Search

Introduction: Non-invasive methods that can characterize a tumor's molecular features could enhance treatment management. Recently, researchers have investigated tumor heterogeneity on imaging to assess how grainy or coarse a tumor seems to be in the search for oncologic prognostic markers and mechanisms. Quantitative computed tomography (CT) based texture analysis (qCTTA) has been used to derive tumor heterogeneity information, and the appearance of the tumors has been shown to relate to patient outcome in esophageal, colorectal, lung and head and neck cancer and treatment response in metastatic renal cell cancer. Furthermore histological assessment has demonstrated an association between qCTTA and hypoxia and angiogenesis in lung cancer and very recently qCTTA in combination with CT blood-flow and PET glucose-uptake identified an imaging signature for K-ras mutation status in colorectal cancer. In this study, we examined the predictive potential of tumoral qCTTA to differentiate K-ras mutant and K-ras wildtype tumors and its prognostic potential using non-contrast CT imaging in non-small cell lung cancer (NSCLC). Methods: Formalin-fixed, paraffin-embedded (FFPE) lung tumor tissues from patients with stage I and II NSCLC, diagnosed between 2001 and 2007 and undergoing definitive surgical resection were obtained. Tumor DNA was extracted and analyzed on the LungCarta Panel that interrogates 213 somatic mutations in 26 oncogenes and/or tumor suppressor genes. Cases with a K-ras mutation or pan-wildtype for all 26 oncogenes and tumor suppressor genes were selected for imaging analysis. Baseline pre-treatment non-contrast CT images that were performed as part of standard of care were retrieved for qCTTA. qCTTA was applied to regions of interest in the primary tumor and comprised an image filtration-histogram technique; where the filtration technique enhanced features of different sizes based on the spatial scale filter (SSF) value varying from fine (2mm in radius), medium (3-5mm in radius) and coarse (6mm in radius) followed by quantification of the histogram parameters – kurtosis (a measure of peakedness and tailedness) and skewness (a measure of asymmetry of the histogram). Non-parametric Mann Whitney test assessed the ability of the qCTTA, clinical and patient characteristics to differentiate between K-ras mutation from pan-wildtype whereas Kaplan-Meier survival analysis assessed the ability of these markers including K-ras mutation to predict survival. Results: 48 cases were identified and qCTTA was applied to pre-treatment non-contrast CT. The median age was 70.4 years (range 45.1-85.1), 29 were men, and 46 had a smoking history. There were 33 adenocarcinomas, 10 squamous cell carcinomas, and 5 other NSCLC. Stages IA, IB, IIA, and IIB were 16, 21, 6, and 5; respectively. At least 33 patients did not receive adjuvant chemotherapy. There were 27 cases with K-ras mutation and 21 cases that were pan-wildtype. As of the last follow-up, 18 had confirmed disease relapse, including 5 with brain metastasis, and 22 were deceased. The median disease-free survival (DFS) was 39.7 months and median overall survival (OS) was 45.0 months. Positive skewness (p=0.031) and lower kurtosis (p=0.009) were significantly associated with the presence of a K-ras mutation (n=27 vs. 21 pan-wildtype). Kurtosis was a significant predictor of OS (p=0.009) and DFS (p=0.04) with a lower kurtosis value linked with poorer survival. As expected, patients with disease relapse and older age had significantly inferior OS and DFS, whereas adenocarcinomas (n=30) had inferior DFS vs. non-adenocarcinomas (n=14)(p=0.031). In line with other published data in early-stage NSCLC, K-ras was not associated with OS or DFS. Conclusions: The outcomes in this dataset are in sync with published poor clinical prognostic features in early-stage NSCLC. Lower kurtosis (which may reflect increased number of highlighted features and intensity variation in highlighted features) and positive skewness (which may reflect bright highlighted features surrounded by darker background or lower density areas) are significantly associated with K-ras mutations. These features may suggest more tumoral necrosis and these may be due to the hostile microenvironment within a K-ras driven tumor. A qCTTA feature such as lower kurtosis is prognostic for poorer OS and DFS. Non-invasive qCTTA can differentiate the presence of K-ras mutation from pan-wildtype NSCLC and is associated with patient survival. Citation Format: Glen J. Weiss, Balaji Ganeshan, Kenneth A. Miles, David A. Campbell, Philip Y. Cheung, Samuel Frank, Ronald L. Korn. Noninvasive image texture analysis differentiates K-ras mutation from pan-wildtype NSCLC and is prognostic. [abstract]. In: Proceedings of the AACR-IASLC Joint Conference on Molecular Origins of Lung Cancer; 2014 Jan 6-9; San Diego, CA. Philadelphia (PA): AACR; Clin Cancer Res 2014;20(2Suppl):Abstract nr A34.

The classification of benign and malignant masses in mammograms by Computer-Aided Diagnosis (CAD) is one of the most difficult and important tasks in the development of CAD systems. This classification has commonly been automated by extracting a set of handcrafted features from mammograms and relating the responses to breast cancer. Recently, the application of Deep Learning (DL) technology in medical imaging informatics has been attracting extensive research interest. However, limited medical image datasets and feature expression often reduce the performance of DL-based schemes. Therefore, this study aims to develop a new combined feature CAD method based on DL for classifying mammographic masses into three classes: normal, benign and cancer (malignant) masses. Three kinds of breast masses were scored by using Deep Convolution Neural Network (DCNN) as a feature extractor. Then the scoring features are combined with the image texture features as input to the classifier. This features including the scoring features, Gray-Level Co-occurrence Matrix (GLCM) and Histogram of Oriented Gradient (HOT) were employed to extract the breast mass information in mammograms and the classifier of Support Vector Machine (SVM) and Extreme Gradient Boosting (XGBoost) were trained for the classification task. Accuracy (ACC), Precision (Pre), Recall (Rec), F1-score (F1), and Overall Accuracy (Overall ACC) are used to evaluate the performance of the proposed system and the results show that the proposed multi-features combination model performs the best results. The performance of the XGBoost classifier has proved to be better in comparison to the SVM classification algorithms. As a result, when XGBoost was used as a classifier, the correct identification rate of the Overall ACC was 92.80% and that of malignant tumors was 84%, with reasonable and best results. These results indicate that the proposed method may help in more accurately diagnosing cases that are difficult to classify on images.

