Automatic cropping method of chest radiographs based on adaptive binarization

General radiography systems are widely used for medical diagnoses. Since the size of image detectors of general radiography systems is standardized in several fixed sizes, relative size of target body parts differs according to patients. The spread of digitization of X-ray imaging forces radiological technologists to crop radiographs manually after measurement for suitable use on diagnoses. This cropping operation distracts attention of radiological technologists from measurement operations and from care to patients. The purpose of this study is to establish image processing method to achieve automatic cropping of X-ray images, which reduces the radiological technologists´ burden in general radiography workflow. The proposed method utilizes adaptive binarization as pre-processing of radiographs. In this paper, we show the result of the proposed method applied for chest radiographs and the evaluation by radiological technologists.