Anant Dadu

CARDBiomedBench evaluates LLM performance in biomedical research using a benchmark focused on neurodegenerative diseases, revealing significant limitations in response quality and safety.

This study evaluates federated learning models for multi-omics Parkinson's disease prediction, finding they perform comparably to centrally trained models with minor performance differences.

This study clusters patients with neurodegenerative diseases based on genetic risk variants, revealing overlapping genetic etiology and subsets with no common risk factors.

This study uses machine learning to identify Parkinson's disease subtypes and predict progression in two cohorts, finding three subtypes with distinct progression rates and high prediction accuracy.

This paper develops an automated ML model to predict Parkinson's Disease using multi-modal data, achieving high AUC scores and constructing gene networks relevant to PD.

This paper applies Aligned-UMAP to longitudinal biomedical datasets to reveal time-dependent patterns and provides best practices for its use.

The paper presents a fully automated FAIMS-DIA proteomic pipeline for iPSC-derived neurons, showing high protein identification and reproducibility, validated against single-cell RNA-Seq data.

This paper uses machine learning to predict Alzheimer's disease progression rates and classify clinical subtypes based on early clinical data.