Breast cancer is a known heterogeneous disease. Current clinically utilized histopathologic biomarkers may undersample tumor heterogeneity, resulting in higher rates of misdiagnosis for breast cancer. MRI can provide a whole-tumor sampling of disease burden and is widely utilized in clinical care. Texture analysis can provide a localized description of breast cancer, with particular emphasis on quantifying breast lesion heterogeneity. The object of this review is to provide an overview of texture analysis applications towards breast cancer diagnosis, prognosis, and treatment response evaluation and review the role of image-based texture features as noninvasive prognostic and predictive biomarkers. Level of Evidence: 5 Technical Efficacy: Stage 2 J. Magn. Reson. Imaging 2019;49:927-938.

A methodology to texture analysis of masses in digitized mammography is proposed.Our methodology uses only the texture analysis for recognition.We provide the specialist bigger support to the diagnosis of breast cancer.We cooperate to a more precise diagnosis and support in the medical intervention. A World Health Organization (WHO) report estimates that in 2015, at least 561 thousand women will die of breast cancer. Although breast cancer is considered a disease of the developed world, nearly 50% of the cases and 58% of the deaths occur in the less developed countries. A mammogram is a way of discovering not just the palpable tumors that cause cancer but also other lesions that are not perceived during the physical examination performed by the expert physician or during self-exams; however, it is known that this exam is targeted for women after the age of 40 because age is one of the factors that can cause great variations in sensitivity during the exam. Besides the patient's age, the expert's experience and the quality of the images obtained during the exam are decisive factors in the detection of breast cancer. This work presents two novelties. The first is the use of Local Binary Patterns (LBPs) to generate a representation of a Region of Interest (ROI) image. Over this representation, we generate other representations using techniques such as image histograms, gray-level co-occurrence matrices (GLCMs) and gray-level run-length matrices (GLRLMs). These representations allow texture analysis through several perspectives. The second novelty uses these representations as input to the application of indexes adapted from ecology (Shannon, McIntosh, Simpson, Gleason and Menhinick) as texture descriptors. Based on this strategy, we analyze mammographic image textures to classify regions of these images as benign or malignant using a Support Vector Machine (SVM). The best result achieved was of 88.31% accuracy, 85% sensitivity, 91.89% specificity, a positive probability ratio of 10.48, a negative probability ratio of 0.16, and an area under the Receiver Operating Characteristic (ROC) curve of 0.88, obtained through the Shannon index. We believe that the proposed method, with some adaptations, may also be used for image texture analysis of several different lesions such as lung nodules, glaucoma and prostates. This belief is based on the achieved results and the method's simplicity.

Image-guided liver surgery aims to enhance the precision of resection and ablation by providing fast localisation of tumours and adjacent complex vasculature to improve oncologic outcome. This Letter presents a novel end-to-end solution for fast stereo reconstruction and motion estimation that demonstrates high accuracy with phantom and clinical data. The authors' computationally efficient coarse-to-fine (CTF) stereo approach facilitates liver imaging by accounting for low texture regions, enabling precise three-dimensional (3D) boundary recovery through the use of adaptive windows and utilising a robust 3D motion estimator to reject spurious data. To the best of their knowledge, theirs is the only adaptive CTF matching approach to reconstruction and motion estimation that registers time series of reconstructions to a single key frame for registration to a volumetric computed tomography scan. The system is evaluated empirically in controlled laboratory experiments with a liver phantom and motorised stages for precise quantitative evaluation. Additional evaluation is provided through testing with patient data during liver resection.

Accurate optic disc (OD) segmentation is an important step in obtaining cup-to-disc ratio-based glaucoma screening using fundus imaging. It is a challenging task because of the subtle OD boundary, blood vessel occlusion and intensity inhomogeneity. In this Letter, the authors propose an improved version of the random walk algorithm for OD segmentation to tackle such challenges. The algorithm incorporates the mean curvature and Gabor texture energy features to define the new composite weight function to compute the edge weights. Unlike the deformable model-based OD segmentation techniques, the proposed algorithm remains unaffected by curve initialisation and local energy minima problem. The effectiveness of the proposed method is verified with DRIVE, DIARETDB1, DRISHTI-GS and MESSIDOR database images using the performance measures such as mean absolute distance, overlapping ratio, dice coefficient, sensitivity, specificity and precision. The obtained OD segmentation results and quantitative performance measures show robustness and superiority of the proposed algorithm in handling the complex challenges in OD segmentation.

Detection, segmentation and classification of nuclei are fundamental analysis operations in digital pathology. Existing state-of-the-art approaches demand extensive amount of supervised training data from pathologists and may still perform poorly in images from unseen tissue types. We propose an unsupervised approach for histopathology image segmentation that synthesizes heterogeneous sets of training image patches, of every tissue type. Although our synthetic patches are not always of high quality, we harness the motley crew of generated samples through a generally applicable importance sampling method. This proposed approach, for the first time, re-weighs the training loss over synthetic data so that the ideal (unbiased) generalization loss over the true data distribution is minimized. This enables us to use a random polygon generator to synthesize approximate cellular structures (i.e., nuclear masks) for which no real examples are given in many tissue types, and hence, GAN-based methods are not suited. In addition, we propose a hybrid synthesis pipeline that utilizes textures in real histopathology patches and GAN models, to tackle heterogeneity in tissue textures. Compared with existing state-of-the-art supervised models, our approach generalizes significantly better on cancer types without training data. Even in cancer types with training data, our approach achieves the same performance without supervision cost. We release code and segmentation results() on over 5000 Whole Slide Images (WSI) in The Cancer Genome Atlas (TCGA) repository, a dataset that would be orders of magnitude larger than what is available today.

