Dansheng Song

Tree structured wavelet transform segmentation of microcalcifications in digital mammography

A novel multistage algorithm is proposed for the automatic segmentation of microcalcification clusters (MCCs) in digital mammography. First, a previously reported tree structured nonlinear filter is proposed for suppressing image noise, while preserving image details, to potentially reduce the false positive (FP) detection rate for MCCs. Second, a tree structured wavelet transform (TSWT) is applied to the images for microcalcification segmentation. The TSWT employs quadrature mirror filters as basic subunits for both multiresolution decomposition and reconstruction processes, where selective reconstruction of subimages is used to segment MCCs. Third, automatic linear scaling is then used to display the image of the segmented MCCs on a computer monitor for interpretation. The proposed algorithms were applied to an image database of 100 single view mammograms at a resolution of 105 microns and 12 bits deep (4096 gray levels). The database contained 50 cases of biopsy proven malignant MCCs, 8 benign cases, and 42 normal cases. The measured sensitivity (true positive detection rate) was 94% with a low FP detection rate of 1.6 MCCs/image. The image details of the segmented MCCs were reasonably well preserved, for microcalcification of less than 500 microns, with good delineation of the extent of the microcalcification clusters for each case based on visual criteria.

Digital mammography: wavelet transform and Kalman-filtering neural network in mass segmentation and detection.

RATIONALE AND OBJECTIVES The authors developed a new adaptive module to improve their computer-assisted diagnostic (CAD) method for mass segmentation and classification. The goal was an adaptive module that used a novel four-channel wavelet transform with neural network rather than a two-channel wavelet transform with manual subimage selection. The four-channel wavelet transform is used for image decomposition and reconstruction, and a novel Kalman-filtering neural network is used for adaptive subimage selection. MATERIALS AND METHODS The adaptive CAD module was compared with the nonadaptive module by comparing receiver operating characteristic curves for the whole CAD system. An image database containing 800 regions of interest enclosing all mass types and normal tissues was used for the relative comparison of system performance, with electronic ground truth established in advance. RESULTS The receiver operating characteristic curves yield Az values of 0.93 and 0.86 with and without the adaptive module respectively, suggesting that overall CAD performance is improved with the adaptive module. CONCLUSION The results of this study confirm the importance of using a new class of adaptive CAD methods that allow a more generalized application for larger image databases or images generated from different sensors or by means of direct x-ray detection, as required for clinical trials.

Digital mammography: Hybrid four-channel wavelet transform for microcalcification segmentation

RATIONALE AND OBJECTIVES The authors evaluated an algorithm for the automatic segmentation of microcalcification clusters (MCCs) at digital mammography. Two- and four-channel wavelet transforms were evaluated to determine whether sensitivity in the detection of MCCs can be improved and if the selective reconstruction of the higher-order M2 subimages allows better preservation of the segmented MCCs, which is required for their classification. MATERIALS AND METHODS The hybrid method involved the use of a nonlinear filter for image noise suppression coupled with wavelet transforms for image decomposition and an adaptive method for selective subimage reconstruction as a basis for segmentation of MCCs. The two- and four-channel wavelet transforms were implemented with different filter bank structures (i.e., polyphase quadrature mirror filters [QMFs], tree structure, and lattice structure) to determine if their computational efficiency can be improved while retaining properties such as near-perfect reconstruction. The hybrid wavelet transforms were applied to a common image database of biopsy-proved MCCs (100 images, 105-micron resolution, 12 bits deep; 52 cases with at least one MCC of varying subtlety [46 malignant and six benign cases] and eight normal cases). RESULTS The two- and four-channel wavelet transforms yielded sensitivities of 93% and 94% and false-positive (PP) detection rates of 1.58 and 1.35 MCCs per image, respectively. The lattice structure provided greater than fivefold improvement in computational speed compared to the polyphase QMF structure, particularly for the higher order of channels (M = 4). CONCLUSION The four-channel wavelet transform provided better sensitivity and FP detection rates and greater image detail preservation for the segmented MCCs.

