Meta anatomical information-oriented medical image processing - new medical image processing in post big data era

2016 Fiscal Year Final Research Report

• Project/Area Number: 26560255
• Research Category: Grant-in-Aid for Challenging Exploratory Research
• Allocation Type: Multi-year Fund
• Research Field: Medical systems
• Research Institution: Nagoya University
• Principal Investigator: Mori Kensaku 名古屋大学, 情報連携統括本部, 教授 (10293664)
• Research Collaborator: Daniel Ruckert インペリアルカレッジロンドン
• Project Period (FY): 2014-04-01 – 2017-03-31
• Keywords: 医用画像処理 / セグメンテーション / 機械学習 / 解剖構造認識 / メタ解剖

Outline of Final Research Achievements

This project aims to create new academic field called Meta anatomical information-oriented medical image processing under the prediction on big change of medical imaging devices. The research project achieved research outcomes in the topics including: (a) definition of target organs in meta-anatomy and its database format, (b) meta-anatomical structure recognition based on conditional random field, (c) meta-anatomical structure extraction from deep learning, (4) automated estimation of bounding boxes of target organ areas, (5) atlas-based anatomical structure recognition using meta-anatomical information database, (6) automated anatomical label assignments to blood vessels based on conditional random field, and (7) evaluation from clinical viewpoints.

Free Research Field

画像処理