Long-PVSUNet: A Longitudinal Deep Learning Framework for Count-Oriented Perivascular Space Segmentation in Brain MRI.
Authors
Affiliations (6)
Affiliations (6)
- Centre for Healthy Brain Ageing (CHeBA), Discipline of Psychiatry and Mental Health, School of Clinical Medicine, Faculty of Medicine and Health, UNSW Sydney, Randwick, NSW, Australia. [email protected].
- Centre for Healthy Brain Ageing (CHeBA), Discipline of Psychiatry and Mental Health, School of Clinical Medicine, Faculty of Medicine and Health, UNSW Sydney, Randwick, NSW, Australia.
- Research Imaging NSW, Prince of Wales Hospital, UNSW Sydney, Randwick, NSW, Australia.
- School of Computer Science and Engineering, UNSW Sydney, Randwick, NSW, Australia.
- Data61, CSIRO, Sydney, NSW, Australia.
- Euroa Centre, Neuropsychiatric Institute (NPI), Prince of Wales Hospital, Randwick, NSW, Australia.
Abstract
Enlarged perivascular spaces (ePVS) are small, sparse MRI-visible markers of cerebral small vessel disease and brain ageing. Their size, low contrast, and severe foreground-background imbalance make automated segmentation challenging. Existing methods are mainly cross-sectional and do not model temporal consistency, limiting their utility for tracking longitudinal change. We developed Long-PVSUNet, a two-timepoint framework for count-oriented ePVS segmentation on paired baseline/follow-up T1-weighted and FLAIR MRI using baseline-guided attention fusion and imbalance-aware training. After screening/QC, longitudinal data were available from UK Biobank (UKB; n = 4568), ADNI (n = 277), and MAS (n = 403). Expert dot annotations were obtained in labelled UKB, ADNI, and MAS subsets (n = 250, 100, 100) for basal ganglia (BG) and centrum semiovale, regions commonly used for clinically relevant ePVS counting. Performance and generalisability were evaluated against cross-sectional and longitudinal baselines using Dice, lesion-level F1-localisation, and cross-cohort, few-shot, ablation, and data-efficiency analyses. Exploratory analyses tested hypertension associations with model-derived ePVS counts and progression in UKB/MAS. Long-PVSUNet achieved strong UKB performance (Dice 0.802 ± 0.010, F1 0.842 ± 0.011) and generalised to ADNI/MAS. It remained data-efficient with fewer labels. Attention fusion significantly outperformed mean and temporal-difference fusion (Dice 0.800 ± 0.004). Long-PVSUNet outperformed cross-sectional U-Net and longitudinal benchmark model, suggesting its utility for longitudinal ePVS segmentation. In clinical analysis, hypertension was associated with faster BG-ePVS progression in UKB (incidence rate ratio (IRR) = 1.06, 95% CI: 1.02-1.10) and combined UKB + MAS (IRR = 1.05, 95% CI: 1.01-1.09). Long-PVSUNet enables scalable, temporally stable ePVS segmentation and count-based quantification, with exploratory evidence of plausible vascular associations.