文章摘要

前列腺全器官病理切片与前列腺多参数磁共振图像 精准匹配的方法探究

作者: 1,2吕晓宇, 1,2张青, 3付尧, 4李铭, 1,2曹文敏, 1,2江波, 5陆加明, 1,2刘光香, 1,2郭宏骞
1 南京大学医学院附属鼓楼医院泌尿外科;
2 南京大学泌尿外科学研究所，南京 210008
3 南京大学医学院附属鼓楼医院病理科，南京 210008
4 南京大学医学院附属鼓楼医院影像科，南京 210008
5 医学院附属鼓楼医院影像科，南京 210008
通讯: 郭宏骞 Email: dr.ghq@163.com
DOI: 10.3978/j.issn.2095-6959.2017.11.013

摘要

目的: 分析和比较常规方法与改良后方法(IUNU法)在前列腺全器官切片与MRI图像匹配方面 的优劣。方法: 2017年2月至5月期间共收集前列腺癌根治标本48例，术前均行3.0T多参数磁 共振(multiparameter MRI，mpMRI)检查，将标本随机分两组，分别按常规方法和IUNU法进行 取材，对两种方法取材所得的组织块的数量和厚度进行比较分析;再将获得的病理切片按照 两种方法分别与术前的MRI图像进行匹配，比较两种方法获得的病理切片和MRI图像层面匹 配的结果有无差异。最后比较配对的病理和MRI图像两者的相似程度，从而对两种方法在精 准匹配方面进行评估。结果: 常规方法取材获得24(标本尖端)+176个组织块，IUNU法取材获 得24(标本尖端)+198组织块。此外，IUNU法获得的组织块平均厚度较常规法更薄，更均一 [(2.98±0.247) mm vs (3.64±0.49) mm，P<0.0001]。IUNU法取材时剖面的偏斜程度在前后方向 [(1.252±1.080) vs (0.582±0.521) mm，P<0.001]和左右方向[(0.873±0.891) mm vs (0.504±0.444) mm，P=0.07] 均更小。获得的病理切片按照常规方法与mpMRI图像进行匹配所对应的层面和IUNU法进行匹配有 92/374张(24.6%)发生了变化。两种方法的最终配对结果平均分为3.250和4.129，差异有统计学意义 (P<0.001)。结论: IUNU法在前列腺全器官切片制作和匹配上体现出了非常明确的优势，该方法可进一 步提高前列腺全器官切片和mpMRI图像的匹配精确程度。
关键词： 前列腺癌 前列腺全器官切片 多参数磁共振 取材器 图像匹配

Exploration of more accurate methods in matching whole- mount histopathology of prostates to multi-parameters of magnetic resonance images

Authors: 1Lv Xiaoyu, 1Zhang Qing, 2Fu Yao, 3Li Ming, 1Cao Wenmin, 1Jiang Bo, 3Lu Jiaming, 1Liu Guangxiang, 1Guo Hongqian
1 Department of Urology, Drum Tower Hospital, Medical School of Nanjing University; Institute of Urology, Nanjing University, Nanjing 210008, China
2 Department of Pathology, Drum Tower Hospital, Medical School of Nanjing University, Nanjing 210008, China
3 Department of Radiology, Drum Tower Hospital, Medical School of Nanjing University, Nanjing 210008, China

CorrespondingAuthor: Guo Hongqian Email: dr.ghq@163.com

DOI: 10.3978/j.issn.2095-6959.2017.11.013

Abstract

Objective: To analyze and compare the advantages and disadvantages of the routine method and the modified methods in matching the image of multi-parameters MRI (mpMRI) to whole-mount histopathology of prostate precisely. Methods: We collected 48 patients who undergone prostatectomy and conducted 3.0T mpMRI before the surgery between February 2017 and May 2017. e patients were divided into two groups: a control group used routine method to take tissue sample, and an experimental group used modified method. And then we matched the whole-mount histopathology with the MRI images by the two methods respectively. We compared the di erence of results. Finally, we evaluated the accuracy of the two methods through the similarity between the paired images of pathology and MRI. Results: We obtained 176+24 (apex) tissue blocks by the routine method but 198+24(apex) ones were obtained by the modi ed method. Furthermore, tissue blocks obtained by the modi ed method was thinner and more homogeneous [(2.98 ±0.247) mm vs (3.64±0.49) mm, P<0.0001] for average thickness than by the routine method. The gradient of section obtained by modified method was smaller than routine method in coronal and sagittal view. Compared with the routine method, there were 92 pathology sections which are matched to different mpMRI images by two different methods of all 374 whole-mount histopathology. e nal matching scores of the two methods were 3.250 vs 4.129 (P<0.001). Conclusion: e modi ed method has signi cantly advantages in matching hispathology section to mpMRI images. We can improve the alignment a er introduced modi ed method into the future work.
Keywords： prostatic cancer whole-mount histopathology multiple parameter magnetic resonance image draw material device match

文章选项

下载引文

分享