摘要
本研究使用一种新型的级联金字塔宽度学习系统(Cascade Pyramid Broad Learning System,CPBLS)以提升阿尔茨海默病(Alzheimer’s Disease,AD)的早期诊断效率。该系统整合了k-means聚类模块,用于优化特征提取和降维过程。通过功能性磁共振成像(Magnetic Resonance Imaging,MRI)数据,CPBLS模型在分类任务中展现了卓越的性能,包括准确率、精确率、召回率、F_(1)得分以及AUC值。宽度学习技术的应用,结合k-means模块,显著降低了模型的复杂度,并缩短了训练时间,为AD的早期诊断提供了快速且有效的解决方案。
This study employs a novel cascade pyramid broad learning system(CPBLS)to enhance the early diagnosis efficiency of Alzheimer’s disease(AD).The system integrates a k-means clustering module to optimize feature extraction and dimensionality reduction.Using functional magnetic resonance imaging(MRI)data,the CPBLS model demonstrates outstanding performance in classification tasks,including high accuracy,precision,recall,F_(1)-score,and AUC values.The application of broad learning technology,combined with the k-means module,significantly reduces model complexity and shortens training time,providing a fast and effective solution for the early diagnosis of AD.
作者
江洁
郭金兴
才莹
富丹
JIANG Jie;GUO Jin-xing;CAI Ying;FU Dan(Department of Infectious Diseases,Mudanjiang Forestry Hospital,Mudanjiang,Heilongjiang 157011,China;Department of Urology,Hongqi Hospital Affiliated to Mudanjiang Medical University,Mudanjiang,Heilongjiang 157011,China;Department of Critical Care Medicine,Hongqi Hospital Affiliated to Mudanjiang Medical University,Mudanjiang,Heilongjiang 157011,China)
出处
《新一代信息技术》
2025年第3期1-5,共5页
New Generation of Information Technology
基金
牡丹江市应用技术研究与开发计划项目(No.HT2022NS116)
