基于深度神经网络的阿尔兹海默病早期诊断算法

Alzheimer's Disease Early Diagnosis Based on Deep Neural Network

关键词:  阿尔兹海默病,神经退行性疾病,疾病诊断,深度学习,对比学习,多层感知机

Key words:  Alzheimer"s disease  neurodegenerative disease  disease diagnosis  deep learning  contrast learning  multilayer perceptron

摘要:

由于阿尔兹海默病目前无法治愈,只能通过早期干预的方式缓解其恶化,因此使用神经影像学技术对阿尔兹海默病进行早期诊断已经在临床中得到了广泛的应用。本文使用深度学习技术对根据磁共振成像数据进行阿尔兹海默症的早期诊断进行了研究。首先,本文引入样本重加权策略来缓解疾病诊断数据集中普遍存在的类不平衡问题对模型的影响。其次,本文设计了监督对比学习来提升模型的特征提取能力。最后,在四个神经退行性疾病诊断数据集上的实验结果表明,本文提出的方法取得了比现有算法更好的性能。

Abstract:

Since Alzheimer"s disease is currently incurable and its deterioration can only be mitigated by early intervention. Therefore, early diagnosis of Alzheimer"s disease using neuroimaging techniques has been widely used in clinical practice. This paper investigates the early diagnosis of Alzheimer"s disease using magnetic resonance imaging data using deep network models. First, this paper introduces a sample reweighting strategy to mitigate the impact of the class imbalance problem prevalent in disease diagnosis datasets on the model. Second, this paper designs supervised comparison learning to improve the feature extraction capability of the model. Finally, experimental results on four neurodegenerative disease diagnosis datasets show that the method proposed in this paper achieves better performance than existing methods.