相关实验视频

Updated: Jul 26, 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

深度学习的不确定性量化方法的可靠性深度学习在头部和部受风险器官的自我细分

Joëlle E van Aalst¹, Federica C Maruccio², Rita Simoẽs²

¹Department of Radiation Oncology, University Medical Center Groningen, Groningen, The Netherlands.

Physics in medicine and biology

|October 8, 2025

概括

放射治疗自分化的不确定性量化 (UQ) 方法显示了可比的性能,但UQ指标的选择显著影响了可靠性. 预测是评估细分信心和错误的最可靠指标.

科学领域:

辐射疗法辐射疗法
医疗成像医学成像
深度学习 (Deep Learning) 是一种深度学习.

背景情况:

深度学习自分化显著提升了放射治疗轮,但由于性能变化,需要质量保证.
手动质量保证重新引入了可变性,并减少了自动细分的节省时间的好处.
探索不确定性量化 (UQ) 以估计输出信心并提高自动细分的可靠性.

研究的目的:

为了比较常用的UQ方法和辐射疗法自我细分的指标的可靠性.
评估不同UQ方法对细分精度,信心校准和错误定位的影响.

主要方法:

使用nnU-Net框架训练了一个3D U-Net,用于对头癌患者的19个风险器官 (OAR) 进行细分.
评估了三种UQ方法:蒙特卡罗脱,深度组合建模和测试时间增强.
在10名患者中使用细分精度 (表面子),信心校准 (ECE标签) 和错误定位 (U-E重叠) 评估可靠性.

主要成果:

与没有UQ的基线相比,所有UQ方法的细分精度都稳定.
UQ方法在信心校准和错误定位方面表现出可比的可靠性.
选择UQ指标显著影响了可靠性,多类预测的表现优于差异和相互信息.

结论:

选择UQ方法对辐射疗法自分割中的不确定性图的可靠性产生重大影响.
虽然UQ方法的性能类似,但特定的UQ指标严重影响了可靠性.
在临床应用之前,仔细选择和评估UQ指标至关重要.

关键词:

自动细分的自动细分.信任度校准的使用方法深度学习的轮设计头部和部器官面临风险的风险辐射疗法辐射疗法可靠性的可靠性不确定性量化不确定性量化

更多相关视频

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

Author Spotlight: Bridging Gaps in Anatomy and Establishing a Foundation for Algorithmic Studies

Author Spotlight: Bridging Gaps in Anatomy and Establishing a Foundation for Algorithmic Studies

Published on: December 15, 2023

相关概念视频

Uncertainty in Measurement: Accuracy and Precision

Uncertainty in Measurement: Accuracy and Precision

Scientists typically make repeated measurements of a quantity to ensure the quality of their findings and to evaluate both the precision and the accuracy of their results. Measurements are said to be precise if they yield very similar results when repeated in the same manner. A measurement is considered accurate if it yields a result that is very close to the true or the accepted value. Precise values agree with each other; accurate values agree with a true value.

Data Validation

Data Validation

Method validation is a crucial process in analytical chemistry designed to confirm that a given method consistently produces reliable and high-quality results. This process is essential when a method is applied to different sample matrices or when procedural modifications are made, ensuring that the results meet acceptable standards across various applications.
Key parameters for method validation include:

Survival Tree

Survival Tree

Survival trees are a non-parametric method used in survival analysis to model the relationship between a set of covariates and the time until an event of interest occurs, often referred to as the "time-to-event" or "survival time." This method is particularly useful when dealing with censored data, where the event has not occurred for some individuals by the end of the study period, or when the exact time of the event is unknown.
Building a Survival Tree
Constructing a survival tree begins...