The development of probabilistic-atlas-guided DOT of the infant head

2019 Fiscal Year Final Research Report

• Project/Area Number: 16K16651
• Research Category: Grant-in-Aid for Young Scientists (B)
• Allocation Type: Multi-year Fund
• Research Field: Basic / Social brain science
• Research Institution: Tokyo Metropolitan University (2017-2019); Yuge National College of Maritime Technology (2016)
• Principal Investigator: Tsuzuki Daisuke 首都大学東京, 人文科学研究科, 特任准教授 (50646346)
• Project Period (FY): 2016-04-01 – 2020-03-31
• Keywords: 乳幼児 / 拡散光イメージング / 参照脳

Outline of Final Research Achievements

We developed a “probabilistic-atlas-guided DOT (pDOT)” approach that uses multiple individual MRI volumes as referential atlases and combines the resulting images to obtain a probabilistic representation. The pDOT based on multiple referential-atlases model can be used without acquisition of a subject-specific structural MRI and circumvent systematic localization biases involved in single-atlas-guided DOT. Thus, this method offers a significant improvement in spatial localization accuracy, approaching that obtained when using a subject-specific anatomy to guide the DOT image reconstruction. Moreover, we developed a pipeline which linearly transforms adult's head to infant's with anatomical labeling via cortical landmarks. This systematic expansion realizes overwrap the sensitivity profiles and reconstructed images in the anatomical information at the gyrus level.

Free Research Field

コンピュータビジョン

Academic Significance and Societal Importance of the Research Achievements

本研究において提案した確率的 DOT は、複数の脳アトラスを参照して光伝播のシミュレーションならびに脳機能断層像の作成を行う脳機能イメージング法であり、この概念自体に学術的な新規性がある。ならびに、得られた脳機能断層像に脳回レベルの解剖情報を加えてデータを解釈するための基礎的なパイプラインを開発し、精度の高い非侵襲な乳幼児の光脳機能イメージング実現に向けて、嚆矢を放つことができたと考えられる。