Explore the models that McWilliams School of Biomedical Informatics (MSBMI) has developed using artificial intelligence.
Specific projects often involve collaborative teams led by MSBMI principal investigators specializing in areas like clinical natural language processing, cancer informatics, and genomic data science.
The core development of these models occurs in-house at UTHealth Houston MSBMI. While we often build upon foundational AI research from the broader scientific community, the customization, training on specific biomedical datasets (like local clinical data, adhering to privacy regulations), and fine-tuning for specific healthcare tasks are performed by our teams. Some projects may also involve collaborations with other departments at UTHealth Houston or external partners.
Represents our strength in processing and understanding complex clinical text data from electronic health records, which is foundational for many clinical AI applications.
Showcases our application of advanced AI in high-impact disease areas like oncology or genomics, demonstrating translational potential.
Helping biologists automate genetic data analysis by generating protein interaction networks and searching scientific literature.
Disentangling Accelerated Cognitive Decline from the Normal Aging Process and Unraveling Its Genetic Components.
Brain imaging is a high-content modality that offers dense insights into the structure and pathology of the brain.
Folding-Docking-Affinity (FDA) is a framework which folds proteins, determines protein-ligand binding conformations, and predicts binding affinities.
Unsupervised Deep Representation Learning Enables Phenotype Discovery for Genetic Association Studies of Brain Imaging.
The PK-RNN-V model is a recurrent neural network that models the pharmacokinetics of vancomycin and predicts the vancomycin concentration.
Gene2vec is a distributed representation of genes based on co-expression. From a pure data-driven fashion, we trained a 200-dimension vector representation of all human genes.
