Breast Cancer Histopathology Image Analysis: A Review

Veta, Mitko; Pluim, Josien P. W.; Diest, Paul J. van; Viergever, Max A.

doi:10.1109/tbme.2014.2303852

Cited by 549 publications

(315 citation statements)

References 101 publications

Supporting

Mentioning

306

Contrasting

Unclassified

Order By: Relevance

“…The KNN and SVM [16] classifiers were implemented to classify images based on the features extracted. It was proved that SVM provided the maximum accuracy of classification comparatively with 88.67% accuracy.…”

Section: Resultsmentioning

confidence: 99%

See 1 more Smart Citation

Automatic Feature Extraction for Breast Density Segmentation and Classification

Cherian

Poovammal

Malathy

2017

Asian J Pharm Clin Res

View full text Add to dashboard Cite

Objective: Cancer is the uncontrollable multiplication of cells in human body. The expansion of cancerous cells in the breast area of the women is identified as breast cancer. It is mostly identified among women aged above 40. With the current advancement in the medical field, various automatic tests are available for the identification of cancerous tissues. The cancerous cells are spotted by taking the photo imprint in the form of X-ray comprising the breast area of the woman. Such images are called mammograms. Segmentation of mammograms is the primary step toward diagnosis. It involves the pre-processing of the image to identify the region of interest (ROI). Later, features are extracted from the image which involves the learned features that may be statistical and textural features. When these features are used as input to the simple classifier, it helps us to predict the risk of cancer. The support vector machine (SVM) classifier was proved to produce a better accuracy percentage with the features extracted. Methods:The mammograms are subjected to a pre-processing stage, where the images are processed to identify the ROI. Next, the features are extracted from these images to identify the statistical and textural features. Finally, these features are used as input to the simple classifier, it helps us to predict the risk of cancer Results:The SVM classifier was proved to produce the maximum accuracy of about 88.67% considering 13 features including both statistical and textural features. The features taken for the study are mean, inverse difference moment, energy, entropy, root mean square, correlation, homogeneity, variance, skewness, range, contrast, kurtosis, and smoothness. Conclusion:Computer-aided diagnosis is one of the most common methods of detection of cancer with mammograms, and it involves minor human intervention. The dataset of mammograms was analyzed and found that SVM provided the highest accuracy of 88.67%. A wide range of the study is progressing in the field of cancer as this disease causes a high threat of human life in this era.

show abstract

Section: Resultsmentioning

confidence: 99%

“…Some of the textural features can be extracted using the gray-level cooccurrence matrix (GLCM) [16] matrix. MATLAB provides a built-in function to produce this matrix.…”

Section: Feature Extractionmentioning

confidence: 99%

Automatic Feature Extraction for Breast Density Segmentation and Classification

Cherian

Poovammal

Malathy

2017

Asian J Pharm Clin Res

View full text Add to dashboard Cite

show abstract

“…The major problem is that there is a low density of mitosis in a single histological image. Hence, manual identification of mitotic cells is a difficult task even for expert pathologists to make a distinction [7]. Since most current mitosis counting approaches are based on the subjective opinion of pathologists, there is clearly a need for development of an automatic mitosis detection algorithm that works with the routinely clinical practices.…”

Section: Introductionmentioning

confidence: 99%

“…With the recent advent of whole slide digital scanners and advances in computational power, it is now possible to use digitized histopathological images and computer-aided diagnosis (CAD) algorithms for disease detection, diagnosis, and prognosis prediction to complement the opinion of the pathologist [7,8,9,10,11]. Recently, a few CAD techniques have been developed to automatically detect and classify mitosis and non-mitosis using various image features [12].…”

Section: Introductionmentioning

confidence: 99%

Automated mitosis detection in histopathology based on non-gaussian modeling of complex wavelet coefficients

et al. 2017

View full text Add to dashboard Cite

General rightsThis document is made available in accordance with publisher policies. Please cite only the published version using the reference above. AbstractTo diagnose breast cancer, the number of mitotic cells present in histology sections is an important indicator for examining and grading biopsy specimen. This study aims at improving the accuracy of automated mitosis detection by characterizing mitotic cells in wavelet based multi-resolution representations via a non-Gaussian modeling method. The potential mitosis candidates were decomposed into multi-scale forms by an undecimated dual-tree complex wavelet transform. Two non-Gaussian models (the generalized Gaussian distribution (GGD) and the symmetric alpha-stable (SαS) distributions) were used to accurately model the heavy-tailed behavior of wavelet marginal distributions. The method was evaluated on two independent data cohorts, including the benchmark dataset (MITOS), via a support vector machine classifier.

show abstract

“…The latter poses a major deterrent for considering automated mitosis detection with high throughput processing of tissue microarrays. More extensive review can be found in [5]. Recently, random forests with population update [6] has been tried for mitosis detection where tree weights are changed based on classification performance.…”

Section: Introductionmentioning

confidence: 99%

Regenerative Random Forest with Automatic Feature Selection to Detect Mitosis in Histopathological Breast Cancer Images

Paul

Dey

Mukherjee

et al. 2015

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. We propose a fast and accurate method for counting the mitotic figures from histopathological slides using regenerative random forest. Our method performs automatic feature selection in an integrated manner with classification. The proposed random forest assigns a weight to each feature (dimension) of the feature vector in a novel manner based on the importance of the feature (dimension). The forest also assigns a misclassification-based penalty term to each tree in the forest. The trees are then regenerated to make a new population of trees (new forest) and only the more important features survive in the new forest. The feature vector is constructed from domain knowledge using the intensity features of nucleus, features of nuclear membrane and features of the possible stroma region surrounding the cell. The use of domain knowledge improves the classification performance. Experiments show at least 4% improvement in F-measure with an improvement in time complexity on the MITOS dataset from ICPR 2012 grand challenge.

show abstract

Breast Cancer Histopathology Image Analysis: A Review

Cited by 549 publications

References 101 publications

Automatic Feature Extraction for Breast Density Segmentation and Classification

Automatic Feature Extraction for Breast Density Segmentation and Classification

Automated mitosis detection in histopathology based on non-gaussian modeling of complex wavelet coefficients

Regenerative Random Forest with Automatic Feature Selection to Detect Mitosis in Histopathological Breast Cancer Images

Contact Info

Product

Resources

About