Application of artificial neural networks for the classification of liver lesions by image texture parameters
TL;DR: In this article, a multilayered back-propagation neural network was used for liver lesion classification using B-scan ultrasound images for normal, hemangioma and malignant livers.
Abstract: Ultrasound imaging is a powerful tool for characterizing the state of soft tissues; however, in some cases, where only subtle differences in images are seen as in certain liver lesions such as hemangioma and malignancy, existing B-scan methods are inadequate. More detailed analyses of image texture parameters along with artificial neural networks can be utilized to enhance differentiation. From B-scan ultrasound images, 11 texture parameters comprising of first, second and run length statistics have been obtained for normal, hemangioma and malignant livers. Tissue characterization was then performed using a multilayered backpropagation neural network. The results for 113 cases have been compared with a classification based on discriminant analysis. For linear discriminant analysis, classification accuracy is 79.6% and with neural networks the accuracy is 100%. The present results show that neural networks classify better than discriminant analysis, demonstrating a much potential for clinical application.
Citations
More filters
TL;DR: This paper reviews ultrasound segmentation methods, in a broad sense, focusing on techniques developed for medical B-mode ultrasound images, and presents a classification of methodology in terms of use of prior information.
Abstract: This paper reviews ultrasound segmentation methods, in a broad sense, focusing on techniques developed for medical B-mode ultrasound images. First, we present a review of articles by clinical application to highlight the approaches that have been investigated and degree of validation that has been done in different clinical domains. Then, we present a classification of methodology in terms of use of prior information. We conclude by selecting ten papers which have presented original ideas that have demonstrated particular clinical usefulness or potential specific to the ultrasound segmentation problem
1,150 citations
01 Sep 2003
TL;DR: An approach to the detection of tumors in colonoscopic video based on a new color feature extraction scheme to represent the different regions in the frame sequence based on the wavelet decomposition, reaching 97% specificity and 90% sensitivity.
Abstract: We present an approach to the detection of tumors in colonoscopic video. It is based on a new color feature extraction scheme to represent the different regions in the frame sequence. This scheme is built on the wavelet decomposition. The features named as color wavelet covariance (CWC) are based on the covariances of second-order textural measures and an optimum subset of them is proposed after the application of a selection algorithm. The proposed approach is supported by a linear discriminant analysis (LDA) procedure for the characterization of the image regions along the video frames. The whole methodology has been applied on real data sets of color colonoscopic videos. The performance in the detection of abnormal colonic regions corresponding to adenomatous polyps has been estimated high, reaching 97% specificity and 90% sensitivity.
480 citations
01 Sep 2003
TL;DR: A computer-aided diagnostic (CAD) system for the classification of hepatic lesions from computed tomography (CT) images is presented and shows that genetic algorithms result in lower dimension feature vectors and improved classification performance.
Abstract: In this paper, a computer-aided diagnostic (CAD) system for the classification of hepatic lesions from computed tomography (CT) images is presented. Regions of interest (ROIs) taken from nonenhanced CT images of normal liver, hepatic cysts, hemangiomas, and hepatocellular carcinomas have been used as input to the system. The proposed system consists of two modules: the feature extraction and the classification modules. The feature extraction module calculates the average gray level and 48 texture characteristics, which are derived from the spatial gray-level co-occurrence matrices, obtained from the ROIs. The classifier module consists of three sequentially placed feed-forward neural networks (NNs). The first NN classifies into normal or pathological liver regions. The pathological liver regions are characterized by the second NN as cyst or "other disease". The third NN classifies "other disease" into hemangioma or hepatocellular carcinoma. Three feature selection techniques have been applied to each individual NN: the sequential forward selection, the sequential floating forward selection, and a genetic algorithm for feature selection. The comparative study of the above dimensionality reduction methods shows that genetic algorithms result in lower dimension feature vectors and improved classification performance.
280 citations
31 Mar 2016
TL;DR: Radiomics is defined as the high throughput extraction of quantitative imaging features or texture from imaging to decode tissue pathology and creating a high dimensional data set for feature extraction and can be further used to develop computational models using advanced machine learning algorithms that may serve as a tool for personalized diagnosis and treatment guidance.
