| Title:
|
Olfactory Brain Volume and Functional Connectivity as Predictors of Age of Onset in Alzheimer’s Disease: A Machine Learning Approach
|
| Author(s):
|
1
|
| Subject:
|
Artificial Intelligence and Brain Mapping
|
|
|
|
Presenter Author:
|
|
| Name:
|
Reza Sheikhzadeh
|
|
|
|
Author:
|
|
| Name 1:
|
Reza Sheikhzadeh - Presenter
|
| Affiliation 1:
|
Department of medical physics, School of Medicine, Iran University of Medical Sciences, Tehran, Iran
|
|
|
| Name 2:
|
Hamid Aghlmand
|
| Affiliation 2:
|
Department of medical physics, School of Medicine, Iran University of Medical Sciences, Tehran, Iran
|
|
|
| Name 3:
|
Arash Zare-Sadeghi - Corresponding
|
| Affiliation 3:
|
Department of medical physics, School of Medicine, Iran University of Medical Sciences, Tehran, Iran
|
|
|
| Name 4:
|
|
| Affiliation 4:
|
|
|
|
| Name 5:
|
|
| Affiliation 5:
|
|
|
|
| Name 6:
|
|
| Affiliation 6:
|
|
|
|
| Name 7:
|
|
| Affiliation 7:
|
|
|
|
| Name 8:
|
|
| Affiliation 8:
|
|
|
|
| Name 9:
|
|
| Affiliation 9:
|
|
|
|
| Name 10:
|
|
| Affiliation 10:
|
|
|
|
Abstract
|
Original Research
|
|
introduction :
|
Early diagnosis and prediction of Alzheimer’s disease (AD) onset are critical for timely intervention and care planning. This study aims to predict the age of symptom onset in individuals with AD using machine learning (ML) models trained on multimodal MRI data.
|
|
Methods:
|
We extracted both functional MRI (rs-fMRI) and structural MRI features from regions involved in olfactory processing, including the orbitofrontal cortex (OFC), hippocampus, hypothalamus, and entorhinal cortex. The dataset was sourced from the ADNI database, and preprocessing involved standard pipelines for brain region segmentation and feature extraction. Several regression models, including K-Nearest Neighbors (KNN), Support Vector Regression (SVR), Gradient Boosting Machines (GBM), and Random Forests (RF), were optimized using Optuna hyperparameter tuning.
|
|
Results:
|
Our results demonstrated that structural features, particularly cortical thickness in the orbitofrontal and hippocampal regions, yielded the best performance (R² up to 0.63; RMSE as low as 1.85 years). Functional connectivity features, in contrast, exhibited higher variability and lower predictive power. SHAP value analysis confirmed the dominance of anatomical features in
prediction performance.
|
|
Conclusion:
|
These findings highlight the potential of structural MRI markers combined with machine learning for age-related prediction in AD.
|
| Attachment:
|
|
| Keywords:
|
Alzheimer's disease, machine learning, MRI, age prediction, olfactory
|
| Key message:
|
|
| Do you want your abstract to be indexed, if this is selected as one of the top 100
articles?
|
