Computerized classification of intraductal breast lesions using histopathological images

Dündar, Murat; Badve, Sunil; Bilgin, Gökhan; Raykar, Vikas C.; Jain, R. K.; Sertel, Olcay; Gürcan, Metin N.

doi:10.1109/tbme.2011.2110648

Cited by 171 publications

(124 citation statements)

References 11 publications

Supporting

Mentioning

120

Contrasting

Unclassified

Order By: Relevance

“…A similar approach is to manually designate a region in the WSI for analysis (Al-Kadi, 2010;Basavanhally et al, 2010Basavanhally et al, , 2008Doyle et al, 2012aDoyle et al, , 2008Doyle et al, , 2007Dundar et al, 2011Dundar et al, , 2010Qureshi et al, 2008;Sertel et al, 2010). This is the digital equivalent of the subsetting process described above: in both procedures, a human uses their knowledge and time to reduce the data available for analysis based on their evaluation of what is informative.…”

Section: Image Subsetting Methodsmentioning

confidence: 99%

Automated classification of brain tumor type in whole-slide digital pathology images using local representative tiles

Barker

Hoogi

Depeursinge

et al. 2016

Medical Image Analysis

193

110

View full text Add to dashboard Cite

a b s t r a c tComputerized analysis of digital pathology images offers the potential of improving clinical care (e.g. automated diagnosis) and catalyzing research (e.g. discovering disease subtypes). There are two key challenges thwarting computerized analysis of digital pathology images: first, whole slide pathology images are massive, making computerized analysis inefficient, and second, diverse tissue regions in whole slide images that are not directly relevant to the disease may mislead computerized diagnosis algorithms. We propose a method to overcome both of these challenges that utilizes a coarse-to-fine analysis of the localized characteristics in pathology images. An initial surveying stage analyzes the diversity of coarse regions in the whole slide image. This includes extraction of spatially localized features of shape, color and texture from tiled regions covering the slide. Dimensionality reduction of the features assesses the image diversity in the tiled regions and clustering creates representative groups. A second stage provides a detailed analysis of a single representative tile from each group. An Elastic Net classifier produces a diagnostic decision value for each representative tile. A weighted voting scheme aggregates the decision values from these tiles to obtain a diagnosis at the whole slide level. We evaluated our method by automatically classifying 302 brain cancer cases into two possible diagnoses (glioblastoma multiforme (N = 182) versus lower grade glioma (N = 120)) with an accuracy of 93.1 % (p << 0.001). We also evaluated our method in the dataset provided for the 2014 MICCAI Pathology Classification Challenge, in which our method, trained and tested using 5-fold cross validation, produced a classification accuracy of 100% (p << 0.001). Our method showed high stability and robustness to parameter variation, with accuracy varying between 95.5% and 100% when evaluated for a wide range of parameters. Our approach may be useful to automatically differentiate between the two cancer subtypes.

show abstract

Section: Image Subsetting Methodsmentioning

confidence: 99%

Automated classification of brain tumor type in whole-slide digital pathology images using local representative tiles

Barker

Hoogi

Depeursinge

et al. 2016

Medical Image Analysis

193

110

View full text Add to dashboard Cite

show abstract

“…This is the case in which positive bags do not contain any negatively labelled instances and a special MIL algorithm needs to be implemented to handle this. The main contrast between these two frameworks is that, in the MIL based frameworks, the need for expert marking still exits during the classifier training phase (Zhao et al, 2006;Dundar et al, 2008Dundar et al, , 2010Dundar et al, , 2011.…”

Section: Conclusion and Discussionmentioning

confidence: 99%

“…Although there are MIL classifiers that calculate the output labels for each instance (Andrews et al, 2002;Chen et al, 2006) several MIL classifiers were designed to perform slide-level classification (Dundar et al, 2010(Dundar et al, , 2011. In contrast, the QSL algorithm calculates the posterior probabilities for all individual instances.…”

Section: Conclusion and Discussionmentioning

confidence: 99%

“…There were many studies conducted for automated detection of regions having the characteristics of a specific disease. The diseases taken into consideration in this respect include colorectal dysplasia (Hamilton et al, 1997), breast lesions (Sahiner et al, 1996;Dundar et al, 2010Dundar et al, , 2011, renal cell carcinoma (Waheed et al, 2007), cervical (Hallouche et al, 1992), prostate (Diamond et al, 1982;Pitts et al, 1993;Doyle et al, 2006Doyle et al, , 2007Huang and Lee, 2009), oral cancers (Muthu et al, 2012;Mookiah et al, 2011) and colon cancers (Hamilton et al, 1987;Nasser Esgiar et al, 1998;Rajpoot and Rajpoot, 2004;Masood et al, 2006;Filippas et al, 2003;Nwoye et al, 2006).…”