Purpose: Mammographic percent density (PD%) is known to be a strong risk factor for breast cancer. Recent studies also suggest that parenchymal texture features, which are more granular descriptors of the parenchymal pattern, can provide additional information about breast cancer risk. To date, most studies have measured mammographic texture within selected regions of interest (ROIs) in the breast, which cannot adequately capture the complexity of the parenchymal pattern throughout the whole breast. To better characterize patterns of the parenchymal tissue, the authors have developed a fully automated software pipeline based on a novel lattice-based strategy to extract a range of parenchymal texture features from the entire breast region. Methods: Digital mammograms from 106 cases with 318 age-matched controls were retrospectively analyzed. The lattice-based approach is based on a regular grid virtually overlaid on each mammographic image. Texture features are computed from the intersection (i.e., lattice) points of the grid lines within the breast, using a local window centered at each lattice point. Using this strategy, a range of statistical (gray-level histogram, co-occurrence, and run-length) and structural (edge-enhancing, local binary pattern, and fractal dimension) features are extracted. To cover the entire breast, the size of the local window for feature extraction is set equal to the lattice grid spacing and optimized experimentally by evaluating different windows sizes. The association between their lattice-based texture features and breast cancer was evaluated using logistic regression with leave-one-out cross validation and further compared to that of breast PD% and commonly used single-ROI texture features extracted from the retroareolar or the central breast region. Classification performance was evaluated using the area under the curve (AUC) of the receiver operating characteristic (ROC). DeLong’s test was used to compare the different ROCs in terms of AUC performance. Results: The average univariate performance of the lattice-based features is higher when extracted from smaller than larger window sizes. While not every individual texture feature is superior to breast PD% (AUC: 0.59, STD: 0.03), their combination in multivariate analysis has significantly better performance (AUC: 0.85, STD: 0.02, p 0.05). Conclusions: The proposed lattice-based strategy for mammographic texture analysis enables to characterize the parenchymal pattern over the entire breast. As such, these features provide richer information compared to currently used descriptors and may ultimately improve breast cancer risk assessment. Larger studies are warranted to validate these findings and also compare to standard demographic and reproductive risk factors.

Background: To stratify risks of pancreatic adenocarcinoma (PA) patients using pre- and post-radiotherapy (RT) PET/CT images, and to assess the prognostic value of texture variations in predicting therapy response of patients. Methods: Twenty-six PA patients treated with RT from 2011–2013 with pre- and post-treatment 18F-FDG-PET/CT scans were identified. Tumor locoregional texture was calculated using 3D kernel-based approach, and texture variations were identified by fitting discrepancies of texture maps of pre- and post-treatment images. A total of 48 texture and clinical variables were identified and evaluated for association with overall survival (OS). The prognostic heterogeneity features were selected using lasso/elastic net regression, and further were evaluated by multivariate Cox analysis. Results: Median age was 69 y (range, 46–86 y). The texture map and temporal variations between pre- and post-treatment were well characterized by histograms and statistical fitting. The lasso analysis identified seven predictors (age, node stage, post-RT SUVmax, variations of homogeneity, variance, sum mean, and cluster tendency). The multivariate Cox analysis identified five significant variables: age, node stage, variations of homogeneity, variance, and cluster tendency (with P=0.020, 0.040, 0.065, 0.078, and 0.081, respectively). The patients were stratified into two groups based on the risk score of multivariate analysis with log-rank P=0.001: a low risk group (n=11) with a longer mean OS (29.3 months) and higher texture variation (>30%), and a high risk group (n=15) with a shorter mean OS (17.7 months) and lower texture variation ( Conclusions: Locoregional metabolic texture response provides a feasible approach for evaluating and predicting clinical outcomes following treatment of PA with RT. The proposed method can be used to stratify patient risk and help select appropriate treatment strategies for individual patients toward implementing response-driven adaptive RT.

Cataloged from PDF version of article. The histopathological examination of tissue specimens is essential for cancer diagnosis and grading. However, this examination is subject to a considerable amount of observer variability as it mainly relies on visual interpretation of pathologists. To alleviate this problem, it is very important to develop computational quantitative tools, for which image segmentation constitutes the core step. In this paper, we introduce an effective and robust algorithm for the segmentation of histopathological tissue images. This algorithm incorporates the background knowledge of the tissue organization into segmentation. For this purpose, it quantifies spatial relations of cytological tissue components by constructing a graph and uses this graph to define new texture features for image segmentation. This new texture definition makes use of the idea of gray-level run-length matrices. However, it considers the runs of cytological components on a graph to form a matrix, instead of considering the runs of pixel intensities. Working with colon tissue images, our experiments demonstrate that the texture features extracted from "graph run-length matrices" lead to high segmentation accuracies, also providing a reasonable number of segmented regions. Compared with four other segmentation algorithms, the results show that the proposed algorithm is more effective in histopathological image segmentation

Breast cancer is the most common type of cancer among women in the world. Mammography is regarded as an effective tool for early detection and diagnosis of breast cancer. Microcalcification is one of the primary signs of breast cancer. There are various image texture analysis techniques for the detection of the microcalcifications. Screenfilm mammography is still the standard method used to detect early breast cancer, thus leading to early treatment. Digital mammography has recently been designated as the imaging technology with the greatest potential for improving the diagnosis of breast cancer. In this work a feature-based approach is used for analysis and classification of malignancy. Gray-level texture and Wavelet coefficient texture methods are used for feature extraction. Probabilistic Neural Network (PNN) is used for classification of images based on extracted features. The performance of classification by PNN based on features by texture method, wavelet method and combined methods are compared. The Receiver Operating Characteristics (ROC) Analysis is used for performance evaluation.

