Limits...
Staining pattern classification of antinuclear autoantibodies based on block segmentation in indirect immunofluorescence images.

Li J, Tseng KK, Hsieh ZY, Yang CW, Huang HN - PLoS ONE (2014)

Bottom Line: The aim of this paper is to design an automatic system to identify the staining patterns based on block segmentation compared to the cell segmentation most used in previous research.Various feature descriptors and classifiers are tested and compared in the classification of the staining pattern of blocks and it is found that the technique of the combination of the local binary pattern and the k-nearest neighbor algorithm achieve the best performance.Relying on the results of block pattern classification, experiments on the whole images show that classifier fusion rules are able to identify the staining patterns of the whole well (specimen image) with a total accuracy of about 94.62%.

View Article: PubMed Central - PubMed

Affiliation: Department of Computer Science and Technology, Harbin Institute of Technology, Shenzhen Graduate School, Shenzhen, China.

ABSTRACT
Indirect immunofluorescence based on HEp-2 cell substrate is the most commonly used staining method for antinuclear autoantibodies associated with different types of autoimmune pathologies. The aim of this paper is to design an automatic system to identify the staining patterns based on block segmentation compared to the cell segmentation most used in previous research. Various feature descriptors and classifiers are tested and compared in the classification of the staining pattern of blocks and it is found that the technique of the combination of the local binary pattern and the k-nearest neighbor algorithm achieve the best performance. Relying on the results of block pattern classification, experiments on the whole images show that classifier fusion rules are able to identify the staining patterns of the whole well (specimen image) with a total accuracy of about 94.62%.

Show MeSH

Related in: MedlinePlus

Accuracies of different combinations of classifier, feature and fusion rule: from right to left sequentially GLCM+SVM+WMR, GLCM+KNN+MR, LBP+BPNN+MR, LBP+BPNN+WMR, LBP+KNN+MR, LBP+KNN+WMR, LBP+KNN+WSR, LDA+KNN+MR, SIFT(vlfeat)+MR and SIFT(vlfeat)+WMR.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4256175&req=5

pone-0113132-g009: Accuracies of different combinations of classifier, feature and fusion rule: from right to left sequentially GLCM+SVM+WMR, GLCM+KNN+MR, LBP+BPNN+MR, LBP+BPNN+WMR, LBP+KNN+MR, LBP+KNN+WMR, LBP+KNN+WSR, LDA+KNN+MR, SIFT(vlfeat)+MR and SIFT(vlfeat)+WMR.

Mentions: In this experiment, various combinations of classifier, feature and fusion rule were utilized to evaluate the performance of the staining pattern recognition of the HEp-2 cell image. Figure 9 presents the accuracies of 10 combinations mainly focusing on the LBP feature and KNN classifier, with some passive combinations omitted, such as LDA feature and BPNN classifier, LDA feature and SVM classifier etc. LBP+BPNN+MR and LBP+KNN+WSR achieve the same accuracy, 94.62%, and the classification results based on the LBP feature is more positive than other features with a maximum accuracy of 76.15% using GLCM+KNN+MR (Figure 9). This indicates that the LBP feature is the most suitable characteristic to identify ANA patterns.


Staining pattern classification of antinuclear autoantibodies based on block segmentation in indirect immunofluorescence images.

Li J, Tseng KK, Hsieh ZY, Yang CW, Huang HN - PLoS ONE (2014)

Accuracies of different combinations of classifier, feature and fusion rule: from right to left sequentially GLCM+SVM+WMR, GLCM+KNN+MR, LBP+BPNN+MR, LBP+BPNN+WMR, LBP+KNN+MR, LBP+KNN+WMR, LBP+KNN+WSR, LDA+KNN+MR, SIFT(vlfeat)+MR and SIFT(vlfeat)+WMR.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4256175&req=5

pone-0113132-g009: Accuracies of different combinations of classifier, feature and fusion rule: from right to left sequentially GLCM+SVM+WMR, GLCM+KNN+MR, LBP+BPNN+MR, LBP+BPNN+WMR, LBP+KNN+MR, LBP+KNN+WMR, LBP+KNN+WSR, LDA+KNN+MR, SIFT(vlfeat)+MR and SIFT(vlfeat)+WMR.
Mentions: In this experiment, various combinations of classifier, feature and fusion rule were utilized to evaluate the performance of the staining pattern recognition of the HEp-2 cell image. Figure 9 presents the accuracies of 10 combinations mainly focusing on the LBP feature and KNN classifier, with some passive combinations omitted, such as LDA feature and BPNN classifier, LDA feature and SVM classifier etc. LBP+BPNN+MR and LBP+KNN+WSR achieve the same accuracy, 94.62%, and the classification results based on the LBP feature is more positive than other features with a maximum accuracy of 76.15% using GLCM+KNN+MR (Figure 9). This indicates that the LBP feature is the most suitable characteristic to identify ANA patterns.

Bottom Line: The aim of this paper is to design an automatic system to identify the staining patterns based on block segmentation compared to the cell segmentation most used in previous research.Various feature descriptors and classifiers are tested and compared in the classification of the staining pattern of blocks and it is found that the technique of the combination of the local binary pattern and the k-nearest neighbor algorithm achieve the best performance.Relying on the results of block pattern classification, experiments on the whole images show that classifier fusion rules are able to identify the staining patterns of the whole well (specimen image) with a total accuracy of about 94.62%.

View Article: PubMed Central - PubMed

Affiliation: Department of Computer Science and Technology, Harbin Institute of Technology, Shenzhen Graduate School, Shenzhen, China.

ABSTRACT
Indirect immunofluorescence based on HEp-2 cell substrate is the most commonly used staining method for antinuclear autoantibodies associated with different types of autoimmune pathologies. The aim of this paper is to design an automatic system to identify the staining patterns based on block segmentation compared to the cell segmentation most used in previous research. Various feature descriptors and classifiers are tested and compared in the classification of the staining pattern of blocks and it is found that the technique of the combination of the local binary pattern and the k-nearest neighbor algorithm achieve the best performance. Relying on the results of block pattern classification, experiments on the whole images show that classifier fusion rules are able to identify the staining patterns of the whole well (specimen image) with a total accuracy of about 94.62%.

Show MeSH
Related in: MedlinePlus