• Search Research Projects
  • Search Researchers
  • How to Use
  1. Back to project page

2022 Fiscal Year Final Research Report

An advanced image registration technique for supporting rapid medical image diagnoses

Research Project

  • PDF
Project/Area Number 19K12219
Research Category

Grant-in-Aid for Scientific Research (C)

Allocation TypeMulti-year Fund
Section一般
Review Section Basic Section 62010:Life, health and medical informatics-related
Research InstitutionKyushu Institute of Technology

Principal Investigator

Tokunaga Terumasa  九州工業大学, 大学院情報工学研究院, 准教授 (50614806)

Project Period (FY) 2019-04-01 – 2023-03-31
Keywordsイメージレジストレーション / ベイズ推論 / カーネル密度推定 / 医療画像処理 / マルコフ確率場 / EMアルゴリズム
Outline of Final Research Achievements

Based on the ideas of kernel density estimation and topology preservation, we studied general-purpose image registration techniques that work with easy pre-turning. In particular, we have focused on image registration technique that can work practically even when the images include non-corresponding regions. As a result, we established the practical image registration technique not only for tissues of relatively simple shapes but also for objects with complex structures such as blood vessels, even when there are non-corresponding regions. Our technique can be applied to a wide range of image registration tasks. But the issue of computational cost could not be completely overcome.

Free Research Field

データ科学

Academic Significance and Societal Importance of the Research Achievements

複雑な自由変形と, 対応してない領域を含む物体同士の位置合わせという, チャレンジ性の高い課題に挑んだ。 前者には, ベイズ推論に基づく位置合わせを行う際に, 事前分布としてマルコフ確率場に基づく自由変形モデルを導入した。後者には, マルコフ確率場で表現される制御点に対し, informativeな制御点とそうではない制御点のクラスタリングも同時に行うことで対応した。これは自己教師対照学習などの他分野の技術の進展に伴い得られた着想である。今後, 数理モデルとデータ駆動的アプローチを融合する位置合わせ技術として様々な展開が期待できる。

URL: 

Published: 2024-01-30   Modified: 2025-01-30  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi