An efficient transfer learning based cross model classification (TLBCM) technique for the prediction of breast cancer

Breast cancer has been the most life-threatening disease in women in the last few decades. The high mortality rate among women is due to breast cancer because of less awareness and a minimum number of medical facilities to detect the disease in the early stages. In the recent era, the situation has changed with the help of many technological advancements and medical equipment to observe breast cancer development. The machine learning technique supports vector machines (SVM), logistic regression, and random forests have been used to analyze the images of cancer cells on different data sets. Although the particular technique has performed better on the smaller data set, accuracy still needs to catch up in most of the data, which needs to be fairer to apply in the real-time medical environment. In the proposed research, state-of-the-art deep learning techniques, such as transfer learning, based cross model classification (TLBCM), convolution neural network (CNN) and transfer learning, residual network (ResNet), and Densenet proposed for efficient prediction of breast cancer with the minimized error rating. The convolution neural network and transfer learning are the most prominent techniques for predicting the main features in the data set. The sensitive data is protected using a cyber-physical system (CPS) while using the images virtually over the network. CPS act as a virtual connection between human and networks. While the data is transferred in the network, it must monitor using CPS. The ResNet changes the data on many layers without compromising the minimum error rate. The DenseNet conciliates the problem of vanishing gradient issues. The experiment is carried out on the data sets Breast Cancer Wisconsin (Diagnostic) and Breast Cancer Histopathological Dataset (BreakHis). The convolution neural network and the transfer learning have achieved a validation accuracy of 98.3%. The results of these proposed methods show the highest classification rate between the benign and the malignant data. The proposed method improves the efficiency and speed of classification, which is more convenient for discovering breast cancer in earlier stages than the previously proposed methodologies.

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%.

Towards non-data-hungry and fully-automated diagnosis of breast cancer from mammographic images

Analysing local texture and generating features are two key issues for automatic cancer detection in mammographic images. Recent researches have shown that deep neural networks provide a promising alternative to hand-driven features which suffer from curse of dimensionality and low accuracy rates. However, large and balanced training data are foremost requirements for deep learning-based models and these data are not always available publicly. In this work, we propose a fully-automated method for breast cancer diagnosis that performs training using small sets of data. Feature extraction from mammographic images is performed using a genetic-programming-based descriptor that exploits statistics on a local binary pattern-like local distribution defined in each pixel. The effectiveness of the suggested method is demonstrated on two challenging datasets, (1) the digital database for screening mammography and (2) the mammographic image analysis society digital mammogram database, for content-based image retrieval as well as for abnormality/malignancy classification. The experimental results show that the proposed method outperforms or achieves comparable results with deep learning-based methods even those with transfer learning and/or data-augmentation.

Loss of Mammographic Tissue Homeostasis in Invasive Lobular and Ductal Breast Carcinomas vs. Benign Lesions

The 2D wavelet transform modulus maxima (WTMM) method is used to perform a comparison of the spatial fluctuations of mammographic breast tissue from patients with invasive lobular carcinoma, those with invasive ductal carcinoma, and those with benign lesions. We follow a procedure developed and validated in a previous study, in which a sliding window protocol is used to analyze thousands of small subregions in a given mammogram. These subregions are categorized according to their Hurst exponent values (H): fatty tissue (H ≤ 0.45), dense tissue (H ≥ 0.55), and disrupted tissue potentially linked with tumor-associated loss of homeostasis (0.45 < H < 0.55). Following this categorization scheme, we compare the mammographic tissue composition of the breasts. First, we show that cancerous breasts are significantly different than breasts with a benign lesion (p-value ∼ 0.002). Second, the asymmetry between a patient’s cancerous breast and its contralateral counterpart, when compared to the asymmetry from patients with benign lesions, is also statistically significant (p-value ∼ 0.006). And finally, we show that lobular and ductal cancerous breasts show similar levels of disruption and similar levels of asymmetry. This study demonstrates reproducibility of the WTMM sliding-window approach to help detect and characterize tumor-associated breast tissue disruption from standard mammography. It also shows promise to help with the detection lobular lesions that typically go undetected via standard screening mammography at a much higher rate than ductal lesions. Here both types are assessed similarly.

A Novel Multiresolution-Statistical Texture Analysis Architecture: Radiomics-Aided Diagnosis of PDAC Based on Plain CT Images

Early screening of PDAC (pancreatic ductal adenocarcinoma) based on plain CT (computed tomography) images is of great significance. Therefore, this work conducted a radiomics-aided diagnosis analysis of PDAC based on plain CT images. We explored a novel MSTA (multiresolution-statistical texture analysis) architecture to extract texture features and built machine learning models to classify PDACs and HPs (healthy pancreases). We also performed significance tests of differences to analyze the relationships between histopathological characteristics and texture features. The MSTA architecture originates from the analysis of histopathological characteristics and combines multiresolution analysis and statistical analysis to extract texture features. The MSTA architecture achieved better experimental results than the traditional architecture that scales the coefficient matrices of the multiresolution analysis. In the validation of the classifications, the MSTA architecture achieved an accuracy of 81.19% and an AUC (area under the ROC (receiver operating characteristic) curve) of 0.88 (95% confidence interval: 0.84-0.92). In the test of the classifications, it achieved an accuracy of 77.66% and an AUC of 0.79 (95% confidence interval: 0.71-0.87). Moreover, the significance tests of differences showed that the extracted texture features may be relevant to the histopathological characteristics. The MSTA architecture is beneficial for the radiomics-aided diagnosis of PDAC based on plain CT images. Its texture features can potentially enhance radiologists' imaging interpretation abilities.

Detecting Brain Tumor using Machines Learning Techniques Based on Different Features Extracting Strategies

BACKGROUND

Brain tumor is the leading cause of death worldwide. It is obvious that the chances of survival can be increased if the tumor is identified and properly classified at an initial stage. MRI (Magnetic Resonance Imaging) is one source of brain tumors detection tool and is extensively used in the diagnosis of brain to detect blood clots. In the past, many researchers developed Computer-Aided Diagnosis (CAD) systems that help the radiologist to detect the abnormalities in an efficient manner.

OBJECTIVE

The aim of this research is to improve the brain tumor detection performance by proposing a multimodal feature extracting strategy and employing machine learning techniques.

METHODS

In this study, we extracted multimodal features such as texture, morphological, entropybased, Scale Invariant Feature Transform (SIFT), and Elliptic Fourier Descriptors (EFDs) from brain tumor imaging database. The tumor was detected using robust machine learning techniques such as Support Vector Machine (SVM) with kernels: polynomial, Radial Base Function (RBF), Gaussian; Decision Tree (DT), and Naïve Bayes. Most commonly used Jack-knife 10-fold Cross- Validation (CV) was used for testing and validation of dataset.

RESULTS

The performance was evaluated in terms of specificity, sensitivity, Positive Predictive Value (PPV), Negative Predictive Value (NPV), False Positive Rate (FPR), Total Accuracy (TA), Area under the receiver operating Curve (AUC), and P-value. The highest performance of 100% in terms of Specificity, Sensitivity, PPV, NPV, TA, AUC using Naïve Bayes classifiers based on entropy, morphological, SIFT and texture features followed by Decision Tree classifier with texture features (TA=97.81%, AUC=1.0) and SVM polynomial kernel with texture features (TA=94.63%). The highest significant p-value was obtained using SVM polynomial with texture features (P-value 2.65e-104) followed by SVM RB with texture features (P-value 1.96e-98).

CONCLUSION

The results reveal that Naïve Bayes followed by Decision Tree gives highest detection accuracy based on entropy, morphological, SIFT and texture features.

Assessment and Classification of Mass Lesions Based on Expert Knowledge Using Mammographic Analysis

BACKGROUND

Masses are one of the most important indicators of breast cancer in mammograms, and their classification into two groups as benign and malignant is highly necessary. Computer Aided Diagnosis (CADx) helps radiologists enhance the accuracy of their decision. Hence, the system is required to support and assess with radiologist's interaction as an expert.

METHODS

In this research, classification of breast masses using mammography in the two main views which include MLO and CC, is evaluated with respect to the shape, texture and asymmetry aspect. Additionally, a method was developed and proposed using the classification of breast tissue density based on the decision tree.

DISCUSSION

This study therefore, aims to provide a method based on the human decision-making model that will help in designing the perfect tool for radiologists, regardless of the complexity of computing, costly procedures and also reducing the diagnosis error.

CONCLUSION

Results show that the proposed system for entirely fat, scattered fibroglandular densities, heterogeneously dense, and extremely dense breast achieved 100, 99, 99 and 98% true malignant rate, respectively with cross-validation procedure.

