Using logistic regression to distinguish between fatty and fibroid masses in medical imaging (ultrasound image)‎

Abstract EN

Medical image analysis has great significance in the field of medicine, especially in non-invasive and clinical studies.

Medical imaging techniques and it analysis tools enable the physicians and Radiologists to reach at a specific diagnosis .In this study has been studying the link between (statistical model ,Computer vision and medical images) using the application binary logistic model to analyze medical imaging (Ultrasound image for breast), for distinguishing between shape of mass (Fatty & Fibroid) through selecting regions of interest (ROI) of the mass, and by extracting statistical and geometric measurements ( Mean, Standard Deviation, Circle, Solidity,…..), best logistic model was able to be estimated which is composed of significance parameters (Integrated Density, Median, Skew, Kurt), Then, It was reached a good percentage of the classification and distinction between (Fibroid) and (Fatty) masses (85.7% vs.

82.9%) with the total percentage of classification equal to 84.4