Radiating Hope: Radiology Department Blog


Prediction of Fontan Outcomes Using T2-Weighted MRI Radiomic Features and Machine Learning 

Post Date: March 31, 2023
<strong>Prediction of Fontan Outcomes Using T2-Weighted MRI Radiomic Features and Machine Learning</strong> 

Beginning this past summer, I have been grateful for the opportunity to conduct research under the mentorship of Dr.?Jonathan Dillman in the Cincinnati Children’s Hospital Radiology department. This work has intersected my current interests in pediatric radiology, pediatric cardiology, and artificial intelligence. 

My research project focuses on pediatric patients born with only one pumping chamber, or ventricle, in the heart instead of two pumping chambers. In most cases, children with single-ventricle heart defects require intensive medical intervention shortly after their birth. The Fontan procedure is?a type of open-heart surgery performed in these patients to avoid overworking the single ventricle in pumping blood to the entire body. The surgery allows venous blood returning from the body to enter the lungs by direct blood vessel connections that avoid passing through the single ventricle. However, because of the rerouting of blood in these patients, elevated pressures in the veins can lead to Fontan-associated liver disease, which has become an increasing cause of morbidity and mortality in these post-procedure patients. 

Research Poster

The goal of my research project has been to develop machine learning models using MRI images of the liver as well as clinical data from post-Fontan patients to predict future clinical outcomes in these patients within two years of their MRI examination. Some clinical outcomes we have studied include the need for cardiac transplantation or a ventricular assist device, elevated liver stiffness, and the presence of portal hypertension. Because of the high prevalence of Fontan-associated liver disease, we hope this research can ultimately provide both radiologists and cardiologists with surveillance strategies during childhood for patients with Fontan circulation. I look forward to continuing to work with Dr. Dillman and the research team to refine our machine learning models and be able to one day provide patients and their families with the best diagnostic performance. 

At the end of last summer Ayush created a research poster explaining his research for an upcoming medical conference. Click here to view the poster

Author: Ayush Prasa

Ayush Prasad, BA, UCCOM MS2, author; Glenn Miñano, BFA, editor; Meredith Towbin, copy editor 

Avatar photo

About the author: Glenn Miñano

Glenn Miñano is a media specialist in the Department of Radiology, providing graphic design, photography, printing, video services, and administration of the department’s online properties. His works have been published in several medical articles, such as the American Journal of Radiology and the American Institute of Ultrasound. He has been providing these services to the Radiology Department since 1996.