Vertebral Bone Marrow Heterogeneity Using Texture Analysis of Chemical Shift Encoding-Based MRI: Variations in Age, Sex, and Anatomical Location

Objective: Vertebral bone marrow composition has been extensively studied in the past and shown potential as imaging biomarker for osteoporosis, hematopoietic, and metabolic disorders. However, beyond quantitative assessment of bone marrow fat, little is known about its heterogeneity. Therefore, we investigated bone marrow heterogeneity of the lumbar spine using texture analysis of chemical-shift-encoding (CSE-MRI) based proton density fat fraction (PDFF) maps and its association with age, sex, and anatomical location. Methods: One hundred and fifty-six healthy subjects were scanned (age range: 20-29 years, 12/30 males/females; 30-39, 15/9; 40-49, 5/13; 50-59, 9/27; ≥60: 9/27). A sagittal 8-echo 3D spoiled-gradient-echo sequence at 3T was used for CSE-MRI-based water-fat separation at the lumbar spine. Manual segmentation of vertebral bodies L1-4 was performed. Mean PDFF and texture features (global: variance, skewness, kurtosis; second-order: energy, entropy, contrast, homogeneity, correlation, sum-average, variance, dissimilarity) were extracted at each vertebral level and compared between age groups, sex, and anatomical location. Results: Mean PDFF significantly increased from L1 to L4 (35.89 ± 11.66 to 39.52 ± 11.18%, p = 0.017) and with age (females: 27.19 ± 6.01 to 49.34 ± 7.75%, p < 0.001; males: 31.97 ± 7.96 to 41.83 ± 7.03 %, p = 0.025), but showed no difference between females and males after adjustment for age and BMI (37.13 ± 11.63 vs. 37.17 ± 8.67%; p = 0.199). Bone marrow heterogeneity assessed by texture analysis, in contrast to PDFF, was significantly higher in females compared to males after adjustment for age and BMI (namely contrast and dissimilarity; p < 0.031), demonstrated age-dependent differences, in particular in females (p < 0.05), but showed no statistically significant dependence on vertebral location. Conclusion: Vertebral bone marrow heterogeneity, assessed by texture analysis of PDFF maps, is primarily dependent on sex and age but not on anatomical location. Future studies are needed to investigate bone marrow heterogeneity with regard to aging and disease.

Correlation-Based Ensemble Feature Selection Using Bioinspired Algorithms and Classification Using Backpropagation Neural Network

A framework for clinical diagnosis which uses bioinspired algorithms for feature selection and gradient descendant backpropagation neural network for classification has been designed and implemented. The clinical data are subjected to data preprocessing, feature selection, and classification. Hot deck imputation has been used for handling missing values and min-max normalization is used for data transformation. Wrapper approach that employs bioinspired algorithms, namely, Differential Evolution, Lion Optimization, and Glowworm Swarm Optimization with accuracy of AdaBoostSVM classifier as fitness function has been used for feature selection. Each bioinspired algorithm selects a subset of features yielding three feature subsets. Correlation-based ensemble feature selection is performed to select the optimal features from the three feature subsets. The optimal features selected through correlation-based ensemble feature selection are used to train a gradient descendant backpropagation neural network. Ten-fold cross-validation technique has been used to train and test the performance of the classifier. Hepatitis dataset and Wisconsin Diagnostic Breast Cancer (WDBC) dataset from University of California Irvine (UCI) Machine Learning repository have been used to evaluate the classification accuracy. An accuracy of 98.47% is obtained for Wisconsin Diagnostic Breast Cancer dataset, and 95.51% is obtained for Hepatitis dataset. The proposed framework can be tailored to develop clinical decision-making systems for any health disorders to assist physicians in clinical diagnosis.

Muscle Type and Gender Recognition Utilising High-Level Textural Representation in Musculoskeletal Ultrasonography

Human assistive technology and computer-aided diagnosis is an emerging field in the area of medical imaging. Following the recent advances in this domain, a study for integrating machine learning techniques in musculoskeletal ultrasonography images was conducted. The goal of this attempt was to investigate how feature extraction techniques, that capture higher-level information, perform in identifying human characteristics. The potential success of these techniques could lead to significant improvement of the current assessment methods-as the gray-scale image analysis-for distinguishing healthy and pathologic conditions, that are heavily dependent on the image-acquisition system. The contribution of this work is threefold. First, a new privately held data set of 74 healthy patients was presented. This data set included musculoskeletal ultrasound images from four muscles of the human body, namely the biceps brachii, tibialis anterior, gastrocnemius medialis and rectus femoris, recorded in the transverse and longitudinal plane. Second, two classification tasks were performed, namely, gender and muscle-type recognition, to assess the performance of the proposed method for successfully identifying differences in the texture of the examined muscle sections. Third, a novel method used with great success in the computer vision domain was presented, allowing the extraction of a high-level feature representation, by encoding the distribution of locally invariant texture descriptors. On the muscle-type recognition our method achieved an 87.07% classification rate, and on the task of gender recognition it surpassed state-of-the-art textural representations, reported in the literature in almost all the examined muscle sections.

Texture analysis of vertebral bone marrow using chemical shift encoding-based water-fat MRI: a feasibility study

This feasibility study investigated the spatial heterogeneity of the lumbar vertebral bone marrow using chemical shift encoding-based water-fat MRI. Acquired texture features like contrast and dissimilarity allowed for differentiation of pre- and postmenopausal women and may serve as imaging biomarkers in the future.