Lung needle biopsy image classification is a critical task for computer-aided lung cancer diagnosis. In this study, a novel method, multimodal sparse representation-based classification (mSRC), is proposed for classifying lung needle biopsy images. In the data acquisition procedure of our method, the cell nuclei are automatically segmented from the images captured by needle biopsy specimens. Then, features of three modalities (shape, color, and texture) are extracted from the segmented cell nuclei. After this procedure, mSRC goes through a training phase and a testing phase. In the training phase, three discriminative subdictionaries corresponding to the shape, color, and texture information are jointly learned by a genetic algorithm guided multimodal dictionary learning approach. The dictionary learning aims to select the topmost discriminative samples and encourage large disagreement among different subdictionaries. In the testing phase, when a new image comes, a hierarchical fusion strategy is applied, which first predicts the labels of the cell nuclei by fusing three modalities, then predicts the label of the image by majority voting. Our method is evaluated on a real image set of 4372 cell nuclei regions segmented from 271 images. These cell nuclei regions can be divided into five classes: four cancerous classes (corresponding to four types of lung cancer) plus one normal class (no cancer). The results demonstrate that the multimodal information is important for lung needle biopsy image classification. Moreover, compared to several state-of-the-art methods (LapRLS, MCMI-AB, mcSVM, ESRC, KSRC), the proposed mSRC can achieve significant improvement (mean accuracy of 88.1%, precision of 85.2%, recall of 92.8%, etc.), especially for classifying different cancerous types.

The authors propose a spatiotemporal enhancement pattern (STEP) for comprehensive characterization of breast tumors in contrast-enhanced MR images. By viewing serial contrast-enhanced MR images as a single spatiotemporal image, they formulate the STEP as a combination of (1) dynamic enhancement and architectural features of a tumor, and (2) the spatial variations of pixelwise temporal enhancements. Although the latter has been widely used by radiologists for diagnostic purposes, it has rarely been employed for computer-aided diagnosis. This article presents two major contributions. First, the STEP features are introduced to capture temporal enhancement and its spatial variations. This is essentially carried out through the Fourier transformation and pharmacokinetic modeling of various temporal enhancement features, followed by the calculation of moment invariants and Gabor texture features. Second, for effectively extracting the STEP features from tumors, we develop a graph-cut based segmentation algorithm that aims at refining coarse manual segmentations of tumors. The STEP features are assessed through their diagnostic performance for differentiating between benign and malignant tumors using a linear classifier (along with a simple ranking-based feature selection) in a leave-one-out cross-validation setting. The experimental results for the proposed features exhibit superior performance, when compared to the existing approaches, with the area under the ROC curve approaching 0.97.

In this article a comprehensive set of registration methods is utilized to provide image-to-physical space registration for image-guided neurosurgery in a clinical study. Central to all methods is the use of textured point clouds as provided by laser range scanning technology. The objective is to perform a systematic comparison of registration methods that include both extracranial (skin marker point-based registration (PBR), and face-based surface registration) and intracranial methods (feature PBR, cortical vessel-contour registration, a combined geometry∕intensity surface registration method, and a constrained form of that method to improve robustness). The platform facilitates the selection of discrete soft-tissue landmarks that appear on the patient’s intraoperative cortical surface and the preoperative gadolinium-enhanced magnetic resonance (MR) image volume, i.e., true corresponding novel targets. In an 11 patient study, data were taken to allow statistical comparison among registration methods within the context of registration error. The results indicate that intraoperative face-based surface registration is statistically equivalent to traditional skin marker registration. The four intracranial registration methods were investigated and the results demonstrated a target registration error of 1.6±0.5 mm, 1.7±0.5 mm, 3.9±3.4 mm, and 2.0±0.9 mm, for feature PBR, cortical vessel-contour registration, unconstrained geometric∕intensity registration, and constrained geometric∕intensity registration, respectively. When analyzing the results on a per case basis, the constrained geometric∕intensity registration performed best, followed by feature PBR, and finally cortical vessel-contour registration. Interestingly, the best target registration errors are similar to targeting errors reported using bone-implanted markers within the context of rigid targets. The experience in this study as with others is that brain shift can compromise extracranial registration methods from the earliest stages. Based on the results reported here, organ-based approaches to registration would improve this, especially for shallow lesions.

Malignant breast tumors typically appear in mammograms with rough, spiculated, or microlobulated contours, whereas most benign masses have smooth, round, oval, or macrolobulated contours. Several studies have shown that shape factors that incorporate differences as above can provide high accuracies in distinguishing between malignant tumors and benign masses based upon their contours only. However, global measures of roughness, such as compactness, are less effective than specially designed features based upon spicularity and concavity. We propose a method to derive polygonal models of contours that preserve spicules and details of diagnostic importance. We show that an index of spiculation derived from the turning functions of the polygonal models obtained by the proposed method yields better classification accuracy than a similar measure derived using a previously published method. The methods were tested with a set of 111 contours of 65 benign masses and 46 malignant tumors. A high classification accuracy of 0.94 in terms of the area under the receiver operating characteristics curve was obtained.