A new false positive reduction method for MCCs detection in digital mammography

A new mixed feature multistage false positive (FP) reduction method has been developed for improving the FP reduction performance. Eleven features were extracted from both spatial and morphology domains in order to describe the micro-calcification clusters (MCCs) from different perspectives. These features are grouped into three categories: gray-level description, shape description and clusters description. Two feature sets that focus on describing MCCs on every single calcification and on clustered calcifications, respectively, were combined with a back-propagation (BP) neural network with Kalman filter (KF) to obtain the best performance of FP reduction. First, 9 of the 11 gray-level description and shape description features were employed with BP neural network to eliminate all the obvious FP calcifications in the image. Second, the remaining MCCs will be classified into several clusters by a widely used criterion in clinical practice, and then the two cluster description features will be added to the first feature set to eliminate the FP clusters from the remaining MCCs. The performance results of this approach were obtained using an image database of 100 real cases of patients mammogram images in H. Lee Moffitt Cancer Center imaging program.

Adaptive Neural Network for Nuclear Medicine Image Restoration

A novel adaptive neural network is proposed for image restoration using a nuclear medicine gamma camera based on the point spread function of measured system. The objective is to restore image degradation due to photon scattering and collimator photon penetration with the gamma camera and allow improved quantitative external measurements of radionuclides in-vivo. The specific clinical model proposed is the imaging of bremsstrahlung radiation using 32P and 90Y because of the enhanced image degradation effects of photon scattering, photon penetration and poor signal/noise ratio in measurements of this type with the gamma camera. This algorithm model avoids the common inverse problem associated with other image restoration filters such as the Wiener filter. The relative performance of the adaptive NN for image restoration is compared to a previously reported order statistic neural network hybrid (OSNNH) filter by these investigators, a traditional Weiner filter and a modified Hopfield neural network using simulated degraded images with different noise levels. Quantitative metrics such as the change of signal to noise ratio (ΔSNR) are used to compare filter performance. The adaptive NN yields comparable results for image restoration with a slightly better performance for the images with higher noise level as often encountered in bremsstrahlung detection with the gamma camera. Experimental attenuation measurements were also performed in a water tank using two radionuclides, 32P and 90Y, typically used for antibody therapy. Similar values for an effective attenuation coefficient was observed for the restored images using the OSNNH filters and adaptive NN which demonstrate that the restoration filters preserves the total counts in the image as required for quantitative in-vivo measurements. The adaptive NN was computationally more efficient by a factor 4–6 compared to the OSNNH filter. The filter architecture, in turn, is also optimum for parallel processing or VLSI implementation as required for planar and particularly for tomographic mode of detection using the gamma camera. The proposed adaptive NN method should also prove to be useful for quantitative imaging of single photon emitters for other nuclear medicine tomographic imaging applications using positron emitters and direct X-ray photon detection.

A kernelised fuzzy-Support Vector Machine CAD system for the diagnosis of lung cancer from tissue images

This research describes a non-interactive process that applies several forms of computational intelligence to classifying biopsy lung tissue samples. Three types of lung cancer evaluated (squamous cell carcinoma, adenocarcinoma, and bronchioalveolar carcinoma) together account for 65?70% of diagnoses. Accuracy achieved supports hypothesis that an accurate predictive model is generated from training images, and performance achieved is an accurate baseline for the processs potential scaling to larger datasets. Feature vector performance is good or better than Thiran and Macqs in every case. Except bronchioalveolar carcinomas, each individual cancer classification task experienced improvement, with two groupings showing nearly 20% classification accuracy.