INTRODUCTION

While the vertebral bone marrow fat using chemical shift encoding water-fat magnetic resonance imaging (MRI) has been extensively studied, its spatial heterogeneity has not been analyzed yet. Therefore, this feasibility study investigated the spatial heterogeneity of the lumbar vertebral bone marrow by using texture analysis in proton density fat fraction (PDFF) maps.

METHODS

Forty-one healthy pre- and postmenopausal women were recruited for this study (premenopausal (n = 15) 30 ± 7 years, postmenopausal (n = 26) 65 ± 7 years). An eight-echo 3D spoiled gradient echo sequence was used for chemical shift encoding-based water-fat separation at the lumbar spine. Vertebral bodies L1 to L5 were manually segmented. Mean PDFF values and texture features were extracted at each vertebral level, namely variance, skewness, and kurtosis, using statistical moments and second-order features (energy, contrast, correlation, homogeneity, dissimilarity, entropy, variance, and sum average). Parameters were compared between pre- and postmenopausal women and vertebral levels.

RESULTS

PDFF was significantly higher in post- than in premenopausal women (49.37 ± 8.14% versus 27.76 ± 7.30%, p < 0.05). Furthermore, PDFF increased from L1 to L5 (L1 37.93 ± 12.85%, L2 38.81 ± 12.77%, L3 40.23 ± 12.72%, L4 42.80 ± 13.27%, L5 45.21 ± 14.55%, p < 0.05). Bone marrow heterogeneity based on texture analysis was significantly (p < 0.05) increased in postmenopausal women. Contrast and dissimilarity performed best in differentiating pre- and postmenopausal women (AUC = 0.97 and 0.96, respectively), not significantly different compared with PDFF (AUC = 0.97).

CONCLUSION

Conclusively, an increased bone marrow heterogeneity could be observed in postmenopausal women. In the future, texture parameters might provide additional information to detect and monitor vertebral bone marrow alterations due to aging or hormonal changes beyond conventional anatomic imaging.

False-positive reduction in computer-aided mass detection using mammographic texture analysis and classification

BACKGROUND AND OBJECTIVE

The aim of computer-aided-detection (CAD) systems for mammograms is to assist radiologists by marking region of interest (ROIs) depicting abnormalities. However, the confusing appearance of some normal tissues that visually look like masses results in a large proportion of marked ROIs with normal tissues. This paper copes with this problem and proposes a framework to reduce false positive masses detected by CAD.

METHODS

To avoid the error induced by the segmentation step, we proposed a segmentation-free framework with particular attention to improve feature extraction and classification steps. We investigated for the first time in mammogram analysis, Hilbert's image representation, Kolmogorov-Smirnov distance and maximum subregion descriptors. Then, a feature selection step is performed to select the most discriminative features. Moreover, we considered several classifiers such as Random Forest, Support Vector Machine and Decision Tree to distinguish between normal tissues and masses. Our experiments were carried out on a large dataset of 10168 ROIs (8254 normal tissues and 1914 masses) constructed from the Digital Database for Screening Mammography (DDSM). To simulate practical scenario, our normal regions are false positives asserted by a CAD system from healthy cases.

RESULTS

The combination of all the descriptors yields better results than each feature set used alone, and the difference is statistically significant. Besides, the feature selection steps yields a statistically significant increase in the accuracy values for the three classifiers. Finally, the random forest achieves the highest accuracy (81.09%), outperforming the SVM classifier (80.01%)) and decision tree (79.12%), but the difference is not statistically significant.

CONCLUSIONS

The accuracy of discrimination between normal and abnormal ROIs in mammograms obtained with the proposed gray level texture features sets are encouraging and comparable to these obtained with multiresolution features. Combination of several features as well as feature selection steps improve the results. To improve false positives reduction in CAD systems for breast cancer diagnosis, these features could be combined with multiresolution features.

A New Feature Ensemble with a Multistage Classification Scheme for Breast Cancer Diagnosis

A new and effective feature ensemble with a multistage classification is proposed to be implemented in a computer-aided diagnosis (CAD) system for breast cancer diagnosis. A publicly available mammogram image dataset collected during the Image Retrieval in Medical Applications (IRMA) project is utilized to verify the suggested feature ensemble and multistage classification. In achieving the CAD system, feature extraction is performed on the mammogram region of interest (ROI) images which are preprocessed by applying a histogram equalization followed by a nonlocal means filtering. The proposed feature ensemble is formed by concatenating the local configuration pattern-based, statistical, and frequency domain features. The classification process of these features is implemented in three cases: a one-stage study, a two-stage study, and a three-stage study. Eight well-known classifiers are used in all cases of this multistage classification scheme. Additionally, the results of the classifiers that provide the top three performances are combined via a majority voting technique to improve the recognition accuracy on both two- and three-stage studies. A maximum of 85.47%, 88.79%, and 93.52% classification accuracies are attained by the one-, two-, and three-stage studies, respectively. The proposed multistage classification scheme is more effective than the single-stage classification for breast cancer diagnosis.

