Tissue characterization from X-ray images

Early detection of radiographic knee osteoarthritis using computer-aided analysis

OBJECTIVE

To determine whether computer-based analysis can detect features predictive of osteoarthritis (OA) development in radiographically normal knees.

METHOD

A systematic computer-aided image analysis method weighted neighbor distances using a compound hierarchy of algorithms representing morphology (WND-CHARM) was used to analyze pairs of weight-bearing knee X-rays. Initial X-rays were all scored as normal Kellgren-Lawrence (KL) grade 0, and on follow-up approximately 20 years later either developed OA (defined as KL grade=2) or remained normal.

RESULTS

The computer-aided method predicted whether a knee would change from KL grade 0 to grade 3 with 72% accuracy (P<0.00001), and to grade 2 with 62% accuracy (P<0.01). Although a large part of the predictive signal comes from the image tiles that contained the joint, the region adjacent to the tibial spines provided the strongest predictive signal.

CONCLUSION

Radiographic features detectable using a computer-aided image analysis method can predict the future development of radiographic knee OA.

Multi-scaled morphological features for the characterization of mammographic masses using statistical classification schemes

OBJECTIVE

A comprehensive signal analysis approach on the mammographic mass boundary morphology is presented in this article. The purpose of this study is to identify efficient sets of simple yet effective shape features, employed in the original and multi-scaled spectral representations of the boundary, for the characterization of the mammographic mass. These new methods of mass boundary representation and processing in more than one domain greatly improve the information content of the base data that is used for pattern classification purposes, introducing comprehensive spectral and multi-scale wavelet versions of the original boundary signals. The evaluation is conducted against morphological and diagnostic characterization of the mass, using statistical methods, fractal dimension analysis and a wide range of classifier architectures.

METHODS AND MATERIALS

This study consists of (a) the investigation of the original radial distance measurements under the complete spectrum of signal analysis, (b) the application of curve feature extractors of morphological characteristics and the evaluation of the discriminative power of each one of them, by means of statistical significance analysis and dataset fractal dimension, and (c) the application of a wide range of classifier architectures on these morphological datasets, in order to conduct a comparative evaluation of the efficiency and effectiveness of all architectures, for mammographic mass characterization. Radial distance signal was exploited using the discrete Fourier transform (DFT) and the discrete wavelet transform (DWT) as additional carrier signals. Seven uniresolution feature functions were applied over these carrier signals and multiple shape descriptors were created. Classification was conducted against mass shape type and clinical diagnosis, using a wide range of linear and non-linear classifiers, including linear discriminant analysis (LDA), least-squares minimum distance (LSMD), k-nearest neighbor (k-NN), radial basis function (RBF) and multi-layered perceptron (MLP) neural networks (NN), and support vector machines (SVM). Fractal analysis was employed as a dataset analysis tool in the feature selection phase. The discriminative power of the features produced by this composite analysis is subsequently analyzed by means of multivariate analysis of variance (MANOVA) and tested against two distinct classification targets, namely (a) the morphological shape type of the mass and (b) the histologically verified clinical diagnosis for each mammogram.

RESULTS

Statistical analysis and classification results have shown that the discrimination value of the features extracted from the DWT components and especially the DFT spectrum, are of great importance. Furthermore, much of the information content of the curve features in the case of DFT and DWT datasets is directly related to the texture and fine-scale details of the corresponding envelope signal of the spectral components. Neural classifiers outperformed all other methods (SVM not used because they are mainly two-class classifiers) with overall success rate of 72.3% for shape type identification, while SVM achieved the overall highest 91.54% for clinical diagnosis. Receiver operating characteristic (ROC) analysis has been employed to present the sensitivity and specificity of the results of this study.

