相关实验视频

Updated: Mar 3, 2026

Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping

Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping

Published on: September 25, 2019

在神经成像中基于学习的注册的等级不确定性估计

Xiaoling Hu¹, Karthik Gopinath¹, Peirong Liu¹

¹Massachusetts General Hospital and Harvard Medical School.

... International Conference on Learning Representations

|March 2, 2026

概括

这项研究引入了一种基于深度学习的图像注册的不确定性估计的新方法,提高了脑MRI扫描的准确性. 与现有技术相比,新方法可以更好地预测注册错误.

科学领域:

医疗成像医学成像
神经成像是一种神经成像.
人工智能的人工智能

背景情况:

基于深度学习的图像注册在医学成像中显示出高精度,特别是在使用MRI进行的人类神经成像时.
目前的不确定性估计方法,如蒙特卡洛学,是通用的,不利用特定领域的空间建模.

研究的目的:

提出一个原则框架,用于将不确定性从空间位置传播到图像注册中的全球转换模型.
通过结合不确定性意识的适配来提高脑MRI扫描注册的准确性.

主要方法:

开发了一个框架,从局部空间估计到全球转换模型传播认识或异常的不确定性.
利用高斯分布进行局部不确定性建模,并探索分层不确定性传播.
将拟议的不确定性估计与蒙特卡洛退出与参考登记错误进行了比较.

主要成果:

拟议的不确定性估计显示,与登记错误的相关性要比蒙特卡洛退出更好.
对变换的不确定性意识调整显著提高了脑MRI扫描的记录准确性.
证明了从下游神经成像任务的变换后部分布采样的实用性.

结论:

拟议的方法提供了一种原则和有效的方式来估计和传播基于深度学习的图像注册中的不确定性.
这种方法可以提高注册准确性,并为下游神经成像应用中不确定性传播提供基础.

更多相关视频

Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images

Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images

Published on: April 13, 2013

Author Spotlight: An Efficient and Robust Software for Automated Fusion of Multiple Preclinical Imaging Modalities

Author Spotlight: An Efficient and Robust Software for Automated Fusion of Multiple Preclinical Imaging Modalities

Published on: October 27, 2023

相关概念视频

Uncertainty: Overview

Uncertainty: Overview

In analytical chemistry, we often perform repetitive measurements to detect and minimize inaccuracies caused by both determinate and indeterminate errors. Despite the cares we take, the presence of random errors means that repeated measurements almost never have exactly the same magnitude. The collective difference between these measurements - observed values - and the estimated or expected value is called uncertainty. Uncertainty is conventionally written after the estimated or expected value.