to determine the normal image or disease type. Experimental results indicate that the proposed algorithm achieves high classification rate and outperforms recently introduced methods while it needs less number of features for classification.     

Analysis of machine-selected cells with an image analysis system in normal and abnormal cervical specimens

An analysis has been performed of visual diagnostic criteria used in cervical cytology applied to machine selected cells in relation to automated classification based on variables, which can be recorded in an image system with automated cell search and segmentation, feature extraction and classification. A 98% accuracy could be obtained with the choice of the most ideal statistical methods for discrimination and the use of the most powerful variables recorded in the image system when compared with consensus of the visual diagnoses based on established cytological criteria for diagnosis of cancer and precancer of the cervix uteri. The most powerful discriminatory variables in the image system (of 17 recorded) for discrimination between normal and abnormal epithelial cells were, in addition to nuclear extinction, cytoplasmic extinction and cytoplasmic shape. It is concluded that the visual classification of cervical cells is highly accurate with experienced observers and that imaging microscopes can be trained to nearly equal this accuracy with appropriate statistical methods of discrimination. The problem of creating fully automated systems, however, also requires the inclusion of even more effective discriminatory variables and also the solution of such problems as automatic cell search, segmentation, artifact rejection, feature extraction, classification and electronic stability in order to become cost-effective.     

Computer assistance in multiparameter flow microphotometry of mammalian cells

Flow system technology enables the biological and medical experimenter to investigate the distribution spectra of various cellular characteristics separately or in parameter combination on the basis of ultrarapid single cell measurements. A typical rate of recognition is about 1000 to 5000 cells per second and the precision of measurements and their statistical relevance has been previously unobtainable. According to the approach of the multiparameter analysis and high data rate, computer assistance in flow system technology is given primary consideration. In this study three different kinds of software controlled modes in data acquisition are demonstrated: normal acquisition and linear accumulation of single parameters, spectra accumulation of two correlated parameters of each single cell and documentation as linear, two- or three-dimensional distribution pattern, and linear accumulation of two correlated parameters simultaneously with their actual signal-to-signal ratio. A first attempt to analyze distribution spectra was the application of the entropy of the structure routinely used in cybernetics. This function seems to be a measure for determining the degree of synchrony in an appropriate pretreated cell population. A special mathematical strategy has been applied to the linear spectra of cellular DNA content, whose advantage is the quantitative extraction of the fractions concerning the various phases of the life cycle cells. The validity of this special curve fitting procedure has been proven on various experimental cell populations.     

Tissue counter analysis of histologic sections of melanoma: influence of mask size and shape, feature selection, statistical methods and tissue preparation.

BACKGROUND: Tissue counter analysis is an image analysis tool designed for the detection of structures in complex images at the macroscopic or microscopic scale. As a basic principle, small square or circular measuring masks are randomly placed across the image and image analysis parameters are obtained for each mask. Based on learning sets, statistical classification procedures are generated which facilitate an automated classification of new data sets. OBJECTIVE: To evaluate the influence of the size and shape of the measuring masks as well as the importance of feature selection, statistical procedures and technical preparation of slides on the performance of tissue counter analysis in microscopic images. As main quality measure of the final classification procedure, the percentage of elements that were correctly classified was used. STUDY DESIGN: HE-stained slides of 25 primary cutaneous melanomas were evaluated by tissue counter analysis for the recognition of melanoma elements (section area occupied by tumour cells) in contrast to other tissue elements and background elements. Circular and square measuring masks, various subsets of image analysis features and classification and regression trees compared with linear discriminant analysis as statistical alternatives were used. The percentage of elements that were correctly classified by the various classification procedures was assessed. In order to evaluate the applicability to slides obtained from different laboratories, the best procedure was automatically applied in a test set of another 50 cases of primary melanoma derived from the same laboratory as the learning set and two test sets of 20 cases each derived from two different laboratories, and the measurements of melanoma area in these cases were compared with conventional assessment of vertical tumour thickness. RESULTS: Square measuring masks were slightly superior to circular masks, and larger masks (64 or 128 pixels in diameter) were superior to smaller masks (8 to 32 pixels in diameter). As far as the subsets of image analysis features were concerned, colour features were superior to densitometric and Haralick texture features. Statistical moments of the grey level distribution were of least significance. CART (classification and regression tree) analysis turned out to be superior to linear discriminant analysis. In the best setting, 95% of melanoma tissue elements were correctly recognized. Automated measurement of melanoma area in the independent test sets yielded a correlation of r=0.846 with vertical tumour thickness (p<0.001), similar to the relationship reported for manual measurements. The test sets obtained from different laboratories yielded comparable results. CONCLUSIONS: Large, square measuring masks, colour features and CART analysis provide a useful setting for the automated measurement of melanoma tissue in tissue counter analysis, which can also be used for slides derived from different laboratories.     

