Detection of Small Lesions in Medical X-Ray Image Based on Multi-Scale Fractal Feature

Abstract

X-ray technology is being increasingly employed for the clinical diagnosis and remedies in recent years. In order to distinguish small lesions in a medical X-ray image correctly, digital image processing technology is linked with clinical diagnosis. A method in medical image processing based on multiscale fractal feature (D <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">MF</sub> ) was applied in this paper. The D <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">MF</sub> values of different regions in a medical image are normally different. According to the values of the D <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">MF</sub> , small lesions in a medical X-ray image can be detected. This method has been applied to a digital bone fracture X-ray image. Experimental result showed that the D <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">MF</sub> values of the background regions are between 0.005 and 0.020; the D <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">MF</sub> values of the normal regions are between 0.040 and 0.070; and the D <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">MF</sub> values of the knot regions are between 0.100 and 0.200. This method has stronger distinguishing ability, showing better performance in clinical diagnosis.