Comparison of Ensemble Techniques for Early Prediction of Alzheimer Disease
Author Information
Author(s): Orlunwo Placida Orochi, Onuodu Friday Eleonu
Primary Institution: Ignatius Ajuru University of Education
Hypothesis
To predict the transition from mild cognitive impairment to probable Alzheimer's disease using ensemble learning approaches.
Conclusion
The study found that ensemble methods like Stacking, Adaptive Boost, and Bagging can accurately identify individuals at risk of developing Alzheimer's disease.
Supporting Evidence
- Ensemble methods improved prediction accuracy for Alzheimer's disease compared to traditional methods.
- Adaptive Boost, Stacking, and Bagging achieved an accuracy of 97% in classifying Alzheimer's stages.
- Machine learning techniques can identify individuals at risk of developing Alzheimer's disease before symptoms appear.
Takeaway
This study used smart computer programs to help find out who might get Alzheimer's disease before they show any signs.
Methodology
The study applied five ensemble learning techniques on neuroimaging and cognitive performance data to predict Alzheimer's disease.
Potential Biases
Potential for model overfitting due to data imbalance and the exclusion of certain biomarkers.
Limitations
The study was limited to 14 biomarkers and excluded others due to high null values, which may affect the model's performance.
Participant Demographics
Participants ranged in age from 55 to 96, including both males and females.
Statistical Information
P-Value
p<0.05
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website