Machine learning improves the precision and robustness of high-content screens: using nonlinear multiparametric methods to analyze screening results

Imaging-based high-content screens often rely on single cell-based evaluation of phenotypes in large data sets of microscopic images. Traditionally, these screens are analyzed by extracting a few image-related parameters and use their ratios (linear single or multiparametric separation) to classify the cells into various phenotypic classes. In this study, the authors show how machine learning-based classification of individual cells outperforms those classical ratio-based techniques. Using fluorescent intensity and morphological and texture features, they evaluated how the performance of data analysis increases with increasing feature numbers. Their findings are based on a case study involving an siRNA screen monitoring nucleoplasmic and nucleolar accumulation of a fluorescently tagged reporter protein. For the analysis, they developed a complete analysis workflow incorporating image segmentation, feature extraction, cell classification, hit detection, and visualization of the results. For the classification task, the authors have established a new graphical framework, the Advanced Cell Classifier, which provides a very accurate high-content screen analysis with minimal user interaction, offering access to a variety of advanced machine learning methods.     

Leukemia-related morphological features in blast cells

This paper investigates the use of image-processing methods to detect leukemia-related morphological differences in mononuclear blast cells. Routinely prepared Pappenheim-stained blood smears were scanned in a high-resolution color TV-microscope system. Eleven blast-cell classes (OMSBC, T-ALL, OMS, ALL, LBL, IBL, AUL, AML, AMOL, AMMOL, and CML) were analyzed with the nonparametric statistical software program "Classification and Regression Trees" (CART). This paper documents the initial statistical evaluation of 62 leukemia-related morphological features that directly measure and analyze the cell-related quantifiable differences occurring in the various blast cells. The 62 cell image features include both common cytophotometric features, and new texture and color features developed for this project. This study found that each leukemia specimen contains a dominant class of blasts that correlates with the specific leukemia, plus a distribution of blasts from related diseases. The present data suggest the existence of a distribution fingerprint pattern for each leukemia.     

Three-dimensional image processing for morphometric analysis of epithelium sections.

The reproducible classification of poorly differentiated abnormal epithelium specimens is still a diagnostic problem. The computer-aided method described here improves the differentiation between benign and malignant epithelium specimens. Hematoxylin and eosin-stained sections of normal squamous epithelium, dysplasia, carcinoma in situ, and carcinoma were scanned in a TV microscope system and analyzed by means of image processing methods on a DEC 5000/200 workstation. From the 15-20 microns thick histological sections, 3-5 focus positions in steps of 1-4 microns were scanned. The segmentation of the cell nuclei was performed automatically by color analysis and geometric operations. For each nucleus the best focus level was selected and at this level the center of the cell was calculated. Graph theoretical methods were applied to analyze the morphometry of the epithelium specimens. The minimal spanning tree was computed in the three-dimensional (3D) space of the sections with the selected centers of the nuclei as vertices. The best feature found for discrimination of the specimens is the average length of all edges in a tree. In the two-dimensional (2D) analysis we had to accept an error probability of about 20% in differentiation of dysplasia and carcinoma. In contrast to this we differentiated normal squamous epithelium, dysplasia, and carcinoma with a correct classification rate of 100% in the 3D analysis.     

Discriminant analysis on cells from developing squamous cancer of the respiratory tract.

