MetaExplainer:从可解释性的角度重新审视功能连接组分析的域概括
Xinmei Qiu1, Yongheng Sun1, Yilin Shi2
1School of Mathematics and Statistics, Xi'an Jiaotong University, Xi'an 710049, China; Research Center for Intelligent Medical Equipment and Devices (IMED), Xi'an Jiaotong University, Xi'an 710049, China.
Medical image analysis
|June 25, 2025
概括
本研究介绍了MetaExplainer,这是一种新的框架,可以增强图形神经网络 (GNN) 的解释性和概括性,用于使用fMRI数据诊断神经精神疾病. 它通过学习域异性生物标志物来实现跨中心的诊断准确度,从而获得最先进的结果.
科学领域:
- 神经科学是一个神经科学.
- 机器学习 机器学习
- 医疗成像医学成像
背景情况:
- 图形神经网络 (GNN) 对于在神经精神疾病诊断中从fMRI数据中分析功能连接组至关重要.
- GNNs的可解释性和通用性对于可靠性至关重要,但通常是独立地解决的.
- 可解释性和可概括性之间的内在联系对于检测微妙的,早期的疾病变化至关重要.
研究的目的:
- 弥合基于fMRI的神经精神疾病诊断的GNN中可解释性和可概括性之间的差距.
- 开发一个元学习框架 (MetaExplainer),以提高跨中心诊断的准确性和可解释性.
- 为了改善诊断性能,识别域异性,与疾病相关的连接组变化.
主要方法:
- 提出了MetaExplainer,这是一个具有解释性概括性规范化的元学习框架.
- 采用双循环元学习过程,从fMRI BOLD信号中学习面向任务的非线性功能网络.
- 专注于在不同临床中心捕捉稳定,域无定,与疾病相关的连接组变化.
主要成果:
- 在两个多中心数据集 (ABIDE,REST-meta-MDD) 上实现了最先进的性能.
- 证明了高诊断准确度:自闭症谱系障碍 (ASD) 的AUC为76.32%,主要抑郁症 (MDD) 的AUC为65.31%.
- 确定了神经生物学上可信的生物标志物,例如MDD中的脑下网络 (SMN) 功能障碍,得到了现有研究的支持.
结论:
- MetaExplainer有效地整合了可解释性和可概括性,用于基于fMRI的稳健GNN诊断.
- 该框架通过学习稳定,可解释的生物标志物来提高跨中心诊断的准确性.
- 这项研究强调了可解释的AI在促进神经精神疾病的精准医学方面的潜力.
更多相关视频
相关概念视频
Functional Divisions of the Nervous System
6.4K
The nervous system, responsible for sensing, integrating, and responding to various stimuli, is divided into the central nervous system (CNS) and the peripheral nervous system (PNS). The PNS has two functional divisions: the sensory or afferent division and the motor or efferent division.
The sensory division transmits information from sensory receptors in the body to the CNS. It provides the CNS with knowledge about somatic senses (such as tactile, thermal, pain, and proprioceptive sensations)...
The sensory division transmits information from sensory receptors in the body to the CNS. It provides the CNS with knowledge about somatic senses (such as tactile, thermal, pain, and proprioceptive sensations)...
6.4K
Organization of the Brain
1.2K
The brain is an integral component of the nervous system and serves as the center for processing sensory inputs, making decisions, and directing bodily actions. This complex organ is organized into three primary sections: the hindbrain, midbrain, and forebrain, each responsible for a range of vital functions.
Hindbrain
The hindbrain, located at the base of the brain, plays a vital role in regulating automatic processes that sustain life. It includes the medulla oblongata, which is essential for...
Hindbrain
The hindbrain, located at the base of the brain, plays a vital role in regulating automatic processes that sustain life. It includes the medulla oblongata, which is essential for...
1.2K
Neural Circuits
1.6K
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...
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...
1.6K
Multicompartment Models: Overview
261
Multicompartment models are mathematical constructs that depict how drugs are distributed and eliminated within the body. They segment the body into several compartments, symbolizing various physiological or anatomical areas connected through drug transfer processes such as absorption, metabolism, distribution, and elimination.
These models offer a more comprehensive representation of drug behavior in the body than one-compartment models. They accommodate the complexity of drug distribution,...
These models offer a more comprehensive representation of drug behavior in the body than one-compartment models. They accommodate the complexity of drug distribution,...
261
Functional Brain Systems: Reticular Formation
2.7K
The reticular formation is a complex network of gray and white matter located within the brainstem extending from the medulla to the midbrain.
Within the reticular formation, there are several distinct nuclei that can be classified into three broad categories. The Raphe nuclei are located along the midline of the brainstem. They are primarily known for their role in synthesizing and releasing serotonin, a neurotransmitter involved in regulating mood, appetite, sleep, and circadian rhythms. The...
Within the reticular formation, there are several distinct nuclei that can be classified into three broad categories. The Raphe nuclei are located along the midline of the brainstem. They are primarily known for their role in synthesizing and releasing serotonin, a neurotransmitter involved in regulating mood, appetite, sleep, and circadian rhythms. The...
2.7K
Mechanistic Models: Overview of Compartment Models
172
Mechanistic models, a category encompassing both physiological and compartmental modeling, differ from empirical models' approaches to incorporating known factors about the systems being modeled. Empirical models describe data with minimal assumptions, while mechanistic models aim to provide a robust description of available data by specifying assumptions and integrating known factors about the system. Compartmental analysis is a key example of a mechanistic model in pharmacokinetics and...
172


