Neural network based texture analysis of CT images for fatty and cirrhosis liver classification

• Medical image analysis.
• Provides additional information for the doctors apart from visual interpretation.
• Proposed system classifies fatty and cirrhosis liver with 95% accuracy using PNN classifier.

Computational methods are useful for medical diagnosis because they provide additional information that cannot be obtained by simple visual interpretation of clinical presentations and radiologic imaging. As a result an enormous amount of research effort has been targeted at achieving automated medical image analysis. This work reports the texture analysis of Computed Tomography (CT) images and development of Probabilistic Neural Network (PNN), Linear Vector Quantization (LVQ) Neural Network and Back Propagation Neural Network (BPN) for classification of fatty and cirrhosis liver from CT abdominal images. Neural networks are supported by more conventional image processing operations in order to achieve the objective set. To evaluate the classifiers, Receiver Operating Characteristic (ROC) analysis is done and the results are also evaluated by the radiologists. Experimental results show that PNN is a good classifier, giving an accuracy of 95% for classifying fatty and cirrhosis liver using wavelet based statistical texture features.

Figure optionsDownload as PowerPoint slide