Cytologic preparations made from the tracheobronchial tree taken by the Schreiber catheter have been scanned by three color microphotometry. The digitized cell images were processed by the analytical cytodiagnostic programs of the TICAS system. Cells were sorted into two control groups and five groups of increasing atypia ranging from normal epithelium to invasive squamous cell carcinoma. Standard statistical tests, including Wilk's Lambda, Rao's V, and the Kruskal-Wallis tests are performed on these subsets of cell image features. This study demonstrates that discriminant analyses permit differentiation between normal cells and those from marked atypia or carcinoma and that the classification achieves a high degree of agreement with visual assignment.     

Discriminant analysis on cells from developing squamous cancer of the respiratory tract

Cytologic preparations made from the tracheobronchial tree taken by the Schreiber catheter have been scanned by three color microphotometry. The digitized cell images were processed by the analytical cytodiagnostic programs of the TICAS system. Cells were sorted into two control groups and five groups of increasing atypia ranging from normal epithelium to invasive squamous cell carcinoma. Standard statistical tests, including Wilk's Lambda, Rao's V, and the Kruskal-Wallis tests are performed on these subsets of cell image features. This study demonstrates that discriminant analyses permit differentiation between normal cells and those from marked atypia or carcinoma and that the classification achieves a high degree of agreement with visual assignment.     

Boosting multiclass learning with repeating codes and weak detectors for protein subcellular localization

MOTIVATION: Determining locations of protein expression is essential to understand protein function. Advances in green fluorescence protein (GFP) fusion proteins and automated fluorescence microscopy allow for rapid acquisition of large collections of protein localization images. Recognition of these cell images requires an automated image analysis system. Approaches taken by previous work concentrated on designing a set of optimal features and then applying standard machine-learning algorithms. In fact, trends of recent advances in machine learning and computer vision can be applied to improve the performance. One trend is the advances in multiclass learning with error-correcting output codes (ECOC). Another trend is the use of a large number of weak detectors with boosting for detecting objects in images of real-world scenes. RESULTS: We take advantage of these advances to propose a new learning algorithm, AdaBoost.ERC, coupled with weak and strong detectors, to improve the performance of automatic recognition of protein subcellular locations in cell images. We prepared two image data sets of CHO and Vero cells and downloaded a HeLa cell image data set in the public domain to evaluate our new method. We show that AdaBoost.ERC outperforms other AdaBoost extensions. We demonstrate the benefit of weak detectors by showing significant performance improvements over classifiers using only strong detectors. We also empirically test our method's capability of generalizing to heterogeneous image collections. Compared with previous work, our method performs reasonably well for the HeLa cell images. AVAILABILITY: CHO and Vero cell images, their corresponding feature sets (SSLF and WSLF), our new learning algorithm, AdaBoost.ERC, and Supplementary Material are available at http://aiia.iis.sinica.edu.tw/     

衰老标记蛋白30 mRNA在癌细胞中的表达研究

目的:检测衰老标记蛋白(SMP)30 mRNA在不同癌细胞系中的表达情况,探讨其在不同细胞中的表达差异。方法:分别采用RT-PCR与荧光定量PCR检测SMP30 mRNA在正常肝细胞、肝癌细胞、胃癌细胞、乳腺癌细胞、宫颈癌细胞中的表达,并用SPSS13.0进行统计学分析。结果:SMP30 mRNA在所有被检测的细胞株中均有表达,在癌细胞中的相对表达量分别为肝癌细胞(0.926±0.340)、胃癌细胞(0.922±0.379)、乳腺癌细胞(0.614±0.356)、宫颈癌细胞(0.608±0.346),而在正常肝细胞中为0.175±0.158,显示SMP30 mRNA在癌细胞中的表达量较正常肝细胞中高(P0.05),且在肝癌细胞中的表达量比在其他癌细胞中更高。结论:SMP30 mRNA在癌细胞中的表达高于正常肝细胞,且在肝癌细胞中的表达高于其他癌细胞,具有临床应用价值。     

Automated Characterization and Parameter-Free Classification of Cell Tracks Based on Local Migration Behavior

