Automated whole breast ultrasound (ABUS) has been widely used as a screening modality for examination of breast abnormalities. Reviewing hundreds of slices produced by ABUS, however, is time consuming. Therefore, in this paper, a fast and effective computer-aided detection system based on 3-D convolutional neural networks (CNNs) and prioritized candidate aggregation is proposed to accelerate this reviewing. First, an efficient sliding window method is used to extract volumes of interest (VOIs). Then, each VOI is estimated the tumor probability with a 3-D CNN, and VOIs with higher estimated probability are selected as tumor candidates. Since the candidates may overlap each other, a novel scheme is designed to aggregate the overlapped candidates. During the aggregation, candidates are prioritized based on estimated tumor probability to alleviate over-aggregation issue. The relationship between the sizes of VOI and target tumor is optimally exploited to effectively perform each stage of our detection algorithm. On evaluation with a test set of 171 tumors, our method achieved sensitivities of 95% (162/171), 90% (154/171), 85% (145/171), and 80% (137/171) with 14.03, 6.92, 4.91, and 3.62 false positives per patient (with six passes), respectively. In summary, our method is more general and much faster than preliminary works and demonstrates promising results.

Automated whole breast ultrasound (ABUS) is becoming a popular screening modality for whole breast examination. Compared to conventional handheld ultrasound, ABUS achieves operator-independent and is feasible for mass screening. However, reviewing hundreds of slices in an ABUS image volume is time-consuming. A computer-aided detection (CADe) system based on watershed transform was proposed in this study to accelerate the reviewing. The watershed transform was applied to gather similar tissues around local minima to be homogeneous regions. The likelihoods of being tumors of the regions were estimated using the quantitative morphology, intensity, and texture features in the 2-D/3-D false positive reduction (FPR). The collected database comprised 68 benign and 65 malignant tumors. As a result, the proposed system achieved sensitivities of 100% (133/133), 90% (121/133), and 80% (107/133) with FPs/pass of 9.44, 5.42, and 3.33, respectively. The figure of merit of the combination of three feature sets is 0.46 which is significantly better than that of other feature sets ( [Formula: see text]). In summary, the proposed CADe system based on the multi-dimensional FPR using the integrated feature set is promising in detecting tumors in ABUS images.

Automated whole breast ultrasound (ABUS) is an emerging screening tool for detecting breast abnormalities. In this study, a computer-aided detection (CADe) system based on multi-scale blob detection was developed for analyzing ABUS images. The performance of the proposed CADe system was tested using a database composed of 136 breast lesions (58 benign lesions and 78 malignant lesions) and 37 normal cases. After speckle noise reduction, Hessian analysis with multi-scale blob detection was applied for the detection of tumors. This method detected every tumor, but some nontumors were also detected. The tumor like lihoods for the remaining candidates were estimated using a logistic regression model based on blobness, internal echo, and morphology features. The tumor candidates with tumor likelihoods higher than a specific threshold (0.4) were considered tumors. By using the combination of blobness, internal echo, and morphology features with 10-fold cross-validation, the proposed CAD system showed sensitivities of 100%, 90%, and 70% with false positives per pass of 17.4, 8.8, and 2.7, respectively. Our results suggest that CADe systems based on multi-scale blob detection can be used to detect breast tumors in ABUS images.

Computer-aided diagnosis (CAD) systems in gray-scale breast ultrasound images have the potential to reduce unnecessary biopsy of breast masses. The purpose of our study is to develop a robust CAD system based on the texture analysis. First, gray-scale invariant features are extracted from ultrasound images via multi-resolution ranklet transform. Thus, one can apply linear support vector machines (SVMs) on the resulting gray-level co-occurrence matrix (GLCM)-based texture features for discriminating the benign and malignant masses. To verify the effectiveness and robustness of the proposed texture analysis, breast ultrasound images obtained from three different platforms are evaluated based on cross-platform training/testing and leave-one-out cross-validation (LOO-CV) schemes. We compare our proposed features with those extracted by wavelet transform in terms of receiver operating characteristic (ROC) analysis. The AUC values derived from the area under the curve for the three databases via ranklet transform are 0.918 (95% confidence interval [CI], 0.848 to 0.961), 0.943 (95% CI, 0.906 to 0.968), and 0.934 (95% CI, 0.883 to 0.961), respectively, while those via wavelet transform are 0.847 (95% CI, 0.762 to 0.910), 0.922 (95% CI, 0.878 to 0.958), and 0.867 (95% CI, 0.798 to 0.914), respectively. Experiments with cross-platform training/testing scheme between each database reveal that the diagnostic performance of our texture analysis using ranklet transform is less sensitive to the sonographic ultrasound platforms. Also, we adopt several co-occurrence statistics in terms of quantization levels and orientations (i.e., descriptor settings) for computing the co-occurrence matrices with 0.632+ bootstrap estimators to verify the use of the proposed texture analysis. These experiments suggest that the texture analysis using multi-resolution gray-scale invariant features via ranklet transform is useful for designing a robust CAD system.

