Heart and cardiovascular system

Cervical lymph node segmentation and labeling

This project applies deep learning to segment lymph nodes in the neck. This approach is often useful as a marker of disease progression in several types of cancers. Because there are many lymph nodes, we also are working to make it easier to track each node to confirm the change in size of each node.

Classification of cardiac diseases

This project applies deep learning methods to cardiac ultrasound imaging, also known as echocardiography, to classify diseases of the heart and cardiovascular system.

Coronary artery disease

This project uses algebraic topology to identify patterns and extract important information from imaging data related to coronary artery disease.

Detection of focal cortical dysplasia and posttreatment outcomes of epilepsy

The goal of this project is to develop a deep learning tool that combines MRI scans, positron emission tomography (PET) scans, electroencephalograms and pathology data to automatically detect and segment focal cortical dysplasia lesions. The Radiology Informatics Lab's approach involves conducting radiomics and image-based deep learning analyses of MRI to evaluate the outcomes of drug or surgical treatments in people with epilepsy.

Glioblastoma progression and pseudoprogression biomarkers

This project is developing a biomarker that uses early MRI to tell if early enhancement is due to tumor progression or caused by treatment effects that mimic tumor progression, also known as pseudoprogression. An accurate determination of true progression versus pseudoprogression is critical. Effective therapy should continue during treatment, but if the tumor is progressing, second line agents can be beneficial.

Multiagentic framework

The lab developed a multiagentic evaluation framework for echocardiography through a cardiology-specialized multiagent architecture. The framework extends the original six error categories with a seventh category dedicated to significant measurement differences. This is defined as numerical deviations exceeding 10% from reference values. Each specialist agent is augmented with a tool-based measurement calculator to eliminate arithmetic hallucination.

Parotid tumor classifier

The lab is developing a deep learning model trained on preoperative CT imaging, using surgical pathology as ground truth, to predict tumor type before the patient enters the operating room. This could reduce costs and complications in this group of patients.

Thyroid ultrasound evaluation of nodules

Thyroid nodules are common, but it can be hard to tell noncancerous nodules from cancerous ones. And not all cancerous nodules need to be treated aggressively. This project attempts to use ultrasound imaging to identify whether nodules are cancerous and aggressive.