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

Search research articles

相关实验视频

Updated: Jan 12, 2026

Quantifying the Brain Metastatic Tumor Micro-Environment using an Organ-On-A Chip 3D Model, Machine Learning, and Confocal Tomography

Quantifying the Brain Metastatic Tumor Micro-Environment using an Organ-On-A Chip 3D Model, Machine Learning, and Confocal Tomography

Published on: August 16, 2020

基于神经网络的可解释的修改卷积混合器深度学习框架,用于准确的脑瘤检测和分类.

S Selva Birunda¹, M Kaliappan¹, R Ramana¹

¹Department of Artificial Intelligence and Data Science, Ramco Institute of Technology, Rajapalayam, India.

Neurological research

|October 31, 2025

概括

本研究介绍了一种可解释的修改卷积混合器神经网络 (EM-ConvMixer+Net),用于使用MRI图像进行高效和准确的脑瘤检测和分类. 这种新型模型实现了高精度,为临床诊断提供了可靠的工具.

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

An Intelligent IoT-Compatible Arrhythmia Detection System Using a Hybrid Vision Transformer-LSTM Framework.

Cardiovascular engineering and technology·2026

Same author

Banana Plant Disease Classification Using Hybrid Convolutional Neural Network.

Computational intelligence and neuroscience·2022

Same author

Cervical Cancer Identification with Synthetic Minority Oversampling Technique and PCA Analysis using Random Forest Classifier.

Journal of medical systems·2019

Same author

Development of Big Data Predictive Analytics Model for Disease Prediction using Machine learning Technique.

Journal of medical systems·2019

Same author

Secure and fair cluster head selection protocol for enhancing security in mobile ad hoc networks.

TheScientificWorldJournal·2014

Same author

Aftercare of depressed inpatients--service delivery and unmet needs.

Social psychiatry and psychiatric epidemiology·2003

科学领域:

医疗成像医学成像
人工智能的人工智能
计算神经科学是一种神经科学.

背景情况:

大脑瘤 (BTs) 由于其复杂性和相关的低存活率,对健康构成重大挑战.
传统的诊断模型通常是计算密集型,需要长时间的培训,阻碍了及时的临床应用.
需要先进,高效和可解释的AI模型来分析脑瘤.

研究的目的:

开发和验证一种用于脑瘤检测和分类的新型可解释修改卷积混合器神经网络 (EM-ConvMixer+Net).
提高神经瘤学中AI模型的计算效率和可解释性.
从MRI数据提高自动化脑瘤诊断的准确性和可靠性.

主要方法:

利用了来自两个公共来源的10,087张MRI图像的数据集,采用了像无声化和增强这样的预处理技术.
通过采用区域生长方法实现瘤细分,然后使用卷积运算混合块 (ConvMixer+) 进行特征增强.
使用交叉注意网络 (CANet) 和外部注意网络 (EANet) 进行特征融合和改进,通过修改的人工神经网络 (M-ANN) 进行分类,并通过Grad-CAM进行解释.

主要成果:

在实验验证中,EM-ConvMixer+Net模型获得了98.86%的高精度和98.44%的F1得分.
对比和废除研究证实了该模型的有效性和稳定性.

关键词:

脑瘤检测脑瘤检测脑瘤检测交叉注意力网络交叉注意力网络外部关注网络外部关注网络梯度加权类激活映射映射的梯度加权类激活映射修改的人工神经网络细分和地区增长的细分和地区增长.

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

相关实验视频

Last Updated: Jan 12, 2026

Quantifying the Brain Metastatic Tumor Micro-Environment using an Organ-On-A Chip 3D Model, Machine Learning, and Confocal Tomography

Quantifying the Brain Metastatic Tumor Micro-Environment using an Organ-On-A Chip 3D Model, Machine Learning, and Confocal Tomography

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

梯度加权类激活映射 (Grad-CAM) 提供了视觉解释,提高了模型的解释性.

结论:

拟议的EM-ConvMixer+Net框架为脑瘤诊断提供了一个计算高效和可解释的解决方案.
这种技术显示出在神经瘤学中推进临床决策的巨大潜力.
该模型的高性能和可解释性为更可靠的AI辅助医学诊断铺平了道路.