Search research articles

关于 JoVE

概览领导团队博客 JoVE 帮助中心

作者

出版流程编辑委员会范围与政策同行评审常见问题投稿

图书馆员

用户评价订阅访问资源图书馆顾问委员会常见问题

研究

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments 存档

教育

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual 教师资源中心教师网站

使用条款与条件

相关概念视频

您也可能阅读

相关文章

通过共同作者、期刊和引用图与本文相关的文章。

排序

Same author

Surgical outcomes of percutaneous pedicle screw fixation for unstable traumatic thoracolumbar fractures: a meta-analysis focused on randomized controlled trials from the past 15 years.

Journal of orthopaedic surgery and research·2026

Same author

Latitudinal variation in morphological patterns of lens opacity among patients with cataracts.

International ophthalmology·2026

Same author

Copper induces depressive-like behavior through neuronal lipid accumulation mediated by Lipin-1.

Neurotoxicology·2026

Same author

Association between intrarenal venous flow and postoperative acute kidney injury in cardiac surgery patients following cardiopulmonary bypass: a prospective cohort study.

BMC cardiovascular disorders·2026

Same author

Structural basis for TCR recognition of a Rac1 neoantigen arising from anchor residue mutation.

Journal of structural biology·2026

Same author

Hierarchically Chiral Silver Nanoclusters Mediated Enantioselective Glutathione Depletion and Intracellular Self-Assembly for Enhanced Anticancer Therapy.

Angewandte Chemie (International ed. in English)·2026

Search research articles

相关实验视频

Updated: Jun 17, 2025

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

证据建模用于可靠性学习和可解释的决策在多模态医疗图像细分的医疗图像细分下.

Jianfeng Zhao¹, Shuo Li²

¹School of Biomedical Engineering, Western University, London, ON, Canada.

Computerized medical imaging and graphics : the official journal of the Computerized Medical Imaging Society

|August 8, 2024

概括

本研究介绍了用于多模式医疗图像细分的上下文折扣证据网络 (CDE-Net),增强可靠性学习和可解释的决策. CDE-Net有效地融合了来自不同成像类型的信息,提高了细分的准确性和可解释性.

关键词:

可以解释性解释性多种方式的多样性.可靠性学习学习的可靠性分段化分段化分段化分段化视觉解释视觉解释

更多相关视频

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

Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application

Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application

Published on: April 14, 2023

相关实验视频

Last Updated: Jun 17, 2025

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

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

Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application

Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application

Published on: April 14, 2023

科学领域:

医疗成像医学成像
人工智能的人工智能
计算机视觉计算机视觉

背景情况:

多模式医疗图像细分至关重要,但在可靠性评估和决策解释性方面面临挑战.
现有的方法往往缺乏透明度,他们如何权衡不同的成像模式.
决策过程的可解释性,例如使用软max,在多模式融合中是有限的.

研究的目的:

提出一种新的方法,即上下文折扣证据网络 (CDE-Net),用于可靠和可解释的多模式医疗图像细分.
开发一个框架,学习每个成像模式的可靠性.
在多模式图像融合中提高决策的可解释性.

主要方法:

CDE-Net使用证据决策模块模拟语义证据,并测量不确定性.
使用上下文折扣的融合层来学习模式特定的可靠性.
多级损失函数优化了证据建模和可靠性学习.

主要成果:

在多模式细分任务中,CDE-Net实现了高性能.
获得了0.914的脑瘤细分和0.913的肝瘤细分的平均子得分.
该框架通过像素归因图的一致性和学习可靠性系数来证明可解释性.

结论:

在多模式医疗图像细分中,CDE-Net为可靠性学习和可解释的决策提供了一个强大的解决方案.
拟议的方法显示了在医学图像融合中推进人工智能的巨大潜力.
CDE-Net的可解释性功能有助于更好地理解人工智能驱动的医学图像分析.