相关实验视频

Updated: Jul 2, 2026

High-Resolution Endocardial and Epicardial Optical Mapping in a Sheep Model of Stretch-Induced Atrial Fibrillation

High-Resolution Endocardial and Epicardial Optical Mapping in a Sheep Model of Stretch-Induced Atrial Fibrillation

Published on: July 29, 2011

增强学习的3D纤维导向映射在厚心脏组织中

Eda Nur Saruhan¹, Hakancan Ozturk², Demet Kul³

¹Computer Science and Engineering, Koç University, Rumelifeneri, Istanbul, 34450, Turkey.

Biomedical optics express

|August 14, 2025

概括

机器学习和深度学习准确地分析厚组织中的3D心血管纤维结构. 与传统技术相比,这些先进的方法为计算建模和组织表征提供了更高的空间精度.

科学领域:

生物医学工程生物医学工程
计算生物学计算生物学
材料科学材料科学材料科学

背景情况:

纤维蛋白,如弹性素和原蛋白,对于心血管结构完整性至关重要.
对于薄组织,已经建立了2D纤维定向映射,但对于厚样品 (例如胚胎心脏) 缺乏强大的3D分析工具.
3D纤维定向数据对于计算血管建模和特征组织微观结构至关重要.

研究的目的:

开发和评估机器学习 (ML) 和深度学习 (DL) 技术,用于分析厚组织样本中的3D心血管纤维结构.
为了比较ML / DL方法的性能与传统的福里埃转换和定向波器方法.
评估这些先进技术在提高空间精度和减少纤维定向分析中人工预处理方面的潜力.

主要方法:

采用了ML / DL模型,包括支持向量回归 (SVR),卷积神经网络 (CNN) 和剩余网络-50 (ResNet50),在合成和真实对焦成像数据集上接受了培训.
利用注意力机制,特别是通过ResNet50引导注意力,以提高模型性能.
在混合数据集 (1200个样本) 和生物数据集 (400个猪/牛样本) 上评估模型.

主要成果:

支持向量回归 (SVR) 在混合数据集上达到5.0%,在生物数据集上达到13.0%的正常化平均绝对误差 (nMAE).
深度学习模型CNN和ResNet50显示nMAE分别为12.0%和11.0% (混合) 和23.0%和22.0% (生物).
频道注意力ResNet50提高了性能,达到5.8% (混合) 和21.0% (生物) 的nMAE,证明了注意力机制的好处.

结论:

ML和DL技术显示出在厚组织样本中精确3D心血管纤维方向检测的巨大潜力.
与传统技术相比,这些方法提供了更好的空间精度,并减少了对手工预处理的依赖.
这项研究可以进行更详细的心血管微结构评估,这对于计算建模和组织工程至关重要.

更多相关视频

Patient-specific Modeling of the Heart: Estimation of Ventricular Fiber Orientations

Patient-specific Modeling of the Heart: Estimation of Ventricular Fiber Orientations

Published on: January 8, 2013

Simultaneous Brightfield, Fluorescence, and Optical Coherence Tomographic Imaging of Contracting Cardiac Trabeculae Ex Vivo

Simultaneous Brightfield, Fluorescence, and Optical Coherence Tomographic Imaging of Contracting Cardiac Trabeculae Ex Vivo

Published on: October 2, 2021

相关概念视频

Studying the Cytoskeleton

Studying the Cytoskeleton

The cytoskeletal architecture can be studied using different microscopic and biochemical techniques. Electron microscopy was instrumental in discovering the cytoskeletal architecture around the 1960s, which allowed obtaining structural information at a high-resolution level. However, the sample preparation procedure often limits this ability in biological samples. Several protocols have been developed over the years to optimize sample preparation. In one of the protocols known as rotary...

Three-Dimensional Microscopy in Microbiology

Three-Dimensional Microscopy in Microbiology

Three-dimensional imaging techniques are essential in cell biology, allowing researchers to visualize intricate cellular structures with high resolution. Two prominent methods, Differential Interference Contrast Microscopy (DIC) and Confocal Scanning Laser Microscopy (CSLM), provide distinct advantages for imaging live and thick specimens, respectively.Differential Interference Contrast MicroscopyDIC microscopy enhances contrast in transparent, unstained samples by converting phase...