Explainable Tensor Multi-Task Ensemble Learning Based on Brain Structure Variation for Alzheimer's Disease Dynamic...

Area of Science:

• Computational neuroscience
• Artificial intelligence in medicine
• Biomedical data analysis

Background:

• Accurate prediction of Alzheimer's disease (AD) progression is crucial for developing effective interventions.
• Existing machine learning models require enhancement for improved AD progression prediction accuracy and stability.

Purpose of the Study:

• To propose a novel machine learning algorithm for predicting Alzheimer's disease progression.
• To leverage a multi-task ensemble learning approach incorporating tensor decomposition and gradient boosting.

Main Methods:

• Developed a tensor multi-task learning (MTL) algorithm analyzing spatio-temporal brain biomarker variability.
• Utilized tensor decomposition to identify shared latent factors across patient prediction tasks.
• Integrated gradient boosting for ensemble learning on temporally continuous subject data.

Main Results:

• The proposed model demonstrated superior accuracy and stability in predicting AD progression compared to existing methods.
• Performance was evaluated using Mini Mental State Examination (MMSE) and Alzheimer's Disease Assessment Scale-Cognitive Subscale (ADAS-Cog) scores.
• The model effectively utilizes magnetic resonance imaging (MRI) data and cognitive scores for prediction.

Conclusions:

• The novel tensor multi-task ensemble learning algorithm offers a powerful tool for predicting AD progression.
• This approach can identify individual brain structure variations and improve clinical management strategies for AD.