Ipsilateral multi-view CAD system for mass detection in digital mammography

An ipsilateral multi-view computer-aided diagnosis (CAD) scheme is presented for the earlier mass detection in digital mammograms. Tree structured nonlinear filtering (TSF) is used in image noise suppression. Two wavelet-based methods, directional wavelet transform (DWT) and tree structured wavelet transform (TSWT) are employed for image enhancement. Adaptive fuzzy-C means (FCM) algorithm is conducted for segmentation. Concurrent analysis is employed for iterative analysis of ipsilateral multi-view mammograms to raise detection sensitivity and specificity, and a supervised three-layer artificial neural network (ANN) in which the backpropagation (BP) algorithm combined with Kalman filtering is used as training algorithm is developed as a classifier, which has been trained using the training database with biopsy proven truth files. The application of such CAD system in digital mammography is reported in this article. The test database consists of 200 cases in which the distribution of normal, abnormal cases balanced, and free-response receiver operating characteristic (FROC) analysis method is used to test the performance of the developed unilateral CAD system. The performance comparison has been conducted between the final ipsilateral mufti-view CAD system and current single-view CAD system. The study results have shown that the advantages of ipsilateral mufti-view CAD method over current single-view CAD system express the feasibility of ipsilateral multi-view CAD system combined with concurrent analysis method described in this paper for the improvement of overall performance of CAD system in the early stage mass detection.

Three-class classification in computer-aided diagnosis of breast cancer by support vector machine

Design of classifier in computer-aided diagnosis (CAD) scheme of breast cancer plays important role to its overall performance in sensitivity and specificity. Classification of a detected object as malignant lesion, benign lesion, or normal tissue on mammogram is a typical three-class pattern recognition problem. This paper presents a three-class classification approach by using two-stage classifier combined with support vector machine (SVM) learning algorithm for classification of breast cancer on mammograms. The first classification stage is used to detect abnormal areas and normal breast tissues, and the second stage is for classification of malignant or benign in detected abnormal objects. A series of spatial, morphology and texture features have been extracted on detected objects areas. By using genetic algorithm (GA), different feature groups for different stage classification have been investigated. Computerized free-response receiver operating characteristic (FROC) and receiver operating characteristic (ROC) analyses have been employed in different classification stages. Results have shown that obvious performance improvement in both sensitivity and specificity was observed through proposed classification approach compared with conventional two-class classification approaches, indicating its effectiveness in classification of breast cancer on mammograms.

Prognosis of stage I lung cancer patients through quantitative analysis of centrosomal features

Centrosome amplification leads to the loss of regulated chromosome segregation, aneuploidy, and chromosome instability and has the possibility to be a biomarker of cancer prognosis. To explore this feasibility, resected, stage I non-small cell lung cancer (NSCLC) tissues from six survivor and six fatal cases were immunostained and scanned. Regions of interest were selected to include one cell and its centrosomes. After segmentation, feature abstraction, and optimization, six nonredundant features were used for statistical analysis and classification. Two analytic methods showed that for each feature, centrosomes from survivors differed from centrosomes of fatalities, indicating sampling from different populations. The data were classified using linear discriminant analysis (LDA) and support vector machines (SVM) with 10-fold cross-validation. Classification accuracy was 74% by LDA and 79% by SVM, respectively, and further improved to 85% with bagging. Centrosome can be a biomarker for stage I NSCLC prognosis and has potential for clinical utility.

Fuzzy image segmentation for lung nodule detection

This paper focuses on evaluating three fuzzy image segmentation algorithms in lung nodule detection scenario: fuzzy entropy-based method, multivariate fuzzy C-means method (MFCM), adaptive fuzzy C-means method (AFCM) and comparing them with the iterative threshold selection method. The experimental result shows that all three methods outperform iterative threshold selection method. The two fuzzy C-means clustering based algorithms achieve better segmentation performance without losing true positives. However, fuzzy entropy-based image segmentation removes the false positives at the cost of losing some true positives, which is a risky approach and hence it is not recommended for lung nodule detection. Moreover, although AFCM outperforms MFCM in true positive detection significantly, in the sense of TPR/FP, MFCM is comparable to AFCM in the confidence interval of significant level 0.95, since AFCM brings in more false positives than MFCM.