Section: Introductionmentioning

confidence: 99%

“…The goal of MIL is to classify unseen bags or instances based on the labelled bags as the training data. This classification methodology was used in several applications of the automated diagnosis of breast lesions (Dundar et al, 2010(Dundar et al, , 2011 and classification human histological images (Zhao et al, 2006).…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Automated labelling of cancer textures in colorectal histopathology slides using quasi-supervised learning

Onder¹,

Sarıoğlu²,

Karaçali³

2013

Micron

View full text Add to dashboard Cite

a b s t r a c tQuasi-supervised learning is a statistical learning algorithm that contrasts two datasets by computing estimate for the posterior probability of each sample in either dataset. This method has not been applied to histopathological images before. The purpose of this study is to evaluate the performance of the method to identify colorectal tissues with or without adenocarcinoma. Light microscopic digital images from histopathological sections were obtained from 30 colorectal radical surgery materials including adenocarcinoma and non-neoplastic regions. The texture features were extracted by using local histograms and co-occurrence matrices. The quasi-supervised learning algorithm operates on two datasets, one containing samples of normal tissues labelled only indirectly, and the other containing an unlabeled collection of samples of both normal and cancer tissues. As such, the algorithm eliminates the need for manually labelled samples of normal and cancer tissues for conventional supervised learning and significantly reduces the expert intervention. Several texture feature vector datasets corresponding to different extraction parameters were tested within the proposed framework. The Independent Component Analysis dimensionality reduction approach was also identified as the one improving the labelling performance evaluated in this series. In this series, the proposed method was applied to the dataset of 22,080 vectors with reduced dimensionality 119 from 132. Regions containing cancer tissue could be identified accurately having false and true positive rates up to 19% and 88% respectively without using manually labelled ground-truth datasets in a quasi-supervised strategy. The resulting labelling performances were compared to that of a conventional powerful supervised classifier using manually labelled ground-truth data. The supervised classifier results were calculated as 3.5% and 95% for the same case. The results in this series in comparison with the benchmark classifier, suggest that quasi-supervised image texture labelling may be a useful method in the analysis and classification of pathological slides but further study is required to improve the results.

show abstract

Automatic detection of melanoma progression by histological analysis of secondary sites

et al. 2012

View full text Add to dashboard Cite

We present results from machine classification of melanoma biopsies sectioned and stained with hematoxylin/eosin (H&E) on tissue micro-arrays (TMA). The four stages of melanoma progression were represented by seven tissue types, including benign nevus, primary tumors with radial and vertical growth patterns (stage I) and four secondary metastatic tumors: subcutaneous (stage II), lymph node (stage III), gastrointestinal and soft tissue (stage IV). Our experiment setup comprised 14,208 image samples based on 164 TMA cores. In our experiments we constructed an HE color space by digitally deconvolving the RGB images into separate H (hematoxylin) and E (eosin) channels. We also compared three different classifiers: Weighted Neighbor Distance (WND), Radial Basis Functions (RBF), and k-Nearest Neighbors (kNN). We found that the HE color space consistently outperformed other color spaces with all three classifiers, while the different classifiers did not have as large of an effect on accuracy. This showed that a more physiologically relevant representation of color can have a larger effect on correct image interpretation than downstream processing steps. We were able to correctly classify individual fields of view with an average of 96% accuracy when randomly splitting the dataset into training and test fields. We also obtained a classification accuracy of 100% when testing entire cores that were not previously used in training (four random trials with one test core for each of 7 classes, 28 tests total). Because each core corresponded to a different patient, this test more closely mimics a clinically relevant setting where new patients are evaluated based on training with previous cases. The analysis method used in this study contains no parameters or adjustments that are specific to melanoma morphology, suggesting it can be used for analyzing other tissues and phenotypes, as well as potentially different image modalities and contrast techniques.

show abstract

Computerized classification of intraductal breast lesions using histopathological images

Cited by 171 publications

References 11 publications

Automated classification of brain tumor type in whole-slide digital pathology images using local representative tiles

Automated classification of brain tumor type in whole-slide digital pathology images using local representative tiles

Automated labelling of cancer textures in colorectal histopathology slides using quasi-supervised learning

Automatic detection of melanoma progression by histological analysis of secondary sites

Contact Info

Product

Resources

About