Evrim Acar

The paper uses low-rank tensor decompositions to analyze glymphatic MRI data and identifies spatiotemporal patterns of CSF tracer transport in the human brain, revealing differences between iNPH and reference cohorts.

This paper introduces coupled matrix factorizations to extract subject-specific temporal patterns from longitudinal data, showing improved insights in metabolomics and sensitization studies.

The paper introduces PARAFAC2 to model subject-specific temporal patterns in longitudinal microbiome data, outperforming CP in capturing individual variations and enabling replicable pattern discovery.

The paper introduces a method combining simulated and real metabolomics data using tensor factorization to improve pattern discovery, showing better performance in males for BMI-related phenotypes.

This study uses multiway data analysis to characterize postprandial metabolic responses in humans, revealing BMI-related differences in metabolite groups and temporal profiles between males.

This study evaluates CP models for analyzing postprandial metabolomics data through simulations, showing their effectiveness in revealing metabolic patterns and group differences.

This paper uses structure-revealing fusion of EEG, fMRI, and sMRI data to identify schizophrenia biomarkers, showing improved clustering and reproducible patterns compared to other methods.