Lung Function and Structure Reconstruction Based on MR-CT Image Integration with Motion Field Analysis

2014 Fiscal Year Final Research Report

• Project/Area Number: 24500539
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Medical systems
• Research Institution: Yokohama National University
• Principal Investigator: GOTOH TOSHIYUKI 横浜国立大学, 環境情報研究科(研究院), 教授 (30234991)
• Co-Investigator(Kenkyū-buntansha): KAGEI Seiichiro 横浜国立大学, 環境情報研究院, 名誉教授 (20017966)
• Project Period (FY): 2012-04-01 – 2015-03-31
• Keywords: 画像診断システム / 医用画像処理 / 3次元ＣＴ / 時系列ＭＲＩ / ＣＯＰＤ / モダリティ統合 / 肺野構造解析

Outline of Final Research Achievements

In this study, a registration method for finding correspondences between feature points of MR and CT images was investigated, using a principal motion pattern model which is applicable for even COPD (Chronic Obstructive Pulmonary Disease) or IP (Interstitial Pneumonia). A new pulmonary blood flow analysis for contrast-enhanced MR was then developed, based on a two-input model including the pulmonary artery and the recirculation from the aorta into the lung. We have also proposed a multi-modality MR-CT integration using lung vessel determination. Then, an automated anatomy structure segmentation method, such as the lobe, airway tree and pulmonary vessel tree, in a chest CT was developed, based on newly proposed a failure tracking and recovery algorithm.
Integrating the proposed methods, we have also developed a prototype system to investigate diffuse lung diseases diagnosis and/or blood flow distributions in each lung structure, in order to evaluate the availability for clinical uses.

Free Research Field

総合領域