Three-Dimensional Shape Modeling and Analysis of Brain Structures

Jaeil Kim; Maria  del Carmen Vald&#233;s Hern&#225;ndez; Jinah Park

doi:10.3791/59172

JoVE Journal
Neuroscience

A subscription to JoVE is required to view this content. Sign in or start your free trial.

JoVE Journal Neuroscience

Three-Dimensional Shape Modeling and Analysis of Brain Structures

脳構造の3次元形状モデリングと解析

Full Text

7,617 Views

05:33 min

November 14, 2019

DOI: 10.3791/59172-v

Jaeil Kim¹, Maria del Carmen Valdés Hernández², Jinah Park³

¹School of Computer Science and Engineering,Kyungpook National University, ²Centre for Clinical Brain Sciences,University of Edinburgh, ³School of Computing and KI for Health Science and Technology (KIHST),Korea Advanced Institute of Science and Technology (KAIST)

Please note that some of the translations on this page are AI generated. Click here for the English version.

Overview

This study introduces a semi-automatic protocol for 3D shape analysis of brain structures, focusing on hippocampal segmentation from brain MRI images. The methodology involves open software for image segmentation followed by group-wise shape analysis using an automated modeling package.

Key Study Components

Area of Science

Neuroscience
Image Analysis
Structural Brain Modeling

Background

Accurate shape recovery is essential for anatomical correspondence in brain models.
The framework includes tools for shape modeling and deformity computation.
Used with large human brain datasets for various studies.
The software was developed by Dr. Jaeil Kim and demonstrates user-friendly features.

Purpose of Study

To demonstrate a procedure for hippocampal segmentation and shape analysis.
To provide an automated framework for modeling individual and group brain shapes.
To offer tools for statistical analysis of shape variations.

Methods Used

The method utilizes a graphic user interface for MRI image and segmentation editing.
The study focuses on hippocampal structures using T1-weighted magnetic resonance images.
Users manually edit hippocampal segmentations and construct group templates.
Statistical analysis is performed using shape deformity measurements.
MATLAB code is provided for analysis at the project page.

Main Results

The approach allows for precise shape modeling of the hippocampus and computation of shape deformities.
Results demonstrate differences in hippocampal shape between groups with varying brain tissue volumes.
Individual shape characteristics are restored while minimizing distortion during modeling.
Visualization of aligned models and average shape deformity maps is included.

Conclusions

This protocol enables effective shape analysis and modeling of hippocampal structures.
The methodology enhances understanding of anatomical variations in brain research.
Applications extend to clinical studies involving conditions like Alzheimer's disease and other structural anomalies.

Frequently Asked Questions

What are the advantages of using this protocol for shape analysis?

This semi-automatic protocol enhances accuracy in shape modeling while reducing user effort through automation. It provides a robust framework for large datasets.

How is the hippocampal segmentation performed?

Segmentation begins with automatic results from the MRI, followed by manual editing to ensure that critical structures like the uncus are included in the mask.

What types of data can be obtained from this analysis?

The analysis yields detailed shape models, deformation measurements, and average shape deformity maps that reveal structural differences among populations.

Can the method be adapted for other brain structures?

Yes, while focused on the hippocampus, the framework can be applied to other brain structures requiring similar shape analysis methods.

What are the key considerations for using this approach?

Users should remain involved in critical steps that require confirmation, such as adjusting intensity parameters to fit segmentation results accurately.

How is statistical analysis integrated into this protocol?

Statistical analysis is performed on shape deformities to explore variations and correlations relevant to clinical conditions and patient populations.

オープンソフトウェアを用いた画像セグメンテーションをはじめ、自動モデリングパッケージを用いたグループ単位の形状解析など、脳構造の形状解析用の半自動プロトコルを紹介します。ここでは、脳MR画像からの海馬セグメンテーションを用いた3D形状解析プロトコルの各ステップを示す。

ラフで騒々しいセグメンテーションに対して形状特徴を正確に回復することは、個々の脳形状モデル間の良好な解剖学的対応を達成するために重要です。当社のフレームワークは、個々の形状モデリング、グループ単位のテンプレート構築、形状変形計算のための様々なツールを提供します。そして、それは人間の脳の大規模なデータセットのために使用されています。

この手順を実証することは、脳の形状モデリングのためのソフトウェアを開発した私の研究室の元大学院生、ジェイル・キム博士です。海馬セグメンテーションの手動編集のために、T1重量磁気共鳴画像と自動海馬セグメンテーション結果をグラフィックユーザーインターフェースソフトウェアで開きます。表示ウィンドウのアイコンをクリックしてコロナビューを選択し、ボリュームをスクロールして、uncus が見つかるまでスクロールします。

View the full transcript and gain access to thousands of scientific videos