Medical News Observer

Key Points:

• DETree provides precise predictions of Alzheimer’s disease stages, improving previous prediction models.
• Data from 266 Alzheimer’s patients demonstrated the framework’s effectiveness, with superior performance.
• Its potential extends beyond Alzheimer’s, with possible applications in predicting the stages of diseases like Parkinson’s, Huntington’s, and Creutzfeldt-Jakob.

Overview

According to the World Health Organization, approximately 55 million people globally are affected by dementia, with Alzheimer’s disease being the most common form. This incurable condition progressively impairs brain function, causing not only physical deterioration but also psychological, social, and economic impacts on both the patients and their caregivers. 

Current Alzheimer’s disease prediction models

As Alzheimer’s symptoms intensify over time, it becomes crucial for patients and their caregivers to anticipate the increasing need for support as the disease progresses. Traditional predictive models for Alzheimer’s dementia are often focused on binary or multi-class classification. They do not consider the continuous nature of AD development and its transition states. 

DETree: Disease-embedding tree

To address the limitations, Researchers at The University of Texas at Arlington have developed a groundbreaking learning-based framework to assist Alzheimer’s patients in accurately identifying their current stage within the disease development spectrum. The framework termed the “disease-embedding tree” (DETree), enables precise prediction of the disease’s progression, including the timing of its later stages. 

Results

This novel approach outperforms previous prediction models. Notably, the DETree has been tested using data from 266 individuals with Alzheimer’s disease from the Alzheimer’s Disease Neuroimaging Initiative. 

Implications

Researchers believe that this framework could also be beneficial in predicting the progression of other diseases with multiple clinical stages, such as Parkinson’s disease, Huntington’s disease, and Creutzfeldt-Jakob disease.

References

• Zhang, Lu, Li Wang, Tianming Liu, and Dajiang Zhu. 2024. “Disease2Vec: Encoding Alzheimer’s Progression via Disease Embedding Tree.” Pharmacological Research 199 (January): 107038. https://doi.org/10.1016/j.phrs.2023.107038.
• Zhang, Lu. n.d. “Disease2Vec.” Accessed January 29, 2024. github.com/qidianzl/Disease2Vec.
• “Dementia.” n.d. Accessed January 29, 2024. https://www.who.int/health-topics/dementia.