Related Experiment Video
Updated: Jun 30, 2026
Probe Type II Band Alignment in One-Dimensional Van Der Waals Heterostructures Using First-Principles Calculations
Published on: October 12, 2019
Feynman-Kac Reweighted Schrödinger Bridge Matching for Surface-based Tau PET Harmonization.
Jianwei Zhang1, Xinyu Nie1, Jiaxin Yue1
1Jianwei Zhang, Xinyu Nie, Jiaxin Yue, and Yonggang Shi are with the Stevens Neuroimaging and Informatics Institute, University of Southern California, Los Angeles, CA 90089 USA. Jianwei Zhang, Jiaxin Yue, and Yonggang Shi are with the Ming Hsieh Department of Electrical and Computer Engineering of Viterbi School of Engineering, University of Southern California, Los Angeles, CA 90089 USA. Yonggang Shi is also with the Alfred E. Mann Department of Biomedical Engineering of Viterbi School of Engineering, University of Southern California, Los Angeles, CA, USA.
This study introduces a new method, Feynman Kac Reweighted Schröodinger Bridge Matching (FKRSBM), to harmonize tau PET imaging data. FKRSBM improves data alignment and disease classification by learning direct transport processes, overcoming limitations of existing harmonization techniques.
Area of Science:
- Neuroimaging
- Biomarker Harmonization
- Machine Learning in Medicine
Background:
- Tau PET imaging is crucial for Alzheimer's disease (AD) tracking.
- Site-specific differences in scanners, protocols, and tracers create nonbiological variability in tau PET data.
- Existing harmonization methods struggle with differing subgroup compositions, risking conflation of site effects with biological variation.
Purpose of the Study:
- To develop a novel harmonization model, Feynman Kac Reweighted Schröodinger Bridge Matching (FKRSBM), to address limitations of current methods.
- To accurately remove site-induced shifts in tau PET data while preserving biologically meaningful signals, even with differing cohort subgroup compositions.
- To improve the reliability of tau PET biomarkers for tracking Alzheimer's disease progression and clinical assessments.
Main Methods:
- Proposed FKRSBM model utilizes entropy-regularized optimal transport for direct stochastic transport between source and target distributions.
- Incorporates a subgroup-aware endpoint proposal via Feynman Kac reweighting for biologically consistent transport.
- Employs a spherical convolutional backbone on cortical meshes for vertex-level harmonization of surface-based neuroimaging data.
Main Results:
- FKRSBM demonstrated superior distributional alignment compared to ComBat, CycleGAN, DF, and DSBM when harmonizing PI-2620 data into the AV-1451 domain.
- Achieved reduced tau-positivity sign mismatch and stronger APOE subgroup alignment.
- Showcased improved downstream disease classification performance.
Conclusions:
- FKRSBM offers a robust solution for harmonizing multi-site tau PET data, effectively handling subgroup differences.
- The method enhances the accuracy and sensitivity of tau PET biomarkers for Alzheimer's disease research.
- FKRSBM represents a significant advancement in neuroimaging data harmonization, enabling more reliable clinical assessments and disease tracking.
Related Concept Videos
01:27¹H NMR: Long-Range Coupling
In alkenes, spin information is communicated via σ–π overlap, as seen in allylic (four-bond) and homoallylic (five-bond) couplings. These coupling interactions are stronger when the σ bond is parallel to the alkene π orbitals.
03:35Hybridization of Atomic Orbitals II
03:24Hybridization of Atomic Orbitals I
01:20¹³C NMR: Distortionless Enhancement by Polarization Transfer (DEPT)
01:34Woodward–Hoffmann Selection Rules and Microscopic Reversibility
01:12Double Resonance Techniques: Overview
Spin decoupling is usually achieved by...