Nerisa Banaj

CALM-VLM: CALIBRATION AND SELECTIVE PREDICTION IN VISION-LANGUAGE MODELS FOR RELIABLE BRAIN MRI CLASSIFICATION

CALM-VLM improves confidence and reliability in brain MRI classification by calibrating vision-language models and enabling selective prediction.

Neuroanatomical Correlates of Negative Symptoms in Schizophrenia

This study identifies small but consistent anatomical correlates of negative symptom dimensions in schizophrenia, linking them to prefrontal cortical thickness and limbic subcortical volumes.

Genome-wide analyses identify 30 loci associated with obsessive-compulsive disorder

This study identifies 30 genome-wide significant loci for obsessive-compulsive disorder (OCD) using a large GWAS, linking risk to specific brain cell types and shared genetic risk with other psychiatric and non-psychiatric traits.

Brain-Age Prediction: Systematic Evaluation of Site Effects, and Sample Age Range and Size

The paper evaluates factors affecting brain-age prediction models using structural neuroimaging data from 35,683 individuals, finding optimal model performance with no site harmonization, age-binning, and a sample size of over 1,600.

Estimating multimodal brain variability in schizophrenia spectrum disorders: A worldwide ENIGMA study

This study examines structural brain heterogeneity in schizophrenia using multimodal MRI data from the ENIGMA dataset, finding higher frontotemporal heterogeneity and left parahippocampal homogeneity in patients.

Normative Modeling of Brain Morphometry Across the Lifespan using CentileBrain: Algorithm Benchmarking and Model Optimization

The paper benchmarks and optimizes a normative modeling framework for brain morphometry across the lifespan using MFPR, showing excellent accuracy and stability.

Bridging Big Data: Procedures for Combining Non-equivalent Cognitive Measures from the ENIGMA Consortium

The paper demonstrates a data harmonization process for memory testing across 53 studies, reducing inter-site variance while preserving covariate effects.

Chronic stroke sensorimotor impairment is related to smaller hippocampal volumes: An ENIGMA analysis

This study finds that smaller ipsilesional hippocampal volume is associated with worse upper limb sensorimotor impairment in chronic stroke patients, with a stronger effect in women.

Smaller spared subcortical nuclei are associated with worse post-stroke sensorimotor outcomes in 28 cohorts worldwide

This study finds that smaller spared subcortical nuclei are linked to worse sensorimotor outcomes after stroke, varying by time since stroke and type of impairment.

White Matter Microstructure and its Relation to Clinical Features of Obsessive-Compulsive Disorder: Findings from the ENIGMA OCD Working Group

This study identifies white matter microstructural alterations in adult OCD patients and links them to clinical features, while finding no such abnormalities in pediatric patients.

Deep Learning for Quality Control of Subcortical Brain 3D Shape Models

The paper develops deep learning models to assess the quality of subcortical brain shape models from MRI, reducing human rater time by 46-70%.

Machine Learning for Large-Scale Quality Control of 3D Shape Models in Neuroimaging

This work uses machine learning to automate quality control of 3D brain shape models, reducing human workload by 30-70% across multiple datasets and diseases.