Zhili Chen

Topological Modeling and Classification of Mammographic Microcalcification Clusters

Goal: The presence of microcalcification clusters is a primary sign of breast cancer; however, it is difficult and time consuming for radiologists to classify microcalcifications as malignant or benign. In this paper, a novel method for the classification of microcalcification clusters in mammograms is proposed. Methods: The topology/connectivity of individual microcalcifications is analyzed within a cluster using multiscale morphology. This is distinct from existing approaches that tend to concentrate on the morphology of individual microcalcifications and/or global (statistical) cluster features. A set of microcalcification graphs are generated to represent the topological structure of microcalcification clusters at different scales. Subsequently, graph theoretical features are extracted, which constitute the topological feature space for modeling and classifying microcalcification clusters. k-nearest-neighbors-based classifiers are employed for classifying microcalcification clusters. Results: The validity of the proposed method is evaluated using two well-known digitized datasets (MIAS and DDSM) and a full-field digital dataset. High classification accuracies (up to 96%) and good ROC results (area under the ROC curve up to 0.96) are achieved. A full comparison with related publications is provided, which includes a direct comparison. Conclusion: The results indicate that the proposed approach is able to outperform the current state-of-the-art methods. Significance: This study shows that topology modeling is an important tool for microcalcification analysis not only because of the improved classification accuracy but also because the topological measures can be linked to clinical understanding.

Modelling mammographic microcalcification clusters using persistent mereotopology

In mammographic imaging, the presence of microcalcifications, small deposits of calcium in the breast, is a primary indicator of breast cancer. However, not all microcalcifications are malignant and their distribution within the breast can be used to indicate whether clusters of microcalcifications are benign or malignant. Computer-aided diagnosis (CAD) systems can be employed to help classify such microcalcification clusters. In this paper a novel method for classifying microcalcification clusters is presented by representing discrete mereotopological relations between the individual microcalcifications over a range of scales in the form of a mereotopological barcode. This barcode based representation is able to model complex relations between multiple regions and the results on mammographic microcalcification data shows the effectiveness of this approach. Classification accuracies of 95% and 80% are achieved on the MIAS and DDSM datasets, respectively. These results are comparable to existing state-of-the art methods. This work also demonstrates that mereotopological barcodes could be used to help trained clinicians in their diagnosis by providing a clinical interpretation of barcodes that represent both benign and malignant cases.

Automated Mammographic Risk Classification Based on Breast Density Estimation

This paper presents a method for automated mammographic risk classification based on breast density estimation in mammograms. The overall profile of breast tissue density is represented using a topographic map, which is a hierarchical representation, obtained from the upper level sets of an image. A shape tree is constructed to describe the topological and geometrical structure of the shapes (i.e. connected components) within the topographic map. Two properties, saliency and independency, are defined to detect shapes of interest (i.e. dense regions) based on the shape tree. A density map is further generated focusing on dense regions, which provides a quantitative description of breast density. Finally, mammographic risk classification is performed based on the breast density measures derived from the density map. The validity of this method is evaluated using the full MIAS database and a large dataset taken from the DDSM database. A high agreement with expert radiologists is indicated according to the BIRADS density classification. The obtained classification accuracies are 76.01% and 81.22%, respectively.

A modified fuzzy c-means algorithm for breast tissue density segmentation in mammograms

The fuzzy c-means (FCM) algorithm has been applied in a variety of medical image segmentation applications. The conventional FCM algorithm uses the greylevel information at a single pixel as the feature space and this contains no spatial contextual information, which makes it very sensitive to noise and intensity inhomogeneities. Recently, some modified FCM algorithms with spatial constraints have been published. However, these have individual disadvantages and are not robust enough with different types of noise. In this paper, we propose a modified FCM algorithm incorporating local spatial and intensity information based on an adaptive local window filter whose weighting coefficients differentiate the neighbouring pixels within the local window. Fast clustering is afterwards performed on the intensity histogram of the filtered image. To demonstrate the robustness and insensitivity to noise of the proposed algorithm, it is extensively tested using synthetic images corrupted by a variety of noise. The experimental results are quantitatively evaluated and compared. This algorithm is then applied to mammographic images for breast tissue density segmentation. The segmentation results indicate its effectiveness to the presence of intensity inhomogeneities in mammograms from different density categories.

