Learning contextual relationships in mammograms using a hierarchical pyramid neural network

A Computer Aided Breast Cancer Detection Using Unit-Linking Pulse Coupled Neural Network & Multiphase Level Set Method

Breast cancer is one of the lethal diseases with high mortality rates among women. An early detection and diagnosis of the disease can help increase the survival rate. Distinguishing a normal breast tissue from a cancerous one proves to be ambiguous for a Radiologist. A computer aided system can help a radiologist in better and efficient diagnosis. This paper aims at detection and classification of benign and malignant mammogram images with Unit-linking Pulse Coupled Neural Network combined with Multiphase level set Method. While Unit linking Pulse Coupled Neural Network (PCNN) helps in coarse feature extraction, Multi phase Level Set method helps in extracting minute details and hence, better classification. The proposed method is tested with images from MIAS open-source database. Performance of the proposed method is measured using sensitivity, accuracy, specificity and false positive rate. Experiments show that the proposed method gives satisfactory results when compared to the state-of-art methods. The sensitivity obtained by the proposed method is 95.16%, an accuracy of 96.76%, the False Positive Rate (FPR) is as less as 0.85% and specificity of 97.12%.

A Hybrid Image Filtering Method for Computer-Aided Detection of Microcalcification Clusters in Mammograms

The presence of microcalcification clusters (MCs) in mammogram is a major indicator of breast cancer. Detection of an MC is one of the key issues for breast cancer control. In this paper, we present a highly accurate method based on a morphological image processing and wavelet transform technique to detect the MCs in mammograms. The microcalcifications are firstly enhanced by using multistructure elements morphological processing. Then, the candidates of microcalcifications are refined by a multilevel wavelet reconstruction approach. Finally, MCs are detected based on their distributions feature. Experiments are performed on 138 clinical mammograms. The proposed method is capable of detecting 92.9% of true microcalcification clusters with an average of 0.08 false microcalcification clusters detected per image.

A New Approach to the Detection of Lesions in Mammography Using Fuzzy Clustering

Breast cancer is a leading cause of female mortality and its early detection is an important means of reducing this. The present study investigated an approach, based on fuzzy clustering, to detect small lesions, such as microcalcifications and other masses, that are hard to recognize in breast cancer screening. A total of 180 mammograms were analysed and classified by radiologists into three groups ( n = 60 per group): those with microcalcifications; those with tumours; and those with no lesions. Twenty mammograms were taken as training data sets from each of the groups. The algorithm was then applied to the data not taken for training. Analysis by fuzzy clustering achieved a mean accuracy of 99.7% compared with the radiologists' findings. It was concluded that the fuzzy clustering algorithm allowed for more efficient and accurate detection of breast lesions and may improve the early detection of breast tumours.

A GRAPH-BASED ALGORITHM FOR CLUSTER DETECTION

In some Computer Vision applications there is the need for grouping, in one or more clusters, only a part of the whole dataset. This happens, for example, when samples of interest for the application at hand are present together with several noisy samples.

In this paper we present a graph-based algorithm for cluster detection that is particularly suited for detecting clusters of any size and shape, without the need of specifying either the actual number of clusters or the other parameters.

The algorithm has been tested on data coming from two different computer vision applications. A comparison with other four state-of-the-art graph-based algorithms was also provided, demonstrating the effectiveness of the proposed approach.

Bioinformatic system modeling on hetian uygur natural longevity people

Longevity and life science are active topics in biomedicine and other subjects. In this research, longevity people from Hetian area in Xinjiang, China are used as an example. The cause of longevity is discussed and a bioinformatic longevity model is established based on the medical findings. Human life is a complex multi-variant natural process. It is complicated yet important to extract expert knowledge that can describe the interactions among different factors and influence of the factors on human life. Artificial intelligent (AI) and information processing techniques are used to efficiently process large amount of collected biomedical data and effectively extract hidden information into the longevity model. The test results show that the established model is able to identify individuals who belong to longevity group with over 90 percent accuracy. This research creates a new approach to explore the cause of formation of human longevity based on comprehensive medical data rather than just from one medical subject. More importantly, this research explores a practical way to model complex bioinformatic systems.

