Danilo Bzdok

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.

Cell type-specific logistic regression estimators trained on 427 ROSMAP prefrontal donors project Braak- and β-amyloid-like transcriptomic signatures onto the 108-region Siletti atlas, flagging brainstem, hypothalamic and rostral hippocampal regions.

The paper examines how cognition and personality relate to adolescent mental health, showing these associations vary with brain, environmental, and pubertal factors.

This study shows that cognitive co-activation dominates fMRI synchrony in the awake brain but is replaced by anatomical and molecular constraints during consciousness perturbations.

The paper identifies multiscale principles of brain reorganization under electrical perturbation by integrating neuroimaging, connectomics, and transcriptomics data from 148 experiments.

This study examines how LRRK2 pathogenic variants affect brain structure in Parkinson's disease patients and non-manifesting carriers, finding milder cortical atrophy in LRRK2 PD patients and no atrophy in carriers.

This paper uses AI to analyze neuroscientific literature and discovers conserved cognitive circuits across species.

The paper examines how changes in functional connectivity during rest and task relate to Alzheimer's pathology and episodic memory in older adults.

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 paper improves brain age prediction models by reducing generalization error through preprocessing, data augmentation, and regularization.

The paper identifies latent impulsivity dimensions in adolescents using PCA and links them to substance use and brain network connectivity in the ABCD study.

The paper introduces Brain-DiT, a deep generative model for analyzing task-fMRI data, identifying cognitive regions and subgroups linked to psychiatric conditions like depression and alcohol abuse.

The paper identifies shared molecular mechanisms between Alzheimer's and Parkinson's disease by analyzing gene co-expression at sub-cell-type resolution using single-nucleus datasets.

This study shows psilocybin induces cell-type specific transcriptional changes in mouse medial prefrontal cortex, with sustained neuroplastic effects mediated by 5-HT2C signaling in L5/6 NP neurons.

This paper introduces a hierarchical Bayesian framework that fuses task-based and resting-state neuro-imaging datasets to create a population-based atlas of the human cerebellum, achieving better individual parcellations with limited data.

The paper identifies three ASD subtypes using morphometric similarity networks and finds distinct transcriptomic and functional connectome patterns associated with each subtype.

The paper identifies structural covariation patterns between cerebellum subregions and cerebral cortex in a large cohort, revealing dissociations and novel associations with phenotypes.

The paper investigates the impact of rare high-effect genetic variants on brain morphology and related behavioral traits in adolescents using the ABCD cohort.

The paper investigates how rare genetic mutations affect brain asymmetry by analyzing CNVs in a cohort of individuals and linking genetic alterations to brain structure and function.

The paper shows how the distribution of non-neuronal cell types in the brain predicts tissue damage patterns in thirteen neurodegenerative conditions.

This study uses Bayesian modeling to analyze stroke patient data and identifies distinct brain regions associated with language and executive control impairments.

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

This study shows that prediction performance for brain-imaging-based phenotypes can improve significantly with larger sample sizes and multimodal data.

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.

This study shows that general anaesthesia reduces the uniqueness of human brain connectivity both between individuals and across species.

The paper examines functional brain network organization in alcohol-dependent patients to identify predictors of relapse during a 12-month follow-up.

This study investigates the relationship between prediction accuracy and feature importance reliability in neuroimaging data, showing that feature importance reliability can be high with sufficient sample size and is positively correlated with prediction accuracy.

The paper models sex differences in white matter substrates of aphasia using Bayesian analysis in 1,401 stroke patients, identifying sex-specific lesion effects in language networks.

The paper investigates how brain network structure, neuromodulation, and disease influence transitions between cognitive activation patterns using multimodal neuroimaging data.

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.