Development of non-linear registration of body CT images for subtraction to detect time-dependent changes.

2014 Fiscal Year Final Research Report

• Project/Area Number: 25670530
• Research Category: Grant-in-Aid for Challenging Exploratory Research
• Allocation Type: Multi-year Fund
• Research Field: Radiation science
• Research Institution: Kyoto University
• Principal Investigator: TOGASHI Kaori 京都大学, 医学(系)研究科(研究院), 教授 (90135484)
• Project Period (FY): 2013-04-01 – 2015-03-31
• Keywords: 肺ガン / 微分同相写像 / LDDMM / 経時変化

Outline of Final Research Achievements

This study focused on non-linear registration of serial body CT images to attain high accuracy. The algorithm used was large deformation diffeomorphic metric mapping (LDDMM) besed on the Euler-Lagrange equation.
In lung tumor cases, changes in the size, disappearance or appearance of the tumors could be evaluated by the automatic processing. In the abdomen and pelvis, the proposed method found bone metastasis, which may not necessarily be detected by conventional evaluation. The proposed method is expected to bring objective and accurate evaluation into clinical practice.

Free Research Field

画像診断学