Breast cancer is one of the most common cancers among women. Early detection and treatment of breast cancer can effectively prohibit its progress and decrease mortality rate. Recently, ultrasound imaging plays an important role in the field of breast cancer diagnosis because of its convenience and non-invasive. With the help of computer-aided diagnosis (CAD) system, the characteristics of tumor can be detected and provided to physicians as a critical reference. Because the shape of a tumor may be altered due to the stress caused by the ultrasound probe, the proposed sub-volume registration method can be utilized to analyze the variation of tumor between pre- and post-compression. Then, we can determine whether the tumor is benign or not with several statistical materials. The experimental results will show that this proposed model can efficaciously detect the tumors and support the clinical diagnoses.

Tumor vascularity is an important factor that has been shown to correlate with tumor malignancy and was demonstrated as a prognostic indicator for a wide range of cancers. Three-dimensional (3-D) power Doppler ultrasound (PDUS) offers a convenient tool for investigators to inspect the signals of blood flow and vascular structures in breast cancer. In this paper, a new computer-aided diagnosis (CAD) system for quantifying Doppler ultrasound images based on 3-D thinning algorithm and neural network is proposed. We extracted the skeleton of blood vessels from 3-D PDUS data to facilitate the capturing of morphological changes. Nine features including vessel-to-volume ratio, number of vascular trees, length of vessels, number of branching, mean of radius, number of cycles, and three tortuosity measures, were extracted from the thinning result. Benign and malignant tumors can therefore be differentiated by a score computed by a multilayered perceptron (MLP) neural network using these features as parameters. The proposed system was tested on 221 breast tumors, including 110 benign and 111 malignant lesions. The accuracy, sensitivity, specificity, and positive and negative predictive values were 88.69% (196/221), 91.89% (102/111), 85.45% (94/110), 86.44% (102/118), and 91.26% (94/103), respectively. The Az value of the ROC curve was 0.94. The results demonstrate a correlation between the morphology of blood vessels and tumor malignancy, indicating that the newly proposed method can retrieves a high accuracy in the classification of benign and malignant breast tumors.

Recently, fractal analyses have been applied successfully for the image compression, texture analysis and texture image segmentation. The fractal dimension could be used to quantify the texture information. Several methods including box-counting, fractal Brownian motion, and iterative function system etc. can be used to estimate fractal dimension. In this study, the differences of gray value of neighboring pixels are used to estimate the fractal dimension of an ultrasound image of breast lesion by using the fractal Brownian motion. Further, a computer-aided diagnosis system based on the fractal analysis is proposed to classify the breast lesions into two classes benign and malignant. In order to improve the classification performances, the ultrasound image are pre-processed by using morphology operations and histogram equalization. Finally, k-means classification method is used to classify benign tumors from malignant ones. Experimental results will exhibit and evaluate the accuracy rate of the proposed method.

Traditional region-based image retrieval systems use only the dominant features within each region and ignore the useful relationship of neighboring regions. In this study, an integrated color/texture RBIR system using the region relationship is purposed. An image is first processed with the edgeflow technique to divide the image into several regions and to extract the important texture information. Then, the region adjacency graph is used to represent the regions and their spatial relationships in the segmented image. Furthermore, the image retrieval problem is reduced to a subgraph isomorphism algorithm, which is performed to verify the similarity between two graphs by a subgraph isomorphism heuristic algorithm. In experiments, several query results from the test database are used to evaluate the performance of the proposed system.

