SWI and T2*-GRE Microhemorrhage Counts in Anti-Amyloid Therapy Eligibility: A Real-World-Calibrated Simulation Study
Sato, K.; Ihara, R.; Kurihara, M.; Suzuki, F.; Tokumaru, A.; Niimi, Y.; Iwatsubo, T.
Show abstract
INTRODUCTION: Anti-amyloid therapy eligibility excludes patients with five or more cerebral microhemorrhages (CMHs), but current guidance allows either T2*-GRE or the more sensitive SWI. This may create sequence-dependent differences in eligibility classification. METHODS: We fitted a Bayesian right-censored zero-inflated Poisson model to single-center real-world SWI-based CMH counts from 130 memory clinic patients. We then simulated T2*-GRE counts under a directional binomial detection model across a range of relative detection probabilities and estimated two metrics: P(T2* <5 | SWI [≥]5) and P(SWI [≥]5 | T2* <5). RESULTS: The model estimated a zero-inflation probability of 0.57 and a Poisson mean of 2.29 in the susceptible subpopulation. In the illustrative 60% relative-detection scenario, the model estimated nearly 80% under-detection among SWI-ineligible patients, whereas approximately 5% of T2*-GRE-eligible patients would have been classified as ineligible by SWI. DISCUSSION: A uniform numeric CMH threshold may not be operationally equivalent across SWI and T2*-GRE. Standardized paired-sequence outcome studies are needed to determine whether SWI-defined and T2*-GRE-defined thresholds carry equivalent ARIA risk.
Matching journals
The top 9 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- Fusion of quantitative susceptibility maps and T1-weighted images improve braintissue contrast in primates 92%
- Advancing motion denoising of multiband resting-state functional connectivity fMRI data 92%
- Automated joint skull-stripping and segmentation with Multi-Task U-Net in large mouse brain MRI databases 92%
Similar papers in this journal
- QSMRim-Net: Imbalance-Aware Learning for Identification of Chronic Active Multiple Sclerosis Lesions on Quantitative Susceptibility Maps 93%
- Fluid and White Matter Suppression Contrasts MRI Improves Deep Learning Detection of Multiple Sclerosis Cortical Lesions 93%
- ComBat Harmonization: Empirical Bayes versus Fully Bayes Approaches 93%
Similar papers in this journal
- Evaluation of ComBat harmonization for reducing across-tracer biases in regional amyloid PET analyses 95%
- WMH-DualTasker: A weakly-supervised deep learning model for automated white matter hyperintensities segmentation and visual rating prediction 95%
- OpenMAP-T1: A Rapid Deep Learning Approach to Parcellate 280 Anatomical Regions to Cover the Whole Brain 93%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.