Abstract: The increasing use of biomarkers in cancer have led to the concept of personalized medicine for patients. Personalized medicine provides better diagnosis and treatment options available to clinicians. Radiological imaging techniques provide an opportunity to deliver unique data on different types of tissue. However, obtaining useful information from all radiological data is challenging in the era of ‘big data’. Recent advances in computational power and the use of genomics have generated a new area of research termed Radiomics. Radiomics is defined as the high throughput extraction of quantitative imaging features or texture (radiomics) from imaging to decode tissue pathology and creating a high dimensional data set for feature extraction. Radiomic features provide information about the gray-scale patterns, inter-pixel relationships. In addition, shape and spectral properties can be extracted within the same regions of interest on radiological images. Moreover, these features can be further used t...
249 citations
TL;DR: An experimental study with real images demonstrated the feasibility and promise of the proposed approach in discriminating between cervical texture patterns indicative of different stages of cervical lesions.
Abstract: This paper presents a generalized statistical texture analysis technique for characterizing and recognizing typical, diagnostically most important, vascular patterns relating to cervical lesions from colposcopic images. The contributions of the research include: (1) the introduction of a generalized texture analysis technique based on the combination of the conventional statistical and structural textural analysis approaches by using a statistical description of geometric primitives; (2) the introduction of a set of textural measures that capture the specific characteristics of cervical textures as perceived by humans. An experimental study with real images demonstrated the feasibility and promise of the proposed approach in discriminating between cervical texture patterns indicative of different stages of cervical lesions.
171 citations
References
More filters
TL;DR: It is suggested that ultrasonic image texture analysis is a simple way to markedly reduce the number of benign lesion biopsies without missing additional cancers.
220 citations
TL;DR: In this paper, the difference between the echo intensities of the liver and kidney determined from ultrasonic histograms was used for the diagnosis of fatty liver using ultrasonography, which had a sensitivity of 91.3%, a specificity of 83.8%, and an accuracy of 86.7%.
Abstract: Accurate diagnosis of fatty liver using ultrasonography was attempted based on the difference between the echo intensities of the liver and kidney determined from ultrasonic histograms. Livers were then classified as having fatty infiltration, normal histology, or intermediate histology based on CT ratios established previously in earlier work comparing non-contrast-enhanced liver and spleen. The hepatorenal difference was significantly greater in the fatty liver group than in the normal liver group (8.9 +/- 2.0 dB vs 2.5 +/- 4.5 dB, p or = 7.0 dB was taken as the criterion, this method had a sensitivity of 91.3%, a specificity of 83.8%, and an accuracy of 86.7% for the diagnosis of fatty liver. Thus, quantitative ultrasonic diagnosis of fatty liver can be performed using echo intensity histograms.
173 citations
113 citations
TL;DR: Methods based on neural network classification of texture features show promise for potentially decreasing the number of unnecessary biopsies by a significant amount in patients with sonographically identifiable lesions.
Abstract: A set of ultrasonograms of lesions from 200 patients between the ages of 14 and 93 years who underwent mammography followed by ultrasonographic examination and excisional biopsy has been studied with computer vision techniques to improve the ultrasonographic specificity of the diagnosis. Selected features representing the texture of the lesion were calculated and then classified by an artificial neural network. This network was biased toward correctly classifying all the malignant cases at the expense of some misclassification of the benign cases. The network diagnosed the malignant cases with 100% sensitivity and 40% specificity (compared with 0% specificity for the radiologists diagnosing the same set of cases in the breast imaging setting), and tests performed with a leave-one-out technique indicate that the network will generalize well to new cases. This suggests that methods based on neural network classification of texture features show promise for potentially decreasing the number of unnecessary biopsies by a significant amount in patients with sonographically identifiable lesions.
84 citations
TL;DR: The clinical findings, radionuclide scintigrams, and ultrasonograms of hepatic hemangiomas are described for 11 patients and may help in the selection of the appropriate investigative studies.
Abstract: The clinical findings, radionuclide scintigrams, and ultrasonograms of hepatic hemangiomas are described for 11 patients. Scintigraphy frequently demonstrated a marginal or subcapsular location. The lesions were otherwise nonspecific in appearance and were indistinguishable from metastatic foci. Sonographically, hypoechoic, hyperechoic, and predominantly anechoic patterns were observed, not unlike those described for malignant foci. However, 7 hemangiomas had a central linear septum. Liver function tests were normal in all patients. Since percutaneous biopsy is hazardous and hepatic angiography usually diagnostic, the scintigraphic and ultrasonic appearances described may help in the selection of the appropriate investigative studies.
52 citations