Comparative Multifractal Analysis of Dynamic Infrared Thermograms and X-Ray Mammograms Enlightens Changes in the Environment of Malignant Tumors

There is growing evidence that the microenvironment surrounding a tumor plays a special role in cancer development and cancer therapeutic resistance. Tumors arise from the dysregulation and alteration of both the malignant cells and their environment. By providing tumor-repressing signals, the microenvironment can impose and sustain normal tissue architecture. Once tissue homeostasis is lost, the altered microenvironment can create a niche favoring the tumorigenic transformation process. A major challenge in early breast cancer diagnosis is thus to show that these physiological and architectural alterations can be detected with currently used screening techniques. In a recent study, we used a 1D wavelet-based multi-scale method to analyze breast skin temperature temporal fluctuations collected with an IR thermography camera in patients with breast cancer. This study reveals that the multifractal complexity of temperature fluctuations superimposed on cardiogenic and vasomotor perfusion oscillations observed in healthy breasts is lost in malignant tumor foci in cancerous breasts. Here we use a 2D wavelet-based multifractal method to analyze the spatial fluctuations of breast density in the X-ray mammograms of the same panel of patients. As compared to the long-range correlations and anti-correlations in roughness fluctuations, respectively observed in dense and fatty breast areas, some significant change in the nature of breast density fluctuations with some clear loss of correlations is detected in the neighborhood of malignant tumors. This attests to some architectural disorganization that may deeply affect heat transfer and related thermomechanics in breast tissues, corroborating the change to homogeneous monofractal temperature fluctuations recorded in cancerous breasts with the IR camera. These results open new perspectives in computer-aided methods to assist in early breast cancer diagnosis.

Detection of microcalcification with top-hat transform and the Gibbs random fields

Breast cancer is one of the most common causes of death in women aged 40 and above. Early detection of breast cancer has been one of the prime topics of research in biomedical engineering area. Micro-calcifications (MCs) are the indicators of early stages of breast cancer, and the detection of these MCs will, in turn, lead to diagnosis and treatment of breast cancer at its earliest stages. This paper proposes a new method to detect MCs in a digital mammogram. The approach starts with the segmentation of the digital mammogram to isolate the breast region, using fuzzy C means clustering algorithm. The segmented image is then further segmented using top-hat transform to localize the region of interest. A watershed transform is used to isolate the region of interest from rest of the image. The Gibbs random fields are employed to analyze the pixels in conjunction with the devised clique patterns and detect MCs in the image. A thresholding is performed on the processed image where the MCs are detected. The proposed algorithm is highly effective in reducing the region of interest to the region which has a high probability of finding a calcification or MC. It has an overall detection rate of 94.4% and accuracy of 88.2% with a false negative detection rate of 5.6%, respectively.

Detection and classification of Breast Cancer in Wavelet Sub-bands of Fractal Segmented Cancerous Zones

Recent studies on wavelet transform and fractal modeling applied on mammograms for the detection of cancerous tissues indicate that microcalcifications and masses can be utilized for the study of the morphology and diagnosis of cancerous cases. It is shown that the use of fractal modeling, as applied to a given image, can clearly discern cancerous zones from noncancerous areas. In this paper, for fractal modeling, the original image is first segmented into appropriate fractal boxes followed by identifying the fractal dimension of each windowed section using a computationally efficient two-dimensional box-counting algorithm. Furthermore, using appropriate wavelet sub-bands and image Reconstruction based on modified wavelet coefficients, it is shown that it is possible to arrive at enhanced features for detection of cancerous zones. In this paper, we have attempted to benefit from the advantages of both fractals and wavelets by introducing a new algorithm. By using a new algorithm named F1W2, the original image is first segmented into appropriate fractal boxes, and the fractal dimension of each windowed section is extracted. Following from that, by applying a maximum level threshold on fractal dimensions matrix, the best-segmented boxes are selected. In the next step, the segmented Cancerous zones which are candidates are then decomposed by utilizing standard orthogonal wavelet transform and db2 wavelet in three different resolution levels, and after nullifying wavelet coefficients of the image at the first scale and low frequency band of the third scale, the modified reconstructed image is successfully utilized for detection of breast cancer regions by applying an appropriate threshold. For detection of cancerous zones, our simulations indicate the accuracy of 90.9% for masses and 88.99% for microcalcifications detection results using the F1W2 method. For classification of detected mictocalcification into benign and malignant cases, eight features are identified and utilized in radial basis function neural network. Our simulation results indicate the accuracy of 92% classification using F1W2 method.