Cell migration is the driving force behind the dynamics of many diverse biological processes. Even though microscopy experiments are routinely performed today by which populations of cells are visualized in space and time, valuable information contained in image data is often disregarded because statistical analyses are performed at the level of cell populations rather than at the single-cell level. Image-based systems biology is a modern approach that aims at quantitatively analyzing and modeling biological processes by developing novel strategies and tools for the interpretation of image data. In this study, we take first steps towards a fully automated characterization and parameter-free classification of cell track data that can be generally applied to tracked objects as obtained from image data. The requirements to achieve this aim include: (i) combination of different measures for single cell tracks, such as the confinement ratio and the asphericity of the track volume, and (ii) computation of these measures in a staggered fashion to retrieve local information from all possible combinations of track segments. We demonstrate for a population of synthetic cell tracks as well as for in vitro neutrophil tracks obtained from microscopy experiment that the information contained in the track data is fully exploited in this way and does not require any prior knowledge, which keeps the analysis unbiased and general. The identification of cells that show the same type of migration behavior within the population of all cells is achieved via agglomerative hierarchical clustering of cell tracks in the parameter space of the staggered measures. The recognition of characteristic patterns is highly desired to advance our knowledge about the dynamics of biological processes.     

Automated detection and recognition of live cells in tissue culture using image cytometry

An automated image cytometry device, the Cell Analyzer, was used to locate live V79 cells plated at low densities in a tissue culture flask. Cells and other objects were detected by moving the flask in steps across a linear solid-state image sensor. The step size was selected to be small enough to allow detection of all the cells in the area being scanned but sufficiently large so that most cells would be detected on only one image line. To distinguish cells from other detected objects, a recognition algorithm utilizing 18 characteristic cell signal features was developed. The algorithm first tests whether a set of feature values falls within specified upper and lower bounds, and then applies a linear discriminant function to the remaining data to further discriminate cells from debris. False-positive errors of 5% or less were achieved with this method, whereas 15-35% of cells were misclassified as debris.     

Application of discriminant analysis and quantitative cytologic examination to gastric lesions

OBJECTIVE: To investigate of the potential value of morphometry and discriminant analysis for the classification of benign and malignant gastric cells and lesions. STUDY DESIGN: The data set consisted of 13,300 cells from 120 cases composed of 30 cases of cancer, 26 cases of gastritis and 64 cases of ulcer according to the final histologic diagnosis. The cytologic diagnosis was divided into 5 categories (gastritis, ulcer, inflammatory dysplasia, cancer and true dysplasia). Classification was attempted at 2 levels: the cell level to classify individual cells and the case level to classify individual cases. For the cellular classification the measured cells from 50% of available cases were selected as a training set to construct a model. The cells from the remaining cases were used as a test set to validate the model. Similarly for case classification, the same 50% of cases that were used for cell classification were used as a training set and the remaining cases as a test set. Images of routinely processed gastric smears stained by the Papanicolaou technique were analyzed by a customized image analysis system. RESULTS: Application of discriminant analysis on the test set gave correct classification of 98.4% of benign cells and 67.1% of malignant cells. On case classification, 100% accuracy was achieved for benign and malignant cases, both for the training and test sets. CONCLUSION: The application of discriminant analysis described in this paper could produce significant classification results at the cellular and individual case level.     

Deep learning of diffraction image patterns for accurate classification of five cell types

Development of label‐free methods for accurate classification of cells with high throughput can yield powerful tools for biological research and clinical applications. We have developed a deep neural network of DINet for extracting features from cross‐polarized diffraction image (p‐DI) pairs on multiple pixel scales to accurately classify cells in five types. A total of 6185 cells were measured by a polarization diffraction imaging flow cytometry (p‐DIFC) method followed by cell classification with DINet on p‐DI data. The averaged value and SD of classification accuracy were found to be 98.9% ± 1.00% on test data sets for 5‐fold training and test. The invariance of DINet to image translation, rotation, and blurring has been verified with an expanded p‐DI data set. To study feature‐based classification by DINet, two sets of correctly and incorrectly classified cells were selected and compared for each of two prostate cell types. It has been found that the signature features of large dissimilarities between p‐DI data of correctly and incorrectly classified cell sets increase markedly from convolutional layers 1 and 2 to layers 3 and 4. These results clearly demonstrate the importance of high‐order correlations extracted at the deep layers for accurate cell classification.