Texture analysis of the ovarian lesions by CT scan images

Introduction: To explore diagnostic potential of computerize texture analysis methods in discrimination of the normal, benign and malignant ovarian lesions by CT scan imaging.   Materials and Methods: Ovarian CT image database consists of 10 normal, 10 benign and 3 malignant which were reported by radiologist and proven by clinical examination. Region of interests (ROI) were defined within the lesions part of the images. Gray level intensity within a ROIs in abnormally part of the image normalized by: N1: default, N2: µ+/- 3σ, where µ and σ respectively were the mean value and standard deviation of the gray level intensity and N3: present intensity limited to 1%-99%. Up to 270 texture features parameters computed for each ROI per normalization schemes. Among them, we selected subsets of ten best discriminating features based on two reduction methods: Fisher (F) coefficient and or the probability of classification error plus average correlation coefficient (POE+ACC). The selected features sets under standard and un-standard states applied for texture analysis with principle component analysis (PCA) and linear discrimination analysis (LDA). The first nearest neighbor (1-NN) classifier was performed for features obtained via PCA and LDA respectively to differential diagnosis benign versus malignant ovarian lesions. Finally, the discrimination performance of the applied texture analysis methods were evaluated by Receiver Operating Characteristic (ROC) curve analysis by calculation sensitivity, specificity and are under the ROC curve ( AZ value). Results: In differentiation of normal from benign ovarian lesions, PCA in comparison with LDA, represent excellent performance with a sensitivity 96.7%, specificity 80% and A z value of 0.9. Also in differentiation of benign from malignant ovarian lesions, again PCA represent highly performance with a sensitivity 82.8% , specificity 96.5% and A z value of 0.89. Conclusion: Computerize texture analysis has high potential to promote radiologist’s confidence for discrimination of ovarian lesions on CT scan images with no need other examination.