Breast cancer diagnosis in digitized mammograms using curvelet moments

BACKGROUND

Feature extraction is a key issue in designing a computer aided diagnosis system. Recent researches on breast cancer diagnosis have reported the effectiveness of multiscale transforms (wavelets and curvelets) for mammogram analysis and have shown the superiority of curvelet transform. However, the curse of dimensionality problem arises when using the curvelet coefficients and therefore a reduction method is required to extract a reduced set of discriminative features.

METHODS

This paper deals with this problem and proposes a feature extraction method based on curvelet transform and moment theory for mammogram description. First, we performed discrete curvelet transform and we computed the four first-order moments from curvelet coefficients distribution. Hence, two feature sets can be obtained: moments from each band and moments from each level. In this work, both sets are studied. Then, the t-test ranking technique was applied to select the best features from each set. Finally, a k-nearest neighbor classifier was used to distinguish between normal and abnormal breast tissues and to classify tumors as malignant or benign. Experiments were performed on 252 mammograms from the Mammographic Image Analysis Society (mini-MIAS) database using the leave-one-out cross validation as well as on 11553 mammograms from the Digital Database for Screening Mammography (DDSM) database using 2×5-fold cross validation.

RESULTS

Experimental results prove the effectiveness and the superiority of curvelet moments for mammogram analysis. Indeed, results on the mini-MIAS database show that curvelet moments yield an accuracy of 91.27% (resp. 81.35 %) with 10 (resp. 8) features for abnormality (resp. malignancy) detection. In addition, empirical comparisons of the proposed method against state-of-the-art curvelet-based methods on the DDSM database show that the suggested method does not only lead to a more reduced feature set, but it also statistically outperforms all the compared methods in terms of accuracy.

CONCLUSIONS

In summary, curvelet moments are an efficient and effective way to extract a reduced set of discriminative features for breast cancer diagnosis.

Multi-scale textural feature extraction and particle swarm optimization based model selection for false positive reduction in mammography

The high number of false positives and the resulting number of avoidable breast biopsies are the major problems faced by current mammography Computer Aided Detection (CAD) systems. False positive reduction is not only a requirement for mass but also for calcification CAD systems which are currently deployed for clinical use. This paper tackles two problems related to reducing the number of false positives in the detection of all lesions and masses, respectively. Firstly, textural patterns of breast tissue have been analyzed using several multi-scale textural descriptors based on wavelet and gray level co-occurrence matrix. The second problem addressed in this paper is the parameter selection and performance optimization. For this, we adopt a model selection procedure based on Particle Swarm Optimization (PSO) for selecting the most discriminative textural features and for strengthening the generalization capacity of the supervised learning stage based on a Support Vector Machine (SVM) classifier. For evaluating the proposed methods, two sets of suspicious mammogram regions have been used. The first one, obtained from Digital Database for Screening Mammography (DDSM), contains 1494 regions (1000 normal and 494 abnormal samples). The second set of suspicious regions was obtained from database of Mammographic Image Analysis Society (mini-MIAS) and contains 315 (207 normal and 108 abnormal) samples. Results from both datasets demonstrate the efficiency of using PSO based model selection for optimizing both classifier hyper-parameters and parameters, respectively. Furthermore, the obtained results indicate the promising performance of the proposed textural features and more specifically, those based on co-occurrence matrix of wavelet image representation technique.

Analysis of perceived similarity between pairs of microcalcification clusters in mammograms

PURPOSE

Content-based image retrieval aims to assist radiologists by presenting example images with known pathology that are visually similar to the case being evaluated. In this work, the authors investigate several fundamental issues underlying the similarity ratings between pairs of microcalcification (MC) lesions on mammograms as judged by radiologists: the degree of variability in the similarity ratings, the impact of this variability on agreement between readers in retrieval of similar lesions, and the factors contributing to the readers' similarity ratings.

METHODS

The authors conduct a reader study on a set of 1000 image pairs of MC lesions, in which a group of experienced breast radiologists rated the degree of similarity between each image pair. The image pairs are selected, from among possible pairings of 222 cases (110 malignant, 112 benign), based on quantitative image attributes (features) and the results of a preliminary reader study. Next, the authors apply analysis of variance (ANOVA) to quantify the level of variability in the readers' similarity ratings, and study how the variability in individual reader ratings affects consistency between readers. The authors also measure the extent to which readers agree on images which are most similar to a given query, for which the Dice coefficient is used. To investigate how the similarity ratings potentially relate to the attributes underlying the cases, the authors study the fraction of perceptually similar images that also share the same benign or malignant pathology as the query image; moreover, the authors apply multidimensional scaling (MDS) to embed the cases according to their mutual perceptual similarity in a two-dimensional plot, which allows the authors to examine the manner in which similar lesions relate to one another in terms of benign or malignant pathology and clustered MCs.

