相关实验视频

Updated: Jul 8, 2026

Deep Learning-Based Segmentation of Cryo-Electron Tomograms

Deep Learning-Based Segmentation of Cryo-Electron Tomograms

Published on: November 11, 2022

贝叶斯U-Net的基于不确定性的主动学习,用于多标签圆束CT细分的贝叶斯U-Net

Jiayu Huang¹, Nazbanoo Farpour², Bingjian J Yang²

¹School of Computing and Augmented Intelligence Arizona State University, Tempe, Arizona.

Journal of endodontics

|November 18, 2023

概括

积极学习 (AL) 策略显著提高了人工智能 (AI) 模型的准确性,用于检测和细分圆束CT (CBCT) 中的周围病变. 这种方法减少了在牙科AI开发中需要广泛的注释数据集的需求.

科学领域:

生物医学图像分析
人工智能在牙科中的应用
医学成像和诊断医学成像和诊断

背景情况:

训练人工智能进行生物医学图像分析需要大量的注释数据集.
这项研究的重点是改进人工智能模型培训,用于在CBCT中检测周周病变.
有限的数据集对在牙科中开发准确的人工智能模型构成了挑战.

研究的目的:

评估积极学习 (AL) 策略对训练人工智能模型的有效性.
改进CBCT中的多标签细分和检测周周病变.
用有限的注释数据来评估AL策略.

主要方法:

使用有限视野CBCT卷 (n=20) 进行培训.
采用贝叶斯式U-Net,具有两个AL函数:贝叶斯式通过不同意 (BALD) 和Max_Entropy (ME) 的积极学习.
将AL策略与非AL贝叶斯U-Net基准进行比较,评估细分精度 (Dice指数) 和病变检测.

主要成果:

经过8次AL代,病变检测灵敏度达到BALD的84.0%,ME的76.0%,显著超过非AL贝叶斯U-Net (32.0%).
与非AL基准指标 (0.680 ± 0.155) 相比,所有标签的平均子指数在AL策略中较高 (0.703 ± 0.166对于ME).
在8次AL代后",损伤"的Dice指数为BALD的0.504和ME的0.501,明显高于非AL模型的0.288.1的最大值.

结论:

主动学习策略,特别是BALD和ME,提高AI模型的准确性,用于CBCT的细分和病变检测.
借助不确定性量化,AL有效地减少了对牙科AI广泛数据注释的需求.
这种方法有望在牙科生物医学图像分析中更有效地开发人工智能工具.

关键词:

人工智能的人工智能贝叶斯的U-Net是贝叶斯的U-Net.在CBCT中,CBCT是CBCT.积极学习是积极学习.深度学习是一种深度学习.多标签细分化的多标签细分化.周围的外皮病变.

更多相关视频

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

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.