Longitudinal multi-modal data prediction model for mild cognitive impairment by deep survival analysis

基于深度生存分析的轻度认知障碍纵向多模态数据预测模型

阅读:1

Abstract

BACKGROUND: Timely prediction of cognitive decline in patients with Mild Cognitive Impairment (MCI) is crucial for guiding optimal therapeutic interventions. In this study, we aimed to develop a deep survival analysis model that leverages longitudinal, multi-modal data to estimate the probability of dementia conversion, thereby facilitating personalized treatment planning in clinical practice. METHODS: We employed a deep neural network model specifically designed for survival analysis to predict the progression from MCI to Alzheimer’s Disease (AD). The model integrated longitudinal biomarkers, including neuropsychological assessments and neuroimaging measures, along with baseline demographic characteristics and genetic risk factors, using data from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database. RESULTS: This study enrolled 922 baseline MCI patients for analysis. The predictive performance was evaluated using a test set at time intervals [Formula: see text] = 1, 2, 3, 4 years from the landmark time s = 1. The prognostic model exhibited outstanding predictive capability, attaining cdAUC values of 0.9089 ± 0.01 alongside BS of 0.1651 ± 0.01 with [Formula: see text] = 1 year on the test set, when all variable sets were incorporated into the time-dependent Cox survival neural network (tdCoxSNN) model. Through feature significance evaluation, the Functional Activities Questionnaire (FAQ) emerged as the most influential predictive element. CONCLUSIONS: By systematically integrating diverse longitudinal biomarkers, we developed a dynamic prediction model for MCI using deep survival analysis. This approach enables accurate individual risk stratification, facilitates the early identification of high-risk individuals, and supports informed, personalized clinical decision-making. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1186/s12911-026-03387-3.

特别声明

1、本页面内容包含部分的内容是基于公开信息的合理引用;引用内容仅为补充信息,不代表本站立场。

2、若认为本页面引用内容涉及侵权,请及时与本站联系,我们将第一时间处理。

3、其他媒体/个人如需使用本页面原创内容,需注明“来源:[生知库]”并获得授权;使用引用内容的,需自行联系原作者获得许可。

4、投稿及合作请联系:info@biocloudy.com。