RESULTS

The ANOVA results show that the coefficient of determination in the reader similarity ratings is 0.59. The variability level in the similarity ratings is proved to be a limiting factor, leading to only moderate correlation between the readers in their readings. The Dice coefficient, measuring agreement between readers in retrieval of similar images, can vary from 0.45 to 0.64 with different levels of similarity for individual readers, but is higher for average ratings from a group of readers (from 0.59 to 0.78). More importantly, the fraction of retrieved cases that match the benign or malignant pathology of the query image was found to increase with the degree of similarity among the retrieved images, reaching average value as high as 0.69 for the radiologists (p-value <10(-4) compared to random guessing). Moreover, MDS embedding of all the cases shows that cases having the same pathology tend to cluster together, and that neighboring cases in the plot tend to be similar in their clustered MCs.

CONCLUSIONS

While individual readers exhibit substantial variability in their similarity ratings, similarity ratings averaged from a group of readers can achieve a high level of intergroup consistency and agreement in retrieval of similar images. More importantly, perceptually similar cases are also likely to be similar in their underlying benign or malignant pathology and image features of clustered MCs, which could be of diagnostic value in computer-aided diagnosis for lesions with clustered MCs.

Wavelet-based 3D reconstruction of microcalcification clusters from two mammographic views: new evidence that fractal tumors are malignant and Euclidean tumors are benign

The 2D Wavelet-Transform Modulus Maxima (WTMM) method was used to detect microcalcifications (MC) in human breast tissue seen in mammograms and to characterize the fractal geometry of benign and malignant MC clusters. This was done in the context of a preliminary analysis of a small dataset, via a novel way to partition the wavelet-transform space-scale skeleton. For the first time, the estimated 3D fractal structure of a breast lesion was inferred by pairing the information from two separate 2D projected mammographic views of the same breast, i.e. the cranial-caudal (CC) and mediolateral-oblique (MLO) views. As a novelty, we define the “CC-MLO fractal dimension plot”, where a “fractal zone” and “Euclidean zones” (non-fractal) are defined. 118 images (59 cases, 25 malignant and 34 benign) obtained from a digital databank of mammograms with known radiologist diagnostics were analyzed to determine which cases would be plotted in the fractal zone and which cases would fall in the Euclidean zones. 92% of malignant breast lesions studied (23 out of 25 cases) were in the fractal zone while 88% of the benign lesions were in the Euclidean zones (30 out of 34 cases). Furthermore, a Bayesian statistical analysis shows that, with 95% credibility, the probability that fractal breast lesions are malignant is between 74% and 98%. Alternatively, with 95% credibility, the probability that Euclidean breast lesions are benign is between 76% and 96%. These results support the notion that the fractal structure of malignant tumors is more likely to be associated with an invasive behavior into the surrounding tissue compared to the less invasive, Euclidean structure of benign tumors. Finally, based on indirect 3D reconstructions from the 2D views, we conjecture that all breast tumors considered in this study, benign and malignant, fractal or Euclidean, restrict their growth to 2-dimensional manifolds within the breast tissue.

Computer-aided detection of breast cancer on mammograms: a swarm intelligence optimized wavelet neural network approach

Breast cancer is the second leading cause of cancer death in women. Accurate early detection can effectively reduce the mortality rate caused by breast cancer. Masses and microcalcification clusters are an important early signs of breast cancer. However, it is often difficult to distinguish abnormalities from normal breast tissues because of their subtle appearance and ambiguous margins. Computer aided diagnosis (CAD) helps the radiologist in detecting the abnormalities in an efficient way. This paper investigates a new classification approach for detection of breast abnormalities in digital mammograms using Particle Swarm Optimized Wavelet Neural Network (PSOWNN). The proposed abnormality detection algorithm is based on extracting Laws Texture Energy Measures from the mammograms and classifying the suspicious regions by applying a pattern classifier. The method is applied to real clinical database of 216 mammograms collected from mammogram screening centers. The detection performance of the CAD system is analyzed using Receiver Operating Characteristic (ROC) curve. This curve indicates the trade-offs between sensitivity and specificity that is available from a diagnostic system, and thus describes the inherent discrimination capacity of the proposed system. The result shows that the area under the ROC curve of the proposed algorithm is 0.96853 with a sensitivity 94.167% of and specificity of 92.105%.

