Development of a method of automated extraction of biological fingerprints from chest radiographs as preprocessing of patient recognition and identification
Development of a method of automated extraction of biological fingerprints from chest radiographs as preprocessing of patient recognition and identification
复制标题
开发一种从胸片中自动提取生物指纹作为患者识别和识别预处理的方法
DOI：
10.1007/s12194-017-0400-y
复制
发表时间：
2017
影响因子：
1.6
通讯作者：
Morishita Junji
中科院分区：
文献类型：
--
作者：
Shimizu Yoichiro;Morishita Junji
本文介绍了一种自动化的方法提取的生物指纹（BF），包括在胸部X光片的图像方向检测的发展。通过检查目标图像与具有正确图像取向的平均男性和女性图像之间的归一化互相关值来识别和修改目标图像的图像取向。从平均图像中提取BF模板。然后，目标图像中的每个BF从平均图像中的BF的模板与目标图像中的对应BF之间示出最高互相关值的位置提取。使用我们的方法，100%（200/200）的图像方向被正确识别。如果方向被认为是不适当的，我们的算法修改它到适当的胸部图像方向。此外，对目标图像中的BF自动提取方法进行了改进。这种方法将是有用的预处理系统的病人识别和鉴定。
This paper describes the development of an automated method of extraction of biological fingerprints (BFs), including detection of image orientation in chest radiographs. The image orientation of a target image was recognized and modified by examination of normalized cross-correlation values between a target image and averaged male and female images with correct image orientation. Templates of BFs were extracted from averaged images. Then, each BF in the target image was extracted from locations that showed the highest cross-correlation value between the template of BF in the averaged image and the corresponding BF in the target image. With our method, 100% (200/200) of image orientations were recognized correctly. If the orientation was recognized as inappropriate, our algorithm modified it into the appropriate chest image orientation. In addition, the BFs automatically extracted from target images were improved. This method would be useful in a preprocessing system for patient recognition and identification.