Texture analysis of perimenopausal and post-menopausal endometrial tissue in grayscale transvaginal ultrasonography

The aim of this study was to investigate the feasibility of texture analysis in characterizing endometrial tissue as depicted in two-dimensional (2D) grayscale transvaginal ultrasonography. Digital transvaginal ultrasound endometrial images were acquired from 65 perimenopausal and post-menopausal women prior to gynaecological operations; histology revealed 15 malignant and 50 benign cases. Images were processed with a wavelet-based contrast enhancement technique. Three regions of interest (ROIs) were identified (endometrium, endometrium plus adjacent myometrium, layer containing endometrial-myometrial interface) on each original and processed image. 32 textural features were extracted from each ROI employing first and second order statistics texture analysis algorithms. Textural features-based models were generated for differentiating benign from malignant endometrial tissue using stepwise logistic regression analysis. Models' performance was evaluated by means of receiver operating characteristic (ROC) analysis. The best logistic regression model comprised seven textural features extracted from the ROIs determined on the processed images; three features were extracted from the endometrium, while four features were extracted from the layer containing the endometrial-myometrial interface. The area under the ROC curve (A(z)) was 0.956+/-0.038, providing 86.0% specificity at 93.3% sensitivity using the cut-off level of 0.5 for probability of malignancy. Texture analysis of 2D grayscale transvaginal ultrasound images can effectively differentiate malignant from benign endometrial tissue and may contribute to computer-aided diagnosis of endometrial cancer.

Assessing hip osteoarthritis severity utilizing a probabilistic neural network based classification scheme

A computer-based classification system is proposed for the characterization of hips from pelvic radiographs as normal or osteoarthritic and for the discrimination among various grades of osteoarthritis (OA) severity. Pelvic radiographs of 18 patients with verified unilateral hip OA were evaluated by three experienced physicians, who assessed OA severity employing the Kellgren and Lawrence scale as: normal, mild/moderate and severe. Five run-length, 75 Laws' and 5 novel textural features were extracted from the digitized radiographic images of each patient's osteoarthritic and contralateral normal hip joint spaces (HJSs). Each one of the three sets of textural features (run-lengths, Laws' and novel features) was separately utilized for assigning hips into the three OA severity categories, by means of a probabilistic neural network (PNN) classifier based hierarchical tree structure. The highest classification accuracy (100%) for characterizing hips as normal, of mild/moderate or of severe OA was obtained for the novel textural features set. Additionally, the novel textural features were used to design a mathematical regression model for providing a quantitative estimation of OA severity. Measured OA severity values, as expressed by HJS-narrowing, correlated highly (r=0.85, p<0.001) with the predicted values by the mathematical regression model. The proposed system may be valuable in OA-patient management.

Osteoarthritis severity of the hip by computer-aided grading of radiographic images

A computer-aided classification system was developed for the assessment of the severity of hip osteoarthritis (OA). Sixty-four radiographic images of normal and osteoarthritic hips were digitized and enhanced. Employing the Kellgren and Lawrence scale, the hips were grouped by three experienced orthopaedists into three OA-severity categories: Normal, Mild/Moderate and Severe. Utilizing custom-developed software, 64 ROIs corresponding to the radiographic Hip Joint Spaces were manually segmented and novel textural features were generated. These features were used in the design of a two-level classification scheme for characterizing hips as normal or osteoarthritic (1st level) and as of Mild/Moderate or Severe OA (2nd level). At each classification level, an ensemble of three classifiers was implemented. The proposed classification scheme discriminated correctly all normal hips from osteoarthritic hips (100% accuracy), while the discrimination accuracy between Mild/Moderate and Severe osteoarthritic hips was 95.7%. The proposed system could be used as a diagnosis decision-supporting tool.

Mammographic masses characterization based on localized texture and dataset fractal analysis using linear, neural and support vector machine classifiers

OBJECTIVE

Localized texture analysis of breast tissue on mammograms is an issue of major importance in mass characterization. However, in contrast to other mammographic diagnostic approaches, it has not been investigated in depth, due to its inherent difficulty and fuzziness. This work aims to the establishment of a quantitative approach of mammographic masses texture classification, based on advanced classifier architectures and supported by fractal analysis of the dataset of the extracted textural features. Additionally, a comparison of the information content of the proposed feature set with that of the qualitative characteristics used in clinical practice by expert radiologists is presented.

