相关实验视频

Updated: May 5, 2026

A Novel Dual-Modal Deep Learning Approach for Real-Time Removal of Hepatic Fluorescence in Indocyanine Green-Guided Laparoscopic Cholecystectomy

A Novel Dual-Modal Deep Learning Approach for Real-Time Removal of Hepatic Fluorescence in Indocyanine Green-Guided Laparoscopic Cholecystectomy

Published on: April 17, 2026

神经形态融合:一种神经启发的混合学习框架,用于在物联网支持的医疗保健系统中进行可解释的深层损伤检测

Roseline Oluwaseun Ogundokun^1,2, Rotimi-Williams Bello¹, Pius Adewale Owolawi¹

¹Department of Computer Systems Engineering, Faculty of Information and Communication Technology, Tshwane University of Technology, Pretoria, South Africa.

Technology in cancer research & treatment

|March 12, 2026

概括

NeuroMorphFusion通过将深度学习与神经启发模型相结合来增强医疗成像,以在边缘设备上有效和可解释的病变检测. 这种框架实现了高精度,同时保持了低延迟,以实现现实世界的临床整合.

科学领域:

人工智能的人工智能
医疗成像医学成像
计算神经科学是一种神经科学.

背景情况:

医疗物联网 (IoMT) 的深度学习提高了自动化病变检测.
挑战包括在资源有限的边缘设备上保持诊断准确性,可解释性和效率.

研究的目的:

引入NeuroMorphFusion,这是一个用于可解释和高效的病变检测的混合框架.
为解决边缘计算的 IoMT 中当前深度学习模型的局限性.

主要方法:

神经形态融合集成了ResNet18骨干,尖端神经网络 (SNN) 和形态注意力机制.
一个半监督的强化学习策略和一个遗传算法 (GA) 优化模型的准确性,可解释性和效率.
在Jetson Nano上进行了超参数优化,在8分钟内收.

主要成果:

在IQ-OTHNCCD肺癌CT数据集上,NeuroMorphFusion实现了98.18%的分类准确度.
与VGG16,SqueezeNet,MobileNetV3和ResNet18相比,该框架显示出更高的透明度和效率.
多目标优化平衡了灵敏度,延迟和可解释性,并通过SHAP和Grad-CAM可视化进行了增强.

结论:

神经形态融合有效地结合了神经生物学灵感,数学可解释性和对IoMT的边缘高效计算.
该框架显示了对现实世界的临床集成和可扩展的IoMT部署的潜力,因为它的准确性,可解释性和低延迟优化.

关键词:

在CT成像中使用CT成像.神经形态融合的神经形态融合在 grad-CAM 中使用.医疗事物互联网的互联网.损伤检测检测损伤的检测形态注意力形态注意力尖的神经网络的神经网络.

相关概念视频

Neural Circuits

Neural Circuits

Neural circuits and neuronal pools are two of the main structures found in the nervous system. Neural circuits are networks of neurons that work together to carry out a specific task or process. They consist of interconnected neurons and glial cells, which provide structural and metabolic support.
Neuronal pools are collections of nerve cells with similar functions and interact through chemical and electrical signals. These pools include both interneurons (the central neural circuit nodes that...