LIDC

在处理LIDC_IPDI时，处理到后期发现一个问题，在每一个病例中，切片的命名的编号，跟dicom中给出的命名的编号不一致，这在处理数据时，无形的增加则一些难度，尤其是在运行别人的代码时，这个问题尤为突出，经过不断的搜索，最终在官网的一个页面上发现了这个问题的一种解决方式，先付上官网的连接：

https://wiki.cancerimagingarchive.net/pages/viewpage.action?pageId=24283641 如何下载LIDR_IRDI数据就不多说了，很多博客中都讲的很细致，下面具体的描述一下我遇到的问题。 从NBIA上下载下来数据后，是这样的 打开第一个病例，如图所示： 其中问题出在，此时的第一个文件对应的切片的标号为：80 但是其实我想要的效果是，这样的，标号为000.dcm的文件里存放的时切片号为1的切片：

二、解决

# noinspection PyPep8Naming import SimpleITK as sitk import glob as gb import os import shutil def safe_sitk_read(img_list, *args, **kwargs): """ Since the default function just looks at images 0 and 1 to determine slice thickness and the images are often not correctly alphabetically sorted, much slower :param img_list: :return: """ pimg_list = [] reader = sitk.ImageSeriesReader() seriesIDs = reader.GetGDCMSeriesIDs(x_path) dicom_names = reader.GetGDCMSeriesFileNames(x_path, seriesIDs[0]) reader.SetFileNames(dicom_names) image_list = reader.GetFileNames() pimg_list = [(sitk.ReadImage(x).GetOrigin(), x) for x in image_list] s_img_list = [path for _, path in sorted(pimg_list, key = lambda x: x[0][2])] print('s_img_list',s_img_list) return s_img_list x_path= 'E:/qiepianmingming/LIDC-IDRI-0002/01-01-2000-98329/3000522-04919/' #从NBIA上下载的一个病例 jieguo_path = 'E:/qiepianmingming/LIDC-IDRI-0002/1/0' #随便设的一个路径 z_sort = safe_sitk_read(x_path) i = len(z_sort)+1 for img in z_sort: i -=1 if i !=0: if i <=9: dst = os.path.join(jieguo_path+str(0)+str(0)+str(i)+'.dcm') elif i>=10 and i <= 99: dst = os.path.join(jieguo_path+str(0)+ str(i) + '.dcm') else: dst = os.path.join(jieguo_path+ str(i) + '.dcm') shutil.copy(img, dst) 有待解决的问题是，将所有的数据都进行转化

补充

通过一个程序实现所有的文件内的dcm图片按编号排列

# noinspection PyPep8Naming import SimpleITK as sitk import glob as gb import os import shutil #这个函数是为了提取每个bcm图片中的InstanceNumber，由与图片的编号用过z的大小提现，我们用z轴进行排序，找到编号 def safe_sitk_read(img_list, *args, **kwargs): """ Since the default function just looks at images 0 and 1 to determine slice thickness and the images are often not correctly alphabetically sorted, much slower :param img_list: :return: """ pimg_list = [] reader = sitk.ImageSeriesReader() seriesIDs = reader.GetGDCMSeriesIDs(img_list) dicom_names = reader.GetGDCMSeriesFileNames(img_list, seriesIDs[0]) reader.SetFileNames(dicom_names) image_list = reader.GetFileNames() pimg_list = [(sitk.ReadImage(x).GetOrigin(), x) for x in image_list] s_img_list = [path for _, path in sorted(pimg_list, key = lambda x: x[0][2])] return s_img_list #x_path是导入路径，完整的LIDC-IDRI-0应该如LIDC-IDRI-0001...LIDC-IDRI-0133，每个LIDC-IDRI-0文件夹下是study编号的文件夹，study 下是series文件夹，该文件夹下即为每个病例的所有.dcm文件 x_path = r'E:\zonghe_feijiejie\LIDC-IDRI\LIDC-IDRI-0' #存储路径，为了与原先的路径匹配起来，路径从E:\zonghe_feijiejie\shunxu_LIDC-IDRI之后都是生成的 jieguo_path = r'E:\zonghe_feijiejie\shunxu_LIDC-IDRI\LIDC-IDRI-0' for i in range(1,4): print('i',i) series1 = '' series2 = '' study1 = '' study2 = '' dst = '' jie = '' #给LIDC-IDRI-0补充 if i <= 9: dst = os.path.join(x_path + str(0) + str(0) + str(i)) jie = os.path.join(jieguo_path + str(0) + str(0) + str(i)) elif i >= 10 and i <= 99: dst = os.path.join(x_path + str(0) + str(i)) jie = os.path.join(jieguo_path + str(0) + str(i)) else: dst = os.path.join(x_path + str(i)) jie = os.path.join(jieguo_path + str(i)) ''' #path为完整的路径，例，E:\zonghe_feijiejie\LIDC-IDRI\LIDC-IDRI- 0001\1.3.6.1.4.1.14519.5.2.1.6279.6001.298806137288633453246975630178\1.3.6.1.4.1.14519.5.2.1.6279.6001.179049373636438705059720603192 ''' study1 = os.path.join(str(dst), str(os.listdir(dst)[0])) series1 = os.path.join(str(study1), str(os.listdir(study1)[0])) path = str(series1) print("path",path) z_sort = safe_sitk_read(path) study2 = os.path.join(str(jie), str(os.listdir(dst)[0])) series2 = study2 + '\\' + str(os.listdir(study1)[0]) os.makedirs(series2) z_sort = safe_sitk_read(path) #存储新的数据 #我这里z_sort是按照倒序排序的，即最后一张编号的.dcm图片排在第一位，因此输出反向 j = len(z_sort) for img in z_sort: if j <=9: cunchu = os.path.join(series2+'\\'+str(0)+str(0)+str(0)+str(j)+'.dcm') elif j>=10 and j<= 99: cunchu = os.path.join(series2+'\\'+str(0)+str(0)+ str(j) + '.dcm') else: cunchu = os.path.join(series2+'\\'+ str(0)+str(j) + '.dcm') shutil.copy(img, cunchu) j -=1