Spiculation is a stellate distortion caused by the intrusion of breast cancer into surrounding tissue. Its existence is an important clue to characterizing malignant tumors. Many successful mammographic methods have been proposed to detect tumors with spiculation. Traditional two-dimensional (2-D) ultrasound cannot easily find spiculations because spiculations normally appear parallel to the surface of the skin. Recently, three-dimensional (3-D) ultrasound has been gradually used in clinical applications and it has been proven to be useful in determining the architectural distortion or spiculation that surrounds a breast tumor. This paper aims to identify spiculation from 3-D ultrasonic volume data of a tumor found by a physician. In the proposed method, each coronal slice of volume data is successively extracted and then analyzed as a 2-D ultrasound image by the proposed spiculation detection method. First, in each horizontal slice, the modified rotating structuring element (ROSE) operation is used to find the central region in which spiculation lines converge. Second, the stick algorithm is used to estimate the direction of the edge of each pixel around the central region. A pixel whose edge points toward the central region is marked as a potential spiculation. Finally, the marked pixels are collected around the central region and their distribution is analyzed to determine whether spiculation is present. The 3-D test datasets were obtained using the Voluson 530 or 730, Kretztechnik, Austria. First, the proposed method was tested on 104 2-D typical coronal images (selected by an experienced physician) extracted from 52 3-D ultrasonic datasets. Finally, 225 3-D pathologically proven datasets were tested to evaluate the performance. Spiculations are more easily observed in the coronal view than in the other two views. That is, the 3-D ultrasound is a powerful tool for identifying spiculations. Furthermore, 16% (19/120) of benign cases and 90% (94/105) of malignant cases are detected as spiculations.

The goal of this paper is to use the three-dimensional (3-D) snake technique in 3-D ultrasound to obtain the tumor contour for the pre- and the post-operative malignant breast excision by the vacuum assisted biopsy instrument Mammotome. This technique of assessing the margin of two can help the physician to evaluate the effect of the surgery. By using the anisotropic diffusion filter, the noise and speckles can be reduced. Then the stick detection is adopted for enhancing the edge. Finally, the gradient vector flow (GVF) snake is used to obtain the tumor contour. These techniques are extended to the 3-D techniques to increase the accuracy and robust of segmentation results. We hope that this study can help physicians to improve the minimal invasive operation for a breast tumor.

With the rapid growth of the Internet and multimedia systems, digital images have been enormously used in many applications now. The way of how to retrieve the image we want from a large image database correctly plays a more important role. Currently, most proposed methods for image retrieval can only be applied with images in the same representing domain. We propose a novel methodology that extracts the feature image directly from its original domain without any decompressing procedures. The feature is further modified to achieve the rotation-invariant and scaling property. Our proposed method can be used for image retrieval on the three most frequently adopted types of images nowadays: spatial, DCT compressed, and wavelet-compressed images. Experimental results show that our proposed method is quite effective not only for the performance but also is not afraid of scale and rotation.

The multiresolution wavelet transform has been shown to be an effective technique and achieved very good performance for texture analysis. However, a large number of images are compressed by methods based on the discrete cosine transform (DCT). Hence, the image decompression of the inverse DCT is needed to obtain the texture features based on the wavelet transform for the DCT-coded image. This paper proposes the use of the multiresolution reordered features for texture analysis. The proposed features are directly generated by using the DCT coefficients from the DCT-coded image. Comparisons with the subband-energy features extracted from the wavelet transform, conventional DCT using the Brodatz (1966) texture database indicate that the proposed method provides the best texture pattern retrieval accuracy and obtains a much better correct classification rate. The proposed DCT based features are expected to be very useful and efficient for texture pattern retrieval and classification in large DCT-coded image databases.

We present nonlinear interpolation schemes for image resolution enhancement. The multilayer perceptron (MLP) interpolation schemes based on the wavelet transform and subband filtering are proposed. Because estimating each subimage signal is more effective than estimating the whole image signal, pixels in the low-resolution image are used as the input signal of the MLP to estimate all of the wavelet subimage of the corresponding high-resolution image. The image of increased resolution is finally produced by the synthesis procedure of the wavelet transform. As compared with other popular methods, the results show that the improvement is remarkable.

In this paper, an adaptive predictive multiplicative autoregressive (APMAR) method is proposed for lossless medical image coding. The adaptive predictor is used for improving the prediction accuracy of encoded image blocks in our proposed method. Each block is first adaptively predicted by one of the seven predictors of the JPEG lossless mode and a local mean predictor. It is clear that the prediction accuracy of an adaptive predictor is better than that of a fixed predictor. Then the residual values are processed by the MAR model with Huffman coding. Comparisons with other methods [MAR, SMAR, adaptive JPEG (AJPEG)] on a series of test images show that our method is suitable for reversible medical image compression.

