Disease progression modeling of Alzheimer’s disease based on variational probability principal component analysis
Xin Xiong, Ximin Wang, Chenyang Zhu, Jianfeng He
Abstract
Alzheimer’s disease (AD) is a neurodegenerative disorder and the leading cause of dementia. Early diagnosis and monitoring of disease progression are crucial for effective intervention. This study presents a novel disease progression model based on Variational Probabilistic Principal Component Analysis (VPPCA), which uses a Bayesian framework for dimensionality reduction and uncertainty quantification.
Introduction
Alzheimer’s disease (AD) is a neurodegenerative disease and the leading cause of dementia. The number of patients is expected to exceed 7 million by 2030 [1]; by 2050, the global prevalence will triple and is expected to affect more than 100 million people [2]. Early diagnosis of patients can be achieved by understanding the underlying pathological mechanisms, identifying multiple pathogenic and protective genes.
Materials and methods
The data used in this article are from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database. Our study collected 16,422 records from 2,431 patients, covering data from four stages: ADNI-1, ADNI-2, ADNI-GO, and the latest ADNI-3.
Results
The original data is projected onto the latent space after VPPCA. Fig 1 shows the comparison between the first principal component score and the second principal component score. It can be seen that the data distribution range and interquartile range of VPPCA1 are much larger than those of VPPCA2.
Discussion
This paper proposes a novel approach, namely variational probability principal component analysis (VPPCA), to predict and analyze the disease progression of Alzheimer’s disease (AD). Through the analysis of the ADNI dataset, the experiments demonstrate the effectiveness and robustness of the method in processing high-dimensional biomarker data, coping with missing values, and extracting disease progression features.
Conclusion
The variational probability principal component analysis (VPPCA) method proposed in this paper performs well in dealing with high-dimensional feature data and missing value problems in Alzheimer’s disease (AD).
Citation: Xiong X, Wang X, Zhu C, He J, for the Alzheimer’s Disease Neuroimaging Initiative (2026) Disease progression modeling of Alzheimer’s disease based on variational probability principal component analysis. PLoS One 21(3): e0342549. https://doi.org/10.1371/journal.pone.0342549
Editor: Yong Fan, University of Pennsylvania Perelman School of Medicine, UNITED STATES OF AMERICA
Received: August 4, 2025; Accepted: January 26, 2026; Published: March 30, 2026
Copyright: © 2026 Xiong et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All data are from ADNI database (https://adni.loni.usc.edu/data-samples/adni-data/). If others want to access the data set, they can register an account through this website and submit an application to ADNI. There will usually be a reply within one or two days, and ADNI will allow downloading data for research purposes. We are absolutely sure that anyone can access the data set in the same way. We also confirm that there are no special access rights that others do not have.
Funding: This study was financially supported by the National Natural Science Foundation of China in the form of a grant awarded to XX and JH (82060329). This study was also financially supported by the Yunnan Fundamental Research Projects program in the form of a grant awarded to XX and JH (202201at070108). This study received additional financial support from the First People’s Hospital of Yunnan Province in the form of a grant awarded to XX and JH (2024nmkfkt-09). Further financial support was provided by the Strategic Project of Fuwai Cardiovascular Hospital in the form of a grant awarded to XX and JH (2025yfkt-zl-08). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
