AI and Machine Learning Projects

AI and Machine Learning in the Surgical Innovation Center

A major focus of the center is in harnessing the massive quantities of healthcare data to develop predictive models that can be used for individualized patient care and for supporting clinical decision-making. We focus on developing algorithms that can accurately and in an automated fashion interpret imaging studies, and using AI/ML and natural language processing to accurately and rapidly extract data from the electronic health record. Our team is actively involved in research and educational efforts to help increase the understanding of improving clinical outcomes through the implementation of  AI ML and NLP.

Current Projects in Artificial Intelligence and Machine Learning

Arman Kilic, M.D. 

• Machine Learning and AI approach to risk modeling for cardiac surgery for quality improvement
• Developing a more efficient allocation system in the United States for matching donors and recipients for heart transplantation
• Developing AI methods to extract unstructured data from the electronic health record to automate reporting of data and quality control 
• Utilizing AI and machine learning for clinical decision support, matching recipients and donors for heart transplants in the U.S.

Ian Bostock, M.D. MS

• Utilizing machine learning algorithms to analyze CT and PET scans of the esophagus to determine if the patient's cancer is in remission. 

Tom Brothers, M.D. 

• Developing risk models and providing clinical decision support in peripheral vascular disease.

Thomas Curran, M.D. 

• Utilizing machine learning for DVT personalized risk prediction and management recommendations for DVT prophylaxis. 

Heather Evans, M.D. MS 

• Developing infrastructure for participation in an established federated learning consortium with the University of Minnesota.

Evert Eriksson, M.D. 

• Applying artificial intelligence to chest wall injury for chest wall reconstruction surgery.
• Utilizing AI to determine factors associated with unstable chest wall injuries.
• Utilizing ML to analyze radiology reports to identify patients with non-healing rib fractures that would benefit from rib fixation treatment.

Kevin Hughes, M.D. 

• Using Artificial Intelligence/Natural Language Processing to help organize a Cancer Genetics KnowledgeBase and build Clinical Decision Support Tools 
• Using Graph Database technology for Cancer Susceptibility Gene Visualization to enhance education and understanding of cancer genetics
• Organizing and mapping disparate data sources to a Central Breast Cancer Database that is used for Machine Learning, clinical care and research
• Systematic reviews of the risk of cancer for each gene-cancer association
• Heredity Cancer Management APP

Rupak Mukherjee, M.D. 

• Predicting general surgery applicant board scores using natural language processing of their residency application packet 

Deepak Ozhathil, M.D. 

•  Utilizing Next Generation Sequencing (NGS) technology to collect data about the microbiome and metagenomics of the burn wound and establish a data and bio-repository.  

Ravi Veeraswamy, M.D. 

• AI approaches to detecting the severity of carotid artery disease