METHODS AND MATERIAL

An extensive set of textural feature functions was applied to a set of 130 digitized mammograms, in multiple configurations and scales, constructing compact datasets of textural "signatures" for benign and malignant cases of tumors. These quantitative textural datasets were subsequently studied against a set of a thorough and compact list of qualitative texture descriptions of breast mass tissue, normally considered under a typical clinical assessment, in order to investigate the discriminating value and the statistical correlation between the two sets. Fractal analysis was employed to compare the information content and dimensionality of the textural features datasets with the qualitative information provided through medical diagnosis. A wide range of linear and non-linear classification architectures was employed, including linear discriminant analysis (LDA), least-squares minimum distance (LSMD), K-nearest-neighbors (K-nn), radial basis function (RBF) and multi-layer perceptron (MLP) artificial neural network (ANN), as well as support vector machine (SVM) classifiers. The classification process was used as the means to evaluate the inherent quality and informational content of each of the datasets, as well as the objective performance of each of the classifiers themselves in real classification of mammographic breast tumors against verified diagnosis.

RESULTS

Textural features extracted at larger scales and sampling box sizes proved to be more content-rich than their equivalents at smaller scales and sizes. Fractal analysis on the dimensionality of the textural datasets verified that reduced subsets of optimal feature combinations can describe the original feature space adequately for classification purposes and at least the same detail and quality as the list of qualitative texture descriptions provided by a human expert. Non-linear classifiers, especially SVMs, have been proven superior to any linear equivalent. Breast mass classification of mammograms, based only on textural features, achieved an optimal score of 83.9%, through SVM classifiers.

Quantitative assessment of hip osteoarthritis based on image texture analysis

A non-invasive method was developed to investigate the potential capacity of digital image texture analysis in evaluating the severity of hip osteoarthritis (OA) and in monitoring its progression. 19 textural features evaluating patterns of pixel intensity fluctuations were extracted from 64 images of radiographic hip joint spaces (HJS), corresponding to 32 patients with verified unilateral or bilateral OA. Images were enhanced employing custom developed software for the delineation of the articular margins on digitized pelvic radiographs. The severity of OA for each patient was assessed by expert orthopaedists employing the Kellgren and Lawrence (KL) scale. Additionally, an index expressing HJS-narrowing was computed considering patients from the unilateral OA-group. A textural feature that quantified pixel distribution non-uniformity (grey level non-uniformity, GLNU) demonstrated the strongest correlation with the HJS-narrowing index among all extracted features and utilized in further analysis. Classification rules employing GLNU feature were introduced to characterize a hip as normal or osteoarthritic and to assign it to one of three severity categories, formed in accordance with the KL scale. Application of the proposed rules resulted in relatively high classification accuracies in characterizing a hip as normal or osteoarthritic (90.6%) and in assigning it to the correct KL scale category (88.9%). Furthermore, the strong correlation between the HJS-narrowing index and the pathological GLNU (r = -0.9, p<0.001) was utilized to provide percentages quantifying hip OA-severity. Texture analysis may contribute in the quantitative assessment of OA-severity, in the monitoring of OA-progression and in the evaluation of a chondroprotective therapy.

Significance analysis of qualitative mammographic features, using linear classifiers, neural networks and support vector machines

Advances in modern technologies and computers have enabled digital image processing to become a vital tool in conventional clinical practice, including mammography. However, the core problem of the clinical evaluation of mammographic tumors remains a highly demanding cognitive task. In order for these automated diagnostic systems to perform in levels of sensitivity and specificity similar to that of human experts, it is essential that a robust framework on problem-specific design parameters is formulated. This study is focused on identifying a robust set of clinical features that can be used as the base for designing the input of any computer-aided diagnosis system for automatic mammographic tumor evaluation. A thorough list of clinical features was constructed and the diagnostic value of each feature was verified against current clinical practices by an expert physician. These features were directly or indirectly related to the overall morphological properties of the mammographic tumor or the texture of the fine-scale tissue structures as they appear in the digitized image, while others contained external clinical data of outmost importance, like the patient's age. The entire feature set was used as an annotation list for describing the clinical properties of mammographic tumor cases in a quantitative way, such that subsequent objective analyses were possible. For the purposes of this study, a mammographic image database was created, with complete clinical evaluation descriptions and positive histological verification for each case. All tumors contained in the database were characterized according to the identified clinical features' set and the resulting dataset was used as input for discrimination and diagnostic value analysis for each one of these features. Specifically, several standard methodologies of statistical significance analysis were employed to create feature rankings according to their discriminating power. Moreover, three different classification models, namely linear classifiers, neural networks and support vector machines, were employed to investigate the true efficiency of each one of them, as well as the overall complexity of the diagnostic task of mammographic tumor characterization. Both the statistical and the classification results have proven the explicit correlation of all the selected features with the final diagnosis, qualifying them as an adequate input base for any type of similar automated diagnosis system. The underlying complexity of the diagnostic task has justified the high value of sophisticated pattern recognition architectures.

