相关实验视频

Updated: Jan 6, 2026

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

FEU-Diff:一个扩散模型与模糊的证据驱动的动态不确定性融合用于医疗图像细分

Sheng Geng, Shu Jiang, Tao Hou

IEEE transactions on neural networks and learning systems

|September 16, 2025

概括

通过整合模糊证据和不确定性融合,FEU-Diff通过扩散模型增强了医疗图像细分. 这种方法提高了细分结果的准确性和细节保存.

科学领域:

人工智能的人工智能
医疗成像医学成像
计算机视觉计算机视觉

背景情况:

扩散模型是医疗图像细分的新兴生成框架.
当前的方法在适应性融合条件先验和被否定的特征方面扎.
现有的方法缺乏显式建模像素级不确定性,导致细节丢失.

研究的目的:

引入FEU-Diff,一种基于扩散的新型细分框架.
解决适应性融合和扩散模型中不确定性建模的局限性.
为了提高医疗图像细分的准确性和结构细节的保存.

主要方法:

FEU-Diff集成了模糊证据建模和不确定性融合 (UF).
一个模糊的语义增强 (FSE) 模块使用高斯成员函数和模糊逻辑模型像素级的不确定性.
一个证据动态聚变 (EDF) 模块使用迪里克莱特分布来适应特征聚变,UF量化预测差异.

主要成果:

在四个公共数据集上,FEU-Diff的性能优于最先进的方法.
取得的平均改进:1.42% 子相似系数 (DSC),1.47% 交叉与联盟 (IoU).
将第95百分点的豪斯多夫距离 (HD95) 缩小了2.26毫米,并生成可解释的不确定性图.

结论:

FEU-Diff有效地提高了医疗图像细分的准确性和完整性.
该框架的模糊证据和不确定性融合机制增强了边界划分和细节保存.
通过生成的不确定性地图,FEU-Diff提供了增强的临床解释性.

更多相关视频

A Data-Driven Approach to Quantifying Immune States in Sepsis

A Data-Driven Approach to Quantifying Immune States in Sepsis

Published on: February 7, 2025

相关概念视频

Assessment of Diffusion and Perfusion

Assessment of Diffusion and Perfusion

Understanding and evaluating diffusion and perfusion is critical in assessing a patient's respiratory and circulatory health. These processes play key roles in maintaining the body's internal environment, ensuring that tissues receive adequate oxygen while waste products are efficiently removed.
The Role of Diffusion in Respiration
Diffusion is the process by which molecules move from an area of higher concentration to an area of lower concentration. In the respiratory system, this...