Very low bit rate image sequence coding is very important for video transmission and storage applications. The fundamental goal of image sequence coding is to remove spatial redundancy and only update the moving parts in an image sequence so that the total information bits required for storage or transmission can be greatly reduced. In our proposed approach, each frame within an image sequence will be separated into moving and stationary blocks. Only the moving blocks need to be transmitted to the decoder, so that the total number of bits and computing time are greatly reduced. The moving blocks will be encoded by edge-based side-match finite-state classified vector quantization (EBSMCVQ). Moreover, a quadtree can be also used to represent the moving and edge information for each frame. In order to reduce the number of bits for the quadtree, we propose a new difference quadtree technique that only transmits the different parts between the previous frame's quadtree and the current frame's quadtree. In the proposed interframe difference quadtree EBSMCVQ image sequence coding scheme, the average bit rate of each frame is reduced to 0.0393 b/pixel and the PSNR is still up to 35.40 dB for image sequence Claire.

Vector quantization (VQ) is an effective image coding technique at low bit rate. The side-match finite-state vector quantizer (SMVQ) exploits the correlations between neighboring blocks (vectors) to avoid large gray level transition across block boundaries. A new adaptive edge-based side-match finite-state classified vector quantizer (classified FSVQ) with a quadtree map has been proposed. In classified FSVQ, blocks are arranged into two main classes, edge blocks and nonedge blocks, to avoid selecting a wrong state codebook for an input block. In order to improve the image quality, edge vectors are reclassified into 16 classes. Each class uses a master codebook that is different from the codebooks of other classes. In our experiments, results are given and comparisons are made between the new scheme and ordinary SMVQ and VQ coding techniques. As is shown, the improvement over ordinary SMVQ is up to 1.16 dB at nearly the same bit rate, moreover, the improvement over ordinary VQ can be up to 2.08 dB at the same bit rate for the image, Lena. Further, block boundaries and edge degradation are less visible because of the edge-vector classification. Hence, the perceptual image quality of classified FSVQ is better than that of ordinary SMVQ.

Subband coding (SBC) with vector quantization (VQ) has been shown to be an effective method for coding images at low bit rates. We split the image spectrum into seven nonuniform subbands. Threshold vector quantization (TVQ) and finite state vector quantization (FSVQ) methods are employed in coding the subband images by exploiting interband and intraband correlations. Our new SBC-FSVQ schemes have the advantages of the subband-VQ scheme while reducing the bit rate and improving the image quality. Experimental results are given and comparisons are made using our new scheme and some other coding techniques. In the experiments, it is found that SBC-FSVQ schemes achieve the best peak signal-to-noise ratio (PSNR) performance when compared to other methods at the same bit rate.

Future B-ISDN (broadband integrated services digital network) users will be able to send various kinds of information, such as voice, data, and image, over the same network and send information only when necessary. It has been recognized that variable-rate encoding techniques are more suitable than fixed-rate techniques for encoding images in a B-ISDN environment. A new variable-rate side-match finite-state vector quantization with a block classifier (CSMVQ) algorithm is described. In an ordinary fixed-rate SMVQ, the size of the state codebook is fixed. In the CSMVQ algorithm presented, the size of the state codebook is changed according to the characteristics of the current vector which can be predicted by a block classifier. In experiments, the improvement over SMVQ was up to 1.761 dB at a lower bit rate. Moreover, the improvement over VQ can be up to 3 dB at nearly the same bit rate.

A coding algorithm must have the ability to adapt to changing image characteristics for image sequences. An adaptive finite-state vector quantization (FSVQ) in which the bit rate and the encoding time can be reduced is described. In order to improve the image quality and avoid producing a wrong state for an input vector, a threshold is used in FSVQ to decide whether to switch to a full searching VQ. The codebook is conditionally replenished according to a distortion threshold at a later time to reflect the local statistics of the current frame. After the codebook is replenished, one can quickly reconstruct the state codebooks of FSVQ using the state codebook selection algorithm. In the experiments, the improvement over the static SMVQ is up to 2.40 dB at nearly the same bit rate and the encoding time is only one-ninth the time required by the static SMVQ. Moreover, the improvement over the static VQ is up to 2.91 dB, and the encoding time is only three-fifths the time required by the static VQ for the image sequence 'Claire'.>

A new adaptive tree-structured vector quantization (adaptive TSVQ) with multipath searching is described. The structure of the codebook tree is reorganized at an update interval, without requiring any transmission of side information. The 1-path tree searching algorithm is the simple way to find the closest codeword in TSVQ. But the disadvantage of 1-path searching is that the selected codeworks are not, in general, closest to the input vectors. Hence, the authors propose the multipath searching algorithm to improve the image quality. In an experiment, the improvement over the fixed TSVQ at the same bit rate is up to 2.03 dB when the number of the leaves in the codebook tree is 256 and 4-path searching is applied.>