A combined method for automatic identification of the breast boundary in mammograms

Breast region segmentation is an essential prerequisite in the (semi-)automatic analysis of digital or digitised mammographic images, which aims to separate the breast region from background information in mammograms. It normally consists of two independent segmentations, which are breast-background segmentation and pectoral muscle segmentation, respectively. The first identifies the boundary between the breast and background, and the second identifies the boundary of the pectoral muscle (present in medio-lateral oblique (MLO) views). In this paper, we propose a method for automatic identification of the breast boundary based on a combination of segmentation approaches, including histogram thresholding, edge detection, contour growing, polynomial fitting, and region growing. To demonstrate the validity of the proposed method, it is tested using two mammographic datasets: the full MIAS database and a large dataset taken from the EPIC database. For the breast-background segmentation, 98.8% and 91.5% nearly accurate results are obtained for the MIAS and EPIC data, respectively. For the pectoral muscle segmentation, 92.8% and 87.9% nearly accurate results are achieved for these two datasets. A comparison with two other methods is also provided based on the full MIAS database. These indicate the proposed method performs effectively in identifying the breast boundary in digitised mammograms. The obtained segmentation results can be used for further analysis in computer-aided diagnosis.

Computer-aided diagnosis: detection and localization of prostate cancer within the peripheral zone.

We propose a methodology for prostate cancer detection and localization within the peripheral zone based on combining multiple segmentation techniques. We extract four image features using Gaussian and median filters. Subsequently, we use each image feature separately to generate binary segmentations. Finally, we take the intersection of all four binary segmentations, incorporating a model of the peripheral zone, and perform erosion to remove small false-positive regions. The initial evaluation of this method is based on 275 MRI images from 37 patients, and 86% of the slices were classified correctly with 87% and 86% sensitivity and specificity achieved, respectively. This paper makes two contributions: firstly, a novel computer-aided diagnosis approach, which is based on combining multiple segmentation techniques using only a small number of simple image features, and secondly, the development of the proposed method and its application in prostate cancer detection and localization using a single MRI modality with the results comparable with the state-of-the-art multimodality and advanced computer vision methods in the literature. Copyright

Classification of microcalcification clusters based on morphological topology analysis

The presence of microcalcification clusters is a primary sign of breast cancer. It is difficult and time consuming for radiologists to diagnose microcalcifications. In this paper, we present a novel method for classification of malignant and benign microcalcification clusters in mammograms. We analyse the connectivity/topology between individual microcalcifications within a cluster using multiscale morphology. A microcalcification graph is constructed to represent the topological structure of clusters. A multiscale topological feature vector is generated by extracting two microcalcification graph properties. The validity of the proposed method is evaluated using a dataset taken from the MIAS database. The performance of including SFS feature selection is investigated. Using a k-nearest neighbour classifier, a classification accuracy of 95% and an area under the ROC curve of 0.93 are achieved. A comparison with existing approaches is presented.

Local feature based mammographic tissue pattern modelling and breast density classification

It has been shown that there is a strong correlation between breast tissue density/patterns and the risk of developing breast cancer. Thus, modelling mammographic tissue patterns is important for quantitative analysis of breast density and computer-aided mammographic risk assessment. In this paper, we first review different local feature based texture representation algorithms, where images are represented as occurrence histograms over a dictionary of local features. Subsequently, we use these approaches to model mammographic tissue patterns based on local tissue appearances in mammographic images. We investigate the performance of different breast tissue representations for breast denstiy classification. The evaluation is based on the full MIAS database using BIRADS ground truth. The obtained classification results are comparable with existing work, which indicates the potential capability of local feature based texture representation in mammographic tissue pattern analysis.

Analysis of Mammographic Microcalcification Clusters Using Topological Features

In mammographic images, the presence of microcalcification clusters is a primary indicator of breast cancer. However, not all microcalcification clusters are malignant and it is difficult and time consuming for radiologists to discriminate between malignant and benign microcalcification clusters. In this paper, a novel method for classifying microcalcification clusters in mammograms is presented. The topology/connectivity of microcalcification clusters is analysed by representing their topological structure over a range of scales in graphical form. Graph theoretical features are extracted from microcalcification graphs to constitute the topological feature space of microcalcification clusters. This idea is distinct from existing approaches that tend to concentrate on the morphology of individual microcalcifications and/or global (statistical) cluster features. The validity of the proposed method is evaluated using two well-known digitised datasets (MIAS and DDSM) and a full-field digital dataset. High classification accuracies (up to 96%) and good ROC results (area under the ROC curve up to 0.96) are achieved. In addition, a full comparison with state-of-the-art methods is provided.

A Multiscale Blob Representation of Mammographic Parenchymal Patterns and Mammographic Risk Assessment

Mammographic parenchymal patterns have been found to be a strong indicator of breast cancer risk and play an important role in mammographic risk assessment. In this paper, a novel representation of mammographic parenchymal patterns is proposed, which is based on multiscale blobs. Approximately blob-like tissue patterns are detected over a range of scales and parenchymal patterns are represented as a set of blobs. Spatial relations between blobs are considered to reduce the overlap between connected dense tissue regions. Quantitative measures of breast density are calculated from the resulting blobs and used for mammographic risk assessment. The proposed approach is evaluated using the full MIAS database and a large dataset from the DDSM database. A high agreement with expert radiologists is indicated according to the BIRADS density classification. The classification accuracies for the MIAS and DDSM databases are up to 79.44% and 76.90%, respectively.