In this study, our purpose was to improve the performance of our mass detection system by using a new dual system approach which combines a computer-added detection (CAD) system optimized with "average" masses with another CAD system optimized with "subtle" masses. The two single CAD systems have similar image processing steps, which include prescreening, object segmentation, morphological and texture feature extraction, and false positive (FP) reduction by rule-based and linear discriminant analysis (LDA) classifiers. A feed-forward backpropagation artificial neural network was trained to merge the scores from the LDA classifiers in the two single CAD systems and differentiate true masses from normal tissue. For an unknown test mammogram, the two single CAD systems are applied to the image in parallel to detect suspicious objects. A total of three data sets were used for training and testing the systems. The first data set of 230 current mammograms, referred to as the average mass set, was collected from 115 patients. We also collected 264 mammograms, referred to as the subtle mass set, which were one to two years prior to the current exam from these patients. Both the average and the subtle mass sets were partitioned into two independent data sets in a cross validation training and testing scheme. A third data set containing 65 cases with 260 normal mammograms was used to estimate the FP marker rates during testing. When the single CAD system trained on the average mass set was applied to the test set with average masses, the FP marker rates were 2.2, 1.8, and 1.5 per image at the case-based sensitivities of 90%, 85%, and 80%, respectively. With the dual CAD system, the FP marker rates were reduced to 1.2, 0.9, and 0.7 per image, respectively, at the same case-based sensitivities. Statistically significant (p < 0.05) improvements on the free response receiver operating characteristic curves were observed when the dual system and the single system were compared using the test sets with either average masses or subtle masses.

We propose a method for the detection of masses in mammographic images that employs Gaussian smoothing and subsampling operations as preprocessing steps. The mass portions are segmented by establishing intensity links from the central portions of masses into the surrounding areas. We introduce methods for analyzing oriented flow-like textural information in mammograms. Features based on flow orientation in adaptive ribbons of pixels across the margins of masses are proposed to classify the regions detected as true mass regions or false-positives (FPs). The methods yielded a mass versus normal tissue classification accuracy represented as an area (A/sub z/) of 0.87 under the receiver operating characteristics (ROCs) curve with a dataset of 56 images including 30 benign disease, 13 malignant disease, and 13 normal cases selected from the mini Mammographic Image Analysis Society database. A sensitivity of 81% was achieved at 2.2 FPs/image. Malignant tumor versus normal tissue classification resulted in a higher A/sub z/ value of 0.9 under the ROC curve using only the 13 malignant and 13 normal cases with a sensitivity of 85% at 2.45 FPs/image. The mass detection algorithm could detect all the 13 malignant tumors successfully, but achieved a success rate of only 63% (19/30) in detecting the benign masses. The mass regions that were successfully segmented were further classified as benign or malignant disease by computing five texture features based on gray-level co-occurrence matrices (GCMs) and using the features in a logistic regression method. The features were computed using adaptive ribbons of pixels across the boundaries of the masses. Benign versus malignant classification using the GCM-based texture features resulted in A/sub z/=0.79 with 19 benign and 13 malignant cases.

Current diagnosis of prostatic adenocarcinoma is done by manual analysis of biopsy tissue samples for tumor presence. However, the recent advent of whole slide digital scanners has made histopathological tissue specimens amenable to computer-aided diagnosis (CAD). In this paper, we present a CAD system to assist pathologists by automatically detecting prostate cancer from digitized images of prostate histological specimens. Automated diagnosis on very large high resolution images is done via a multi-resolution scheme similar to the manner in which a pathologist isolates regions of interest on a glass slide. Nearly 600 image texture features are extracted and used to perform pixel-wise Bayesian classification at each image scale to obtain corresponding likelihood scenes. Starting at the lowest scale, we apply the AdaBoost algorithm to combine the most discriminating features, and we analyze only pixels with a high combined probability of malignancy at subsequent higher scales. The system was evaluated on 22 studies by comparing the CAD result to a pathologist’s manual segmentation of cancer (which served as ground truth) and found to have an overall accuracy of 88%. Our results show that (1) CAD detection sensitivity remains consistently high across image scales while CAD specificity increases with higher scales, (2) the method is robust to choice of training samples, and (3) the multi-scale cascaded approach results in significant savings in computational time.

This paper presents an algorithm for classification of non-melanoma skin lesions based on a novel hierarchical K-Nearest Neighbors (K-NN) classifier. The K-NN classifier is simple, quick and effective. The hierarchical structure decomposes the classification task into a set of simpler problems, one at each node of the classification. Feature selection is embedded in the hierarchical framework that chooses the most relevant feature subsets at each node of the hierarchy. Colour and texture features are extracted from skin lesions. The accuracy of the proposed hierarchical scheme is higher than 93% in discriminating cancer and pre-malignant lesions from benign lesions, and it reaches an overall classification accuracy of 74% over five common classes of skin lesions, including two non-melanoma cancer types. This is the most extensive published result on non-melanoma skin cancer classification from colour images acquired by a standard camera (non-dermoscopy).

In lung cancer screening, benign and malignant nodules can be classified through nodule growth assessment by the registration and, then, subtraction between follow-up computed tomography scans. During the registration, the volume of nodule regions in the floating image should be preserved, whereas the volume of other regions in the floating image should be aligned to that in the reference image. However, ground glass opacity (GGO) nodules are very elusive to automatically segment due to their inhomogeneous interior. In other words, it is difficult to automatically define the volume-preserving regions of GGO nodules. In this paper, we propose an accurate and fast nonrigid registration method. It applies the volume-preserving constraint to candidate regions of GGO nodules, which are automatically detected by gray-level cooccurrence matrix (GLCM) texture analysis. Considering that GGO nodules can be characterized by their inner inhomogeneity and high intensity, we identify the candidate regions of GGO nodules based on the homogeneity values calculated by the GLCM and the intensity values. Furthermore, we accelerate our nonrigid registration by using Compute Unified Device Architecture (CUDA). In the nonrigid registration process, the computationally expensive procedures of the floating-image transformation and the cost-function calculation are accelerated by using CUDA. The experimental results demonstrated that our method almost perfectly preserves the volume of GGO nodules in the floating image as well as effectively aligns the lung between the reference and floating images. Regarding the computational performance, our CUDA-based method delivers about 20× faster registration than the conventional method. Our method can be successfully applied to a GGO nodule follow-up study and can be extended to the volume-preserving registration and subtraction of specific diseases in other organs (e.g., liver cancer).

