Walter Hugo Lopez Pinaya

Confounder-aware foundation modeling for accurate phenotype profiling in cell imaging

This paper introduces a confounder-aware foundation model for cell imaging that improves phenotype profiling and compound target prediction, achieving state-of-the-art results for both seen and unseen compounds.

A Graph Neural Network Approach to Investigate Brain Critical States Over Neurodevelopment

This paper estimates the Ising temperature of the brain from fMRI data using GNNs, finding a significant negative correlation between age and temperature in typically developing children and ADHD patients.

Inferring the heritability of large-scale functional networks with a multivariate ACE modeling approach

The paper estimates the heritability of functional brain networks using a multivariate ACE model and twin data, finding genetic influences on specific networks.

Normative modelling using deep autoencoders: a multi-cohort study on mild cognitive impairment and Alzheimer's disease

This study applies deep autoencoders to structural neuroimaging data from Alzheimer's and mild cognitive impairment patients to quantify deviations from healthy norms and compare performance with traditional classifiers.