紀要論文

非浸潤性乳管がん(DCIS)の病理画像から三次元再構成

Automated 3D reconstruction of pathological images of Ductal Carcinoma In Situ (DCIS)

jpn

21879923

https://doi.org/10.15002/00010466

法政大学大学院理工学・工学研究科

55

1

2

2014-03-24

Version of Record

Ductal Carcinoma In Situ (DCIS) is a pre-invasive carcinoma of the breast that exhibits sev- eral distinct morphologies but the link between morphology and patient outcome is not clear. We hypothesize that different mechanisms of growth may still result in similar 2D morpholo- gies, which may look different in 3D. In order to clear how DCIS grows and produced similar morphologies, we produced a fully automated algorithm that reconstructs 3D architectures from microscopy images of 2D serial sections from human specimens. Our method consists of two phases: the alignment and the segmentation phases. Our method aligned pathological images with the Scale-Invariant Feature Transform (SIFT) and segmented ducts by thresholding and morphological operations. We confirmed our method appropriately produced 3D reconstruction model of DCIS appropriately and further medical review can be expected with this model.

Article

AA12677220