The Gleason score is the single most important prognostic indicator for prostate cancer candidates and plays a significant role in treatment planning. Histopathological imaging of prostate tissue samples provides the gold standard for obtaining the Gleason score, but the manual assignment of Gleason grades is a labor-intensive and error-prone process. We have developed a texture classification system for automatic and reproducible Gleason grading. Our system characterizes the texture in images belonging to a tumor grade by clustering extracted filter responses at each pixel into textons (basic texture elements). We have used random forests to cluster the filter responses into textons followed by the spatial pyramid match kernel in conjunction with an SVM classifier. We have demonstrated the efficacy of our system in distinguishing between Gleason grades 3 and 4.

Breast cancer accounts for about 30% of all cancers and 15% of cancer deaths in women. Advances in computer-assisted analysis hold promise for classifying subtypes of disease and improving prognostic accuracy. We introduce a grid-enabled decision support system for performing automatic analysis of imaged breast tissue microarrays. To date, we have processed more than 1 00 000 digitized specimens (1200 times 1200 pixels each) on IBM's World Community Grid (WCG). As a part of the Help Defeat Cancer (HDC) project, we have analyzed that the data returned from WCG along with retrospective patient clinical profiles for a subset of 3744 breast tissue samples, and have reported the results in this paper. Texture-based features were extracted from the digitized specimens, and isometric feature mapping was applied to achieve nonlinear dimension reduction. Iterative prototyping and testing were performed to classify several major subtypes of breast cancer. Overall, the most reliable approach was gentle AdaBoost using an eight-node classification and regression tree as the weak learner. Using the proposed algorithm, a binary classification accuracy of 89% and the multiclass accuracy of 80% were achieved. Throughout the course of the experiments, only 30% of the dataset was used for training.

Prostatic adenocarcinoma is the most commonly occurring cancer among men in the United States, second only to skin cancer. Currently, the only definitive method to ascertain the presence of prostatic cancer is by trans-rectal ultrasound (TRUS) directed biopsy. Owing to the poor image quality of ultrasound, the accuracy of TRUS is only 20%-25%. High-resolution magnetic resonance imaging (MRI) has been shown to have a higher accuracy of prostate cancer detection compared to ultrasound. Consequently, several researchers have been exploring the use of high resolution MRI in performing prostate biopsies. Visual detection of prostate cancer, however, continues to be difficult owing to its apparent lack of shape, and the fact that several malignant and benign structures have overlapping intensity and texture characteristics. In this paper, we present a fully automated computer-aided detection (CAD) system for detecting prostatic adenocarcinoma from 4 Tesla ex vivo magnetic resonance (MR) imagery of the prostate. After the acquired MR images have been corrected for background inhomogeneity and nonstandardness, novel three-dimensional (3-D) texture features are extracted from the 3-D MRI scene. A Bayesian classifier then assigns each image voxel a "likelihood" of malignancy for each feature independently. The "likelihood" images generated in this fashion are then combined using an optimally weighted feature combination scheme. Quantitative evaluation was performed by comparing the CAD results with the manually ascertained ground truth for the tumor on the MRI. The tumor labels on the MR slices were determined manually by an expert by visually registering the MR slices with the corresponding regions on the histology slices. We evaluated our CAD system on a total of 33 two-dimensional (2-D) MR slices from five different 3-D MR prostate studies. Five slices from two different glands were used for training. Our feature combination scheme was found to outperform the individual texture features, and also other popularly used feature combination methods, including AdaBoost, ensemble averaging, and majority voting. Further, in several instances our CAD system performed better than the experts in terms of accuracy, the expert segmentations being determined solely from visual inspection of the MRI data. In addition, the intrasystem variability (changes in CAD accuracy with changes in values of system parameters) was significantly lower than the corresponding intraobserver and interobserver variability. CAD performance was found to be very similar for different training sets. Future work will focus on extending the methodology to guide high-resolution MRI-assisted in vivo prostate biopsies.

The potential of cytopathologic diagnoses to detect cancer of a variety of types non-invasively, cost-efficiently and up to three years ahead of conventional histopathologic diagnoses can be increased by the application of adjuvant methods, i.e. the combination of different stainings. For further improvement of cytopathology we introduced multimodal cell analysis (MMCA) which combines specific information about identical cells in different stainings successively applied to the same microscope slide. This requires a precise relocation and coregistration of individual cells under scrutiny. As a precondition for application in daily routine and screening settings the crucial relocation of cells has to be automated. The paper describes a method for an automatic retrieval of images of cells which have already been selected and recorded in a preceding staining together with their coordinates. Due to inevitable mechanical inaccuracies the geometric match is insufficient. The comparison of nuclear constellations based on segmentations in both stains facilitates an automatic correction of the position even if there is a subscene matching only. The process furthermore generates the initial guess for the succeeding coregistration which thereby gains robustness. The success rate of the method described is about 85%.