A novel method for breast cancer prognosis using wavelet packet based neural network

This paper presents an approach for early breast cancer diagnostic by employing combination of artificial neural networks (ANN) and wavelet based subband image decomposition which detect microcalcification in digital mammograms. The microcalcifications correspond to high-frequency components of the image spectrum, detection of microcalcifications is achieved by decomposing the mammograms into different frequency subbands, suppressing the low-frequency subband, and finally, reconstructing the mammogram from the subbands containing only high frequencies. For this approach we employed different types of wavelet packets. We used the result as an input of neural network for classification. The proposed methodology is tested using the Nijmegen and the Mammographic Image Analysis Society (MIAS) mammographic databases and images collected from local hospitals. Results are presented as the receiver operating characteristic (ROC) performance and are quantified by the area under the ROC curve.

A genetic algorithm design for microcalcification detection and classification in digital mammograms

In this paper, we propose a genetic algorithm design to automatically classify and detect micocalcification clusters in digital mammograms. The proposed GA technique is characterised by transforming input images into a feature domain, where each pixel is represented by its mean and standard deviation inside a surrounding window of size 9 x 9 pixel. In the feature domain, chromosomes are constructed to populate the initial generation and further features are extracted to enable the proposed GA to search for optimised classification and detection of microcalcification clusters via regions of 128 x 128 pixels. Extensive experiments show that the proposed GA design is able to achieve high performances in microcalcification classification and detection, which are measured by ROC curves, sensitivity against specificity, areas under ROC curves and benchmarked by existing representative techniques.

MACHINE LEARNING FOR DETECTION AND DIAGNOSIS OF DISEASE

Machine learning offers a principled approach for developing sophisticated, automatic, and objective algorithms for analysis of high-dimensional and multimodal biomedical data. This review focuses on several advances in the state of the art that have shown promise in improving detection, diagnosis, and therapeutic monitoring of disease. Key in the advancement has been the development of a more in-depth understanding and theoretical analysis of critical issues related to algorithmic construction and learning theory. These include trade-offs for maximizing generalization performance, use of physically realistic constraints, and incorporation of prior knowledge and uncertainty. The review describes recent developments in machine learning, focusing on supervised and unsupervised linear methods and Bayesian inference, which have made significant impacts in the detection and diagnosis of disease in biomedicine. We describe the different methodologies and, for each, provide examples of their application to specific domains in biomedical diagnostics.

Knowledge-discovery incorporated evolutionary search for microcalcification detection in breast cancer diagnosis

OBJECTIVES

The presence of microcalcifications (MCs), clusters of tiny calcium deposits that appear as small bright spots in a mammogram, has been considered as a very important indicator for breast cancer diagnosis. Much research has been performed for developing computer-aided systems for the accurate identification of MCs, however, the computer-based automatic detection of MCs has been shown difficult because of the complicated nature of surrounding of breast tissue, the variation of MCs in shape, orientation, brightness and size.

METHODS AND MATERIALS

This paper presents a new approach for the effective detection of MCs by incorporating a knowledge-discovery mechanism in the genetic algorithm (GA). In the proposed approach, called knowledge-discovery incorporated genetic algorithm (KD-GA), the genetic algorithm is used to search for the bright spots in mammogram and a knowledge-discovery mechanism is integrated to improve the performance of the GA. The function of the knowledge-discovery mechanism includes evaluating the possibility of a bright spot being a true MC, and adaptively adjusting the associated fitness values. The adjustment of fitness is to indirectly guide the GA to extract the true MCs and eliminate the false MCs (FMCs) accordingly.

