Texture is an important significant property of medical images based on which images can be characterized and classified in a content-based image retrieval and classification system. This paper examines the feature extraction methods to ameliorate texture recognition accuracy by extracting the rotation-invariant texture feature from liver images by the individual Gabor filter method and by multi-scale Gabor rotation-invariant LBP (MGRLBP) method. The features extracted from both the approaches are tested on a set of 60 liver images of four different classes. The classification algorithms such as support vector machine (SVM) and k-nearest neighbor (KNN) were used to evaluate the extracted features from both methods, showing advancing improvements with the MGRLBP method over the individual method in the classification task.
Medical Diagnosis has been gaining importance in everyday life. The diseases and their symptoms are highly varying and there is always a need for a continuous update of knowledge needed for the doctors. This forces lots of challenges as the diagnostic tools need to visualize organs and soft tissues and further classify them for diagnosis. One such application of diagnostic ultrasound is liver imaging. The existing approaches for classification & retrieval system have the following issues: speckle noise, semantic gap, computational time, dimensionality reduction and accuracy of retrieved images from large dataset. This paper proposes a new method for the classification & retrieval of liver diseases from ultrasound image dataset. The proposed work concentrates on diagnosing both focal and diffuse liver diseases from ultrasound images. The contribution of this paper relies on the following areas. Speckle reduction by Modified Laplacian Pyramid Nonlinear Diffusion (MLPND), Mutual Information (MI) based image registration, Image texture analysis by Haralick's features, Image Classification & retrieval by machine learning algorithms. The dataset used in each phase of the work are authenticated dataset provided by doctors. The results at each phase have been evaluated with doctors in the relevant field.The CNR value for MLPND has improved 95% compared to existing speckle reduction methods. The MI based registration with optimization techniques to reduce the computation time & monitor the growth of the liver diseases. The results retrieved from different machine learning techniques indicate that the proposed methods improve the image quality and overcome the fuzzy nature of dataset.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.