Detection of single and clustered microcalcifications in mammograms using fractals models and neural networks

Microcalcifications (microCas) are often early signs of breast cancer. However, detecting them is a difficult visual task and recognizing malignant lesions is a complex diagnostic problem. In recent years, several research groups have been working to develop computer-aided diagnosis (CAD) systems for X-ray mammography. In this paper, we propose a method to detect and classify microcalcifications. In order to discover the presence of microCas clusters, particular attention is paid to the analysis of the spatial arrangement of detected lesions. A fractal model has been used to describe the mammographic image, thus, allowing the use of a matched filtering stage to enhance microcalcifications against the background. A region growing algorithm, coupled with a neural classifier, detects existing lesions. Subsequently, a second fractal model is used to analyze their spatial arrangement so that the presence of microcalcification clusters can be detected and classified. Reported results indicate that fractal models provide an adequate framework for medical image processing; consequently high correct classification rates are achieved.

Distinct patterns of active and non-active plaques using texture analysis on brain NMR images in multiple sclerosis patients: preliminary results

The benefits of texture analysis of magnetic resonance images have been assessed in multiple sclerosis (MS) patients. Out of thirty-two lesions identified in eight MS patients, nine were considered active, judging from their gadolinium uptake. Texture analysis allowed to obtain forty-two characterizing parameters for each lesion. Using discriminant analysis as a statistical method allowed to classify the lesions into two groups: active or non-active. An attempt to classify their level of activity by using only co-occurrence matrices was unsuccessful. Alternately, the same type of analysis performed on runlength analysis criteria allowed the accurate classification of 88% of active lesions and 96% of non-active lesions. Using incremental discriminate analysis can reduce the number of useful parameters. This method showed that among the 42 parameters, 8 only were highly significant and permitted an accurate classification. Five of these parameters are runlength parameters, and three others are more directly related to the global distribution. The main interest of runlength parameters is that they allowed to demonstrate that the lesion structure was different in active and non-active plaques. This preliminary work suggests that using texture analysis could be of interest in the follow-up of MS patients because it provides an opportunity to identify active lesions without frequent gadolinium injections.

Artificial neural networks in chest radiography: application to the differential diagnosis of interstitial lung disease

RATIONALE AND OBJECTIVES

The authors evaluated the usefulness of artificial neural networks (ANNs) in the differential diagnosis of interstitial lung disease.

MATERIALS AND METHODS

The authors used three-layer, feed-forward ANNs with a back-propagation algorithm. The ANNs were designed to distinguish between 11 interstitial lung diseases on the basis of 10 clinical parameters and 16 radiologic findings extracted by chest radiologists. Thus, the ANNs consisted of 26 input units and 11 output units. One hundred fifty actual clinical cases, 110 cases from previously published articles, and 110 hypothetical cases were used for training and testing the ANNs by using a round-robin (or leave-one-out) technique. ANN performance was evaluated with receiver operating characteristic (ROC) analysis.

RESULTS

The Az (area under the ROC curve) obtained with actual clinical cases was 0.947, and both the sensitivity and specificity of the ANNs were approximately 90% in terms of indicating the correct diagnosis with the two largest output values among the 11 diseases.

CONCLUSION

ANNs using clinical parameters and radiologic findings may be useful for making the differential diagnosis of interstitial lung disease on chest radiographs.