Motivation: Clinical diagnosis of frontotemporal dementia (FTD) can be challenging due to its distinct regional atrophy patterns on MRI. Goal(s): This study aimed to evaluate the effectiveness of a deep learning-based automated brain volumetry tool for the diagnosis of FTD using MRI. Approach: The study included 759 subjects, with brain volumetry performed using VUNO Med DeepBrain software. Key volumetric features, including the frontal lobes, insula, cingulate cortex, and subcortical gray matter, were identified. Results: linear SVM classifier using volumetric features, age and MMSE scores, achieved 89.3% accuracy in training, 92.0% in internal validation, and 84.6% in external validation, demonstrating strong diagnostic performance. Impact: Our automated brain volumetry model demonstrated promising diagnostic accuracy for FTD, offering a potential tool for differentiating FTD from AD and normal in clinical settings.
Lee et al. (Tue,) studied this question.