Joshua Levy

This study uses microspectroscopic techniques to analyze sulfur species and zinc distribution in rat temporomandibular joint condylar zones, linking them to biological pathways and potential diagnostic biomarkers.

TRACE co-registers elemental imaging and histopathology to enable spatial analysis of tissue sections through a web application.

This paper develops a deep learning algorithm for 3D-HDAM to detect dyssynergia, showing comparable accuracy to traditional and hybrid models but better handling of ambiguous cases.

The paper presents a modular pipeline for federated health data analysis using deep learning to predict health outcomes across diseases.

The paper introduces MethylCapsNet and MethylSPWNet, deep learning models that organize DNA methylation data into biologically relevant capsules to improve interpretability in classifying CNS tumors.

The paper introduces GCN4R, an R library for fitting graph neural networks on networks to generate embeddings for network-based tasks like community detection.

The paper develops a hybrid logistic regression model enhanced by machine learning to improve predictive performance while maintaining interpretability.

MethylNet is a deep learning framework for DNA methylation analysis that constructs embeddings, makes predictions, and uncovers heterogeneity with minimal user input.

PolyCRACKER separates closely related polyploid subgenomes using repetitive DNA signatures without prior knowledge or related genomes.