Lijun An

The paper compares three plasma proteomic metrics across organ systems to assess their associations with brain structure and neurodegenerative disease, finding distinct relationships and modest predictive value.

The paper identifies robust multicellular axes associated with Alzheimer's disease pathology and sex using single-cell data from 427 donors, validated in an independent dataset.

97% of 184 high-impact biomedical AI studies compare cross-validated models with tests assuming fold independence; in 69% of re-analyzable studies a significant comparison does not survive correction.

The paper presents a harmonized framework for analyzing large-scale plasma proteomics data across neurodegenerative diseases, identifying disease-specific circulating signatures and biomarkers.

The paper shows longer scans improve prediction accuracy in brain-wide association studies while reducing costs, with 30-minute scans being the most cost-effective.

This study investigates how brain network dynamics and regional vulnerability influence tau propagation and load in Alzheimer's disease using multi-scale simulations.

This study examines the association between parenthood and brain function in mid- to late-life, finding that higher parity is linked to increased functional connectivity and counteracts age-related declines.

DeepResBat improves MRI data harmonization by addressing covariate distribution differences, outperforming existing methods while avoiding false associations.

This paper adapts meta-matching frameworks to translate phenotypic prediction models from large to small MRI datasets, showing significant performance improvements over classical methods.

The paper introduces multilayer meta-matching to improve phenotypic prediction in small datasets by translating models from multiple large datasets, outperforming classical methods and transfer learning.

The paper estimates the whole-cortex excitation-inhibition ratio using fMRI and shows it declines during youth, with lower ratios linked to better cognitive performance in children.

The paper develops homotopic local-global parcellations of the human cerebral cortex using resting-state and task-fMRI, showing improved homogeneity and lateralization compared to existing parcellations.

This paper introduces a goal-specific MRI harmonization framework that improves downstream prediction performance by integrating application-specific constraints during harmonization.

The paper introduces a framework called meta-matching to improve prediction performance in small datasets by leveraging correlations with large-scale phenotypic data.

This paper predicts Alzheimer's disease progression using a minimal recurrent neural network, showing better performance than baseline methods when handling missing data with model-based imputation.