Jove
Visualize
联系我们
JoVE
x logofacebook logolinkedin logoyoutube logo
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 JournalMethods CollectionsJoVE Encyclopedia of Experiments存档
教育
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab Manual教师资源中心教师网站
使用条款与条件
隐私政策
政策

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

Spatiotemporal synergy of a coordination decomposition bifunctional strategy for ultra-stable aqueous zinc anodes.

Chemical science·2026
Same author

Prognostic factors and the role of pelvic lymph node dissection in oligometastatic prostate cancer: a multicenter exploratory study.

Surgical endoscopy·2026
Same author

Anterior cingulate cortex neuron subtypes differentially regulate seizures.

Epilepsia·2026
Same author

A deep learning model based on combining surface and esophageal ECG data for diagnosis of paroxysmal supraventricular tachycardia.

Digital health·2026
Same author

A Near-Infrared-Triggered Luminescence-Activated System for In Vivo Biomacromolecular Tagging and Photocatalytic Crosslinking for Large-Scale Investigation of RNA-Protein Complexes in Living Mice.

Journal of the American Chemical Society·2026
Same author

The exosomal miRNA/FOXN3 axis empowers fibroblasts with tumor-promoting functions via enhanced migration and altered secretome.

Journal of nanobiotechnology·2026

相关实验视频

Updated: May 10, 2025

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention
06:37

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Published on: December 15, 2023

2.4K

MAL-Net:一个多标签深度学习框架,集成LSTM和多头注意力,用于使用临床传感器数据对IgA脏病亚型的增强分类.

Hongyan Wang1, Yuehui Liao1, Li Gao2

  • 1School of Medical Technology and Information Engineering, Zhejiang Chinese Medical University, Hangzhou 310053, China.

Sensors (Basel, Switzerland)
|April 28, 2025
PubMed
概括

一个新的深度学习模型,MAL-Net,使用各种临床数据准确地分类IgA脏病 (IgAN) 亚型. 这一进步有助于Igan患者的早期诊断和个性化治疗.

关键词:
一种IgA脏病 (IgAN) 的治疗方法注意力机制注意力机制临床传感器 临床传感器长时间的短期记忆 (LSTM)多个标签的分类.分类类型分类类型分类类型分类

更多相关视频

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches
09:47

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches

Published on: December 15, 2023

899
Supervised Machine Learning for Semi-Quantification of Extracellular DNA in Glomerulonephritis
09:16

Supervised Machine Learning for Semi-Quantification of Extracellular DNA in Glomerulonephritis

Published on: June 18, 2020

6.7K

相关实验视频

Last Updated: May 10, 2025

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention
06:37

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Published on: December 15, 2023

2.4K
Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches
09:47

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches

Published on: December 15, 2023

899
Supervised Machine Learning for Semi-Quantification of Extracellular DNA in Glomerulonephritis
09:16

Supervised Machine Learning for Semi-Quantification of Extracellular DNA in Glomerulonephritis

Published on: June 18, 2020

6.7K

科学领域:

  • 腎臟病學 (nephrology) 是一種醫學專業.
  • 人工智能的人工智能
  • 医疗信息学 医疗信息学

背景情况:

  • IgA病 (IgAN) 是造成衰竭的主要原因,其特征是复杂的异质性.
  • 目前的分类方法与IGAN的多样化数据和重叠的症状作斗争.
  • 需要先进的工具来准确地分类Igan,以便更好地管理患者.

研究的目的:

  • 引入MAL-Net,这是一个用于多标签Igan亚型分类的深度学习框架.
  • 利用多维临床数据,包括基于传感器的输入,以改进IgAN亚型.
  • 为了应对Igan分类中的数据异质性和类不平衡的挑战.

主要方法:

  • 开发了MAL-Net,集成长期短期记忆 (LSTM) 和多头注意力 (MHA) 网络.
  • 利用内存网络从临床传感器和记录中提取特征.
  • 在500名Igan患者的数据上训练并验证了该模型,包括人口统计,实验室和症状.

主要成果:

  • 马尔-网实现了91%的准确性和0.97的AUC,超过了六个基线模型.
  • 多头注意力显著改善了分类,特别是对于罕见的Igan亚型.
  • Ni-du亚型的F1分数增加了0.8,表明有效地缓解了阶级失衡.

结论:

  • MAL-Net为多标签Igan亚型分类提供了一个强大的解决方案.
  • 该框架有效地处理数据异质性,类不平衡和特征相互依赖.
  • 整合临床传感器数据可以提高IGAN亚型的预测,从而改善诊断和预后.