Machine Learning-Based Age Prediction with Feature Subset Selection from Magnetic Resonance Angiography Data

Abstract

Objectives: The objective of this study was to evaluate the effectiveness of machine learning (ML) models using selectedsubsets of features to predict age based on intracranial arterial segments’ tortuosity and diameter characteristics derived frommagnetic resonance angiography (MRA) data. Additionally, this study aimed to identify key vascular features important forpredicting vascular age. Methods: Three-dimensional time-of-flight MRA image data from 171 subjects were analyzed. Afterannotating the endpoints for each arterial segment, 169 features—comprising tortuosity metrics and arterial segment diameterstatistics—were extracted. Five ML models (random forest, linear regression, AdaBoost, XGBoost, and lightGBM) weretrained and validated. Two feature selection methods, correlation-based feature selection (CFS) and Relief-F, were appliedto identify optimal feature subsets. Results: The random forest model utilizing the CFS-based 50% feature subset achievedthe best performance, with a root mean square error of 14.0 years, a coefficient of determination (R2) of 0.275, and a Pearsoncorrelation coefficient of 0.560. Tortuosity metrics (e.g., triangular index of the left posterior cerebral artery P1 segment)appeared more frequently than diameter statistics among the top five most important features. Conclusions: CFS-based featureselection enhanced the performance of ML-based age prediction compared with using the complete feature set. Linearregression consistently demonstrated the poorest performance across all evaluation metrics. ML-based age prediction usingsegmental tortuosity metrics and diameter statistics is feasible, potentially revealing significant features related to vascular aging.