Multimodal brain age prediction using machine learning: combining structural MRI and 5-HT2AR PET derived features
To better assess the pathology of neurodegenerative disorders and to evaluate the efficacy of neuroprotective interventions, it is required to develop biomarkers that can accurately capture age-related biological changes in the human brain. Brain serotonin 2A receptors (5-HT2AR) show a particularly profound age-related decline and are also widely reduced in, e.g., Alzheimer's disease. Hence, cerebral 5-HT2AR binding measured in vivo using positron emission tomography (PET) is a potentially useful biomarker for age-related changes in the brain. In this study, we investigate the decline in 5-HT2AR binding to evaluate its usefulness as a biomarker for biological aging. Specifically, we aim to 1) predict brain age using 5-HT2AR binding outcomes, 2) compare 5-HT2AR-based predictions of brain age to predictions based on gray matter (GM) volume, as determined with structural magnetic resonance imaging (MRI), and 3) investigate whether combining 5-HT2AR and GM volume data improves prediction. We used PET and MR images from 209 healthy individuals aged between 18 and 85 years (mean=38, std=18). 5-HT2AR binding and GM volume were calculated for 14 cortical and subcortical regions. Different machine learning algorithms were used to predict age based on 5-HT2AR binding, GM volume, and the combined measures. The mean absolute error (MAE) and a cross-validation approach were used for evaluation and model comparison. We find that both the cerebral 5-HT2AR binding (mean MAE=6.63 years, std=0.78 years) and GM volume (mean MAE=7.76 years, std=0.92 years) predict chronological age accurately. Combining the two measures improves the prediction further (mean MAE=5.93 years, std=0.82). We conclude that when it comes to predicting age, in vivo measurements of the cerebral 5-HT2AR binding are more informative than GM volumes.