RESULTS AND CONCLUSIONS

The experimental results demonstrate that the incorporation of knowledge-discovery mechanism into the genetic algorithm is able to eliminate the FMCs and produce improved performance comparing with the conventional GA methods. Furthermore, the experimental results show that the proposed KD-GA method provides a promising and generic approach for the development of computer-aided diagnosis for breast cancer.

3-D Reconstruction of Microcalcification Clusters Using Stereo Imaging: Algorithm and Mammographic Unit Calibration

The three-dimensional (3-D) shape of microcalcification clusters is an important indicator in early breast cancer detection. In fact, there is a relationship between the cluster topology and the type of lesion (malignant or benign). This paper presents a 3-D reconstruction method for such clusters using two 2-D views acquired during standard mammographic examinations. For this purpose, the mammographic unit was modeled using a camera with virtual optics. This model was used to calibrate the acquisition unit and then to reconstruct the clusters in the 3-D space after microcalcification segmentation and matching. The proposed model is hardware independent since it is suitable for digital mammographic units with different geometries and with various physical acquisition principles. Three-dimensional reconstruction results are presented here to prove the validity of the method. Tests were first performed using a phantom with a well-known geometry. The latter contained X-ray opaque glass balls representing microcalcifications. The positions of these balls were reconstructed with a 16.25-microm mean accuracy. This very high inherent algorithm accuracy is more than enough for a precise 3-D cluster representation. Further validation tests were carried out using a second phantom including a spherical cluster. This phantom was built with materials simulating the behavior of both mammary tissue and microcalcifications toward Xrays. The reconstructed shape was effectively spherical. Finally, reconstructions were carried out for real clusters and their results are also presented.

Correspondences between microcalcification projections on two mammographic views acquired with digital systems

In this paper, we have proposed an algorithm for automatic matching of MC projections viewed on two mammograms of the same breast. The implemented algorithm consists in three steps. From five morphological features of the MC, a similarity function was built between each MC of the first image of the pair, and each MC of the second image. These values quantified the resemblance between each pair of MC and permitted, for a given MC of the first image, to sort the MC of the second image which could be matched to it. From the geometry of the system providing the pairs of images being analysed, we derived some geometrical constraints that must be satisfied by corresponding MC. In order to take into account the fact that due to breast deformation, the corresponding MC is often off the classical epipolar line constructed on the basis of stereovision assumption, we instead consider an epipolar strip on both side of the approximated epipolar line. The MC of the second image which was out of that strip was eliminated. Then a coefficient was applied to the remaining MC that took into account their distances to the approximated epipolar line. Finally, the selection procedure was used to pick out the right pair. In order to test our algorithm, we compared the result it yielded with those coming from two operators who matched a number of MC with confidence. A concordance of 78.43% was obtained between confident manual matching and automatic matching. Since this algorithm was designed to be part of a tool for 3D reconstruction of microcalcification clusters, we reconstructed some clusters with manual matching and automatic matching and compared the shapes of the clusters obtained in these two ways. The resemblance in some cases was very good and average in a number of others. This suggests that our algorithm should be improved. Therefore, apart from ongoing effort to introduce other constraints, we believe that taking into account the third view provided by the imaging system could be of great help. We shall soon explore this possibility. Overall, we believe that despite the failure of our tool in some cases it can already at this stage be used with some confidence.

A multi-scale probabilistic network model for detection, synthesis and compression in mammographic image analysis

We develop a probabilistic network model over image spaces and demonstrate its broad utility in mammographic image analysis, particularly with respect to computer-aided diagnosis. The model employs a multi-scale pyramid decomposition to factor images across scale and a network of tree-structured hidden variables to capture long-range spatial dependencies. This factoring makes the computation of the density functions local and tractable. The result is a hierarchical mixture of conditional probabilities, similar to a hidden Markov model on a tree. The model parameters are found with maximum likelihood estimation using the expectation-maximization algorithm. The utility of the model is demonstrated for three applications: (1) detection of mammographic masses for computer-aided diagnosis; (2) qualitative assessment of model structure through mammographic synthesis; and (3) compression of mammographic regions of interest.