Histological grading of pathological images is used to determine the level of malignancy of cancerous tissues. This is a very important task in prostate cancer prognosis, since it is used for treatment planning. If infection of cancer is not rejected by noninvasive diagnostic techniques like magnetic resonance imaging, computed tomography scan, and ultrasound, then biopsy specimens of tissue are tested. For prostate, biopsied tissue is stained by hematoxyline and eosine method and viewed by pathologists under a microscope to determine its histological grade. Human grading is very subjective due to interobserver and intraobserver variations and in some cases difficult and time-consuming. Thus, an automatic and repeatable technique is needed for grading. The Gleason grading system is the most common method for histological grading of prostate tissue samples. According to this system, each cancerous specimen is assigned one of five grades. Although some automatic systems have been developed for analysis of pathological images, Gleason grading has not yet been automated; the goal of this research is to automate it. To this end, we calculate energy and entropy features of multiwavelet coefficients of the image. Then, we select most discriminative features by simulated annealing and use a k-nearest neighbor classifier to classify each image to appropriate grade (class). The leaving-one-out technique is used for error rate estimation. We also obtain the results using features extracted by wavelet packets and co-occurrence matrices and compare them with the multiwavelet method. Experimental results show the superiority of the multiwavelet transforms compared with other techniques. For multiwavelets, critically sampled preprocessing outperforms repeated-row preprocessing and has less sensitivity to noise for second level of decomposition. The first level of decomposition is very sensitive to noise and, thus, should not be used for feature extraction. The best multiwavelet method grades prostate pathological images correctly 97% of the time.

Breast cancer is the most frequently diagnosed malignancy and the second leading cause of mortality in women . In the last decade, ultrasound along with digital mammography has come to be regarded as the gold standard for breast cancer diagnosis. Automatically detecting tumors and extracting lesion boundaries in ultrasound images is difficult due to their specular nature and the variance in shape and appearance of sonographic lesions. Past work on automated ultrasonic breast lesion segmentation has not addressed important issues such as shadowing artifacts or dealing with similar tumor like structures in the sonogram. Algorithms that claim to automatically classify ultrasonic breast lesions, rely on manual delineation of the tumor boundaries. In this paper, we present a novel technique to automatically find lesion margins in ultrasound images, by combining intensity and texture with empirical domain specific knowledge along with directional gradient and a deformable shape-based model. The images are first filtered to remove speckle noise and then contrast enhanced to emphasize the tumor regions. For the first time, a mathematical formulation of the empirical rules used by radiologists in detecting ultrasonic breast lesions, popularly known as the "Stavros Criteria" is presented in this paper. We have applied this formulation to automatically determine a seed point within the image. Probabilistic classification of image pixels based on intensity and texture is followed by region growing using the automatically determined seed point to obtain an initial segmentation of the lesion. Boundary points are found on the directional gradient of the image. Outliers are removed by a process of recursive refinement. These boundary points are then supplied as an initial estimate to a deformable model. Incorporating empirical domain specific knowledge along with low and high-level knowledge makes it possible to avoid shadowing artifacts and lowers the chance of confusing similar tumor like structures for the lesion. The system was validated on a database of breast sonograms for 42 patients. The average mean boundary error between manual and automated segmentation was 6.6 pixels and the normalized true positive area overlap was 75.1%. The algorithm was found to be robust to 1) variations in system parameters, 2) number of training samples used, and 3) the position of the seed point within the tumor. Running time for segmenting a single sonogram was 18 s on a 1.8-GHz Pentium machine.

Clustered microcalcifications on X-ray mammograms are an important sign for early detection of breast cancer. Texture-analysis methods can be applied to detect clustered microcalcifications in digitized mammograms. In this paper, a comparative study of texture-analysis methods is performed for the surrounding region-dependence method, which has been proposed by the authors, and conventional texture-analysis methods, such as the spatial gray level dependence method, the gray-level run-length method, and the gray-level difference method. Textural features extracted by these methods are exploited to classify regions of interest (ROI's) into positive ROI's containing clustered microcalcifications and negative ROI's containing normal tissues. A three-layer backpropagation neural network is used as a classifier. The results of the neural network for the texture-analysis methods are evaluated by using a receiver operating-characteristics (ROC) analysis. The surrounding region-dependence method is shown to be superior to the conventional texture-analysis methods with respect to classification accuracy and computational complexity.

Mammograms can depict most of the significant changes of breast disease. The primary radiographic signs of cancer are masses (its density, size, shape, borders), spicular lesions and calcification content. These features may be extracted according to their coherence and orientation and can provide important visual cues for radiologists to locate suspicious areas without generating false positives. An artifact free enhancement algorithm based on overcomplete multiscale wavelet analysis is presented. The novelty of this algorithm lies in its detection of directional features and removal of unwanted perturbations. Compared to existing multiscale enhancement approaches, images processed with this method appear more familiar to radiologists and naturally close to the original mammogram.

The authors investigated the classification of regions of interest (ROI's) on mammograms as either mass or normal tissue using a convolution neural network (CNN). A CNN is a backpropagation neural network with two-dimensional (2-D) weight kernels that operate on images. A generalized, fast and stable implementation of the CNN was developed. The input images to the CNN were obtained from the ROI's using two techniques. The first technique employed averaging and subsampling. The second technique employed texture feature extraction methods applied to small subregions inside the ROI. Features computed over different subregions were arranged as texture images, which were subsequently used as CNN inputs. The effects of CNN architecture and texture feature parameters on classification accuracy were studied. Receiver operating characteristic (ROC) methodology was used to evaluate the classification accuracy. A data set consisting of 168 ROIs containing biopsy-proven masses and 504 ROI's containing normal breast tissue was extracted from 168 mammograms by radiologists experienced in mammography. This data set was used for training and testing the CNN. With the best combination of CNN architecture and texture feature parameters, the area under the test ROC curve reached 0.87, which corresponded to a true-positive fraction of 90% at a false positive fraction of 31%. The authors' results demonstrate the feasibility of using a CNN for classification of masses and normal tissue on mammograms.

