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

Real-Time Cardiac Arrhythmia Classification Using TinyML on Ultra-Low-Cost Microcontrollers: A Feasibility Study for Resource-Constrained Environments.

Bioengineering (Basel, Switzerland)·2026

Same author

Non-Glycemic Clinical Data for Type 2 Diabetes Detection in Mexican Adults: A Comparative Analysis of Atherogenic Indices, Statistical Transformations, and Machine Learning Algorithms.

Diagnostics (Basel, Switzerland)·2026

Same author

Heart Sound Classification with MFCCs and Wavelet Daubechies Analysis Using Machine Learning Algorithms.

Diagnostics (Basel, Switzerland)·2026

Same author

Assessing ResNeXt and RegNet Models for Diabetic Retinopathy Classification: A Comprehensive Comparative Study.

Diagnostics (Basel, Switzerland)·2025

Same author

Convolutional Neural Network for Depression and Schizophrenia Detection.

Diagnostics (Basel, Switzerland)·2025

Same author

Evaluating Feature Selection Methods for Accurate Diagnosis of Diabetic Kidney Disease.

Biomedicines·2025

Search research articles

相关实验视频

Updated: Jan 7, 2026

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

Published on: August 16, 2020

可解释的深度学习用于数字和对比增强的乳房摄影中的乳房损伤分类.

Samara Acosta-Jiménez¹, Miguel M Mendoza-Mendoza¹, Carlos E Galván-Tejada¹

¹Unidad Académica de Ingeniería Eléctrica, Universidad Autónoma de Zacatecas, Jardín Juárez 147, Centro, Zacatecas 98000, Mexico.

Diagnostics (Basel, Switzerland)

|December 30, 2025

概括

这项研究比较了人工智能 (AI) 在乳腺癌查中的表现,使用数字乳腺学 (DM) 和对比增强光谱乳腺学 (CESM). 结果显示,这两种模式都产生了可解释的AI结果,CESM在某些分类中表现出色,DM在其他分类中表现出色.

关键词:

美国有线电视新闻网 (CNN)乳腺癌乳腺癌乳腺癌与对比度增强的光谱乳房学显微镜.数字乳房造 mammography 数字乳房造 mammography 数字乳房造 mammography 是一个很好的方法.可以解释的人工智能AI

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

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Published on: October 10, 2018

相关实验视频

Last Updated: Jan 7, 2026

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

Published on: August 16, 2020

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

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Published on: October 10, 2018

科学领域:

医疗成像医学成像
人工智能的人工智能
在瘤学瘤学.

背景情况:

人工智能 (AI) 在乳腺癌查方面表现有前途.
需要进一步探索人工智能在数字乳房镜像 (DM) 和对比增强光谱乳房镜像 (CESM) 等乳房镜像模式中的整合.
了解跨DM和CESM的AI模型行为对于公平的临床采用至关重要.

研究的目的:

评估和比较三个深度学习卷积神经网络 (CNN) 架构 (ResNet-18,DenseNet-121,EfficientNet-B0) 在DM和CESM乳腺损伤分类上的性能.
通过使用SHapley添加式扩展 (SHAP) 来分析这两种乳房摄影模式的AI模型的决策模式.

主要方法:

利用公共CDD-CESM数据集,包括2006图像用于二进制分类任务 (正常与良性,良性与恶性,正常与恶性).
在DM和CESM图像上单独训练CNN模型,使用转移学习,加权二进制交叉损失和三倍交叉验证方案.
使用SHAP分析可视化和解释像素级别的模型决策.

主要成果:

在正常与良性和良性与恶性分类中,CESM表现优越.
在正常与恶性疾病的比较中,数字哺乳镜 (DM) 获得了最高的辨别能力 (EfficientNet-B0:AUC=97%,精度=93.15%).
SHAP分析显示,对于这两种模式,一致的,解剖学上相关的决策模式,CESM提供了更清晰的焦点,DM显示了更广泛的关注.

结论:

通过功能对比,CESM增强了损伤歧视,而DM通过CNN提供了非常准确和可解释的结果,尽管获取更简单.
跨模式的基于SHAP的一致相关性证实DM和CESM都保留了用于AI分析的临床上有意义的信息.
这项研究是第一个使用可解释的深度学习模型在相同条件下直接比较DM和CESM的研究.