Construction of highly-functional PET image reconstruction methods for multi-modality age

2014 Fiscal Year Final Research Report

• Project/Area Number: 24601004
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: 医学物理学・放射線技術学
• Research Institution: University of Tsukuba
• Principal Investigator: KUDO HIROYUKI 筑波大学, システム情報系, 教授 (60221933)
• Co-Investigator(Renkei-kenkyūsha): TAKIZAWA Hotaka 筑波大学, システム情報系, 准教授 (40303705) ; WANG Zhen 筑波大学, 図書館情報メディア系, 准研究員 (30593579)
• Project Period (FY): 2012-04-01 – 2015-03-31
• Keywords: 医用画像 / ＰＥＴ / 画像処理 / 画像再構成 / 逆問題 / マルチモダリティ

Outline of Final Research Achievements

Recently, multi-modality PET scanners have been developed and practically used in diagnostic medicine, which allows to generate a PET functional image and an MRI/CT morphological image simultaneously. However, these scanners still employ rather old tomographic image reconstruction methods to generate PET images from measured projection data. In this research, we develop three conceptually new image reconstruction methods for the multi-modality scanners. These methods include 1) a reconstruction method which uses an MRI/CT image as a priori knowledge to improve PET image quality significantly, 2) a reconstruction method which outputs a normal image and a lesion image separately by using techniques of computer-aided-diagnosis field during the reconstruction processing, and 3) a real-time reconstruction method using list-mode PET data which provides an intermediate PET image on line during the measurement and also provides a final PET image immediately after the measurement.

Free Research Field

医用画像工学