A Novel Approach for Detecting and Classifying Breast Cancer in Mammogram Images

This study uses data mining techniques for computer-aided diagnosis that involves the feature extraction for cancer detection, so as to help doctors towards making optimal decisions quickly and accurately. Features play an important role in detecting the cancer in the digital mammogram and feature extraction stage is the most vital and difficult stage. In this paper, an enhanced feature extraction method named Multiscale Surrounding Region Dependence Method (MSRDM) is proposed to be effective in classifying the mammogram images into normal or benign or malignant. This proposed system is based on a four-step procedure: Regions of Interest specification, two dimensional discrete wavelet transformation, and multiscale surrounding region dependence matrix computation and feature extraction. The performance of the proposed feature set is compared with the conventional texture-analysis methods such as gray level cooccurence matrix features and surrounding region dependence method features. Experiments have been conducted on both real and benchmark data and the results have been proved to be progressive.

Texture and moments-based classification of the acrosome integrity of boar spermatozoa images

The automated assessment of the sperm quality is an important challenge in the veterinary field. In this paper, we explore how to describe the acrosomes of boar spermatozoa using image analysis so that they can be automatically categorized as intact or damaged. Our proposal aims at characterizing the acrosomes by means of texture features. The texture is described using first order statistics and features derived from the co-occurrence matrix of the image, both computed from the original image and from the coefficients yielded by the Discrete Wavelet Transform. Texture descriptors are evaluated and compared with moments-based descriptors in terms of the classification accuracy they provide. Experimental results with a Multilayer Perceptron and the k-Nearest Neighbours classifiers show that texture descriptors outperform moment-based descriptors, reaching an accuracy of 94.93%, which makes this approach very attractive for the veterinarian community.

Rapid image stitching and computer-aided detection for multipass automated breast ultrasound

PURPOSE

Breast ultrasound (US) is recently becoming more and more popular for detecting breast lesions. However, screening results in hundreds of US images for each subject. This magnitude of images can lead to fatigue in radiologist, causing failure in the detection of lesions of a subtle nature. In this study, an image stitching technique is proposed for combining multipass images of the whole breast into a series of full-view images, and a fully automatic screening system that works off these images is also presented.

METHODS

Using the registration technique based on the simple sum of absolute block-mean difference (SBMD) measure, three-pass images were merged into full-view US images. An automatic screening system was then developed for detecting tumors from these full-view images. The preprocessing step was used to reduce the tumor detection time of the system and to improve image quality. The gray-level slicing method was then used to divide images into numerous regions. Finally, seven computerized features--darkness, uniformity, width-height ratio, area size, nonpersistence, coronal area size, and region continuity--were defined and used to determine whether or not each region was a part of a tumor.

RESULTS

In the experiment, there was a total of 25 experimental cases with 26 lesions, and each case was composed of 252 images (three passes, 84 images/pass). The processing time of the proposed stitching procedure for each case was within 30 s with a Pentium IV 2.0 processor, and the detection sensitivity of the proposed CAD system was 92.3% with 1.76 false positives per case.

CONCLUSIONS

The proposed automatic screening system can be applied to the whole breast images stitched together via SBMD-based registration in order to detect tumors.

Characterization of breast cancer types by texture analysis of magnetic resonance images

RATIONALE AND OBJECTIVES

This novel study aims to investigate texture parameters in distinguishing healthy breast tissue and breast cancer in breast magnetic resonance imaging (MRI). A specific aim was to identify possible differences in the texture characteristics of histological types (lobular and ductal) of invasive breast cancer and to determine the value of these differences for computer-assisted lesion classification.

MATERIALS AND METHODS

Twenty patients (mean age 50.6 + or - SD 10.6; range 37-70 years), with histopathologically proven invasive breast cancer (10 lobular and 10 ductal) were included in this preliminary study. The median MRI lesion size was 25 mm (range, 7-60 mm). The selected T1-weighted precontrast, post-contrast, and subtracted images were analyzed and classified with texture analysis (TA) software MaZda and additional statistical tests were used for testing the parameters separability.

RESULTS

All classification methods employed were able to differentiate between cancer and healthy breast tissue and also invasive lobular and ductal carcinoma with classification accuracy varying between 80% and 100%, depending on the used imaging series and the type of region of interest. We found several parameters to be significantly different between the regions of interest studied. The co-occurrence matrix based parameters proved to be superior to other texture parameters used.

CONCLUSIONS

The results of this study indicate that MRI TA differentiates breast cancer from normal tissue and may be able to distinguish between two histological types of breast cancer providing more accurate characterization of breast lesions thereby offering a new tool for radiological analysis of breast MRI.