Tumor spatial heterogeneity is an important prognostic factor, which may be reflected in medical imagesImage texture analysis is an approach of quantifying heterogeneity that may not be appreciated by the naked eye. Different methods can be applied including statistical-, model-, and transform-based methods.Early evidence suggests that texture analysis has the potential to augment diagnosis and characterization as well as improve tumor staging and therapy response assessment in oncological practice.This review provides an overview of the application of texture analysis with different imaging modalities, CT, MRI, and PET, to date and describes the technical challenges that have limited its widespread clinical implementation so far. With further efforts to refine its application, image texture analysis has the potential to develop into a valuable clinical tool for oncologic imaging. TEACHING POINTS : • Tumor spatial heterogeneity is an important prognostic factor. • Image texture analysis is an approach of quantifying heterogeneity. • Different methods can be applied, including statistical-, model-, and transform-based methods. • Texture analysis could improve the diagnosis, tumor staging, and therapy response assessment.

Small breast tumors could be detected early by registering two longitudinal breast images and subtracting them. The noise and low contrast in breast images permit few control points for image registration. We propose a new textural transformation, called the “starbyte” transformation, to solve this image registration problem. The starbyte transform is computed using local neighborhood properties around every pixel in an image and converts breast images to texture maps with clearly demarcable regions from which accurate control points can be easily identified. A complete registration algorithm is presented which consists of first starbyte transforming the original images, extracting control points (both manually and automatically), and performing unwarping using thin plate splines. The performance of the algorithm was rigorously tested on portions of mammograms warped by six different types of distortions. Limited studies have also been performed with added noise and simulated tumors. Preliminary results are also presented on roughly corresponding 2D slices taken from 3D MRIs of volunteers imaged twice with breast repositioning between the two images.

In this work, we present a Computer Aided Diagnosis (CAD) based technique for automatic classification of benign and malignant thyroid lesions in 3D contrast-enhanced ultrasound images. The images were obtained from 20 patients. Fine needle aspiration biopsy and histology confirmed malignancy. Discrete Wavelet Transform (DWT) and texture based features were extracted from the thyroid images. The resulting feature vectors were used to train and test three different classifiers: K-Nearest Neighbor (K-NN), Probabilistic Neural Network (PNN), and Decision Tree (DeTr) using ten-fold cross validation technique. Our results show that combination of DWT and texture features in the K-NN classifier resulted in a classification accuracy of 98.9%, a sensitivity of 98%, and a specificity of 99.8%. Thus, the preliminary results of the proposed technique show that it could be adapted as an adjunct tool that can give valuable second opinions to the doctors regarding the nature of the thyroid nodule. The technique is cost-effective, non-invasive, fast, completely automated and gives more objective and reproducible results compared to manual analysis of the ultrasound images. We however intend to establish the clinical applicability of this technique by evaluating it with more data in the future.

The use of multiscale AM-FM analysis systems has been recently demonstrated in a variety of applications in medical image analysis. In all of these applications, a fixed filter-bank is used as a preprocessing step for estimating different AM-FM components from different scales. In this paper, for the first time, we introduce the use of an adaptive, multiscale AM-FM approach that searches for the optimal filter-bank specification for use in image classification. We demonstrate an example application in hysteroscopy imaging, for identification of gynaecological cancer, where the optimal filter-bank turns out to be circularly symmetric. © 2012 IEEE. 744 747 Sponsors: IEEE IEEE Computer Society University of Cyprus Biological and AI Foundation (BAIF) Frederick University Conference code: 95206 Cited By :2

Translational relevance: Many types of cancer are located and assessed via positron emission tomography (PET) using the 18F-fluorodeoxyglucose (FDG) radiotracer of glucose uptake. There is rapidly increasing interest in exploiting the intra-tumor heterogeneity observed in these FDG-PET images as an indicator of disease outcome. If this image heterogeneity is of genuine prognostic value, then it either correlates to known prognostic factors, such as tumor stage, or it indicates some as yet unknown tumor quality. Therefore, the first step in demonstrating the clinical usefulness of image heterogeneity is to explore the dependence of image heterogeneity metrics upon established prognostic indicators and other clinically interesting factors. If it is shown that image heterogeneity is merely a surrogate for other important tumor properties or variations in patient populations, then the theoretical value of quantified biological heterogeneity may not yet translate into the clinic given current imaging technology. Purpose: We explore the relation between pelvic lymph node status at diagnosis and the visually evident uptake heterogeneity often observed in 18F-fluorodeoxyglucose positron emission tomography (FDG-PET) images of cervical carcinomas. Experimental design: We retrospectively studied the FDG-PET images of 47 node negative and 38 node positive patients, each having FIGO stage IIb tumors with squamous cell histology. Imaged tumors were segmented using 40% of the maximum tumor uptake as the tumor-defining threshold and then converted into sets of three-dimensional coordinates. We employed the sphericity, extent, Shannon entropy (S) and the accrued deviation from smoothest gradients (ζ) as image heterogeneity metrics. We analyze these metrics within tumor volume strata via: the Kolmogorov-Smirnov test, principal component analysis and contingency tables. Results: We found no statistically significant difference between the positive and negative lymph node groups for any one metric or plausible combinations thereof. Additionally, we observed that S is strongly dependent upon tumor volume and that ζ moderately correlates with mean FDG uptake. Conclusions: FDG uptake heterogeneity did not indicate patients with differing prognoses. Apparent heterogeneity differences between clinical groups may be an artifact arising from either the dependence of some image metrics upon other factors such as tumor volume or upon the underlying variations in the patient populations compared.