Grants and Partnerships

To create a contemporary, population-based data resource that quantifies real-world imaging surveillance after curative therapy for stage I NSCLC, identifies gaps across rural and urban settings in Florida and Georgia, and informs targeted interventions to reduce recurrence and mortality.

Comparison of Anti-coagulation and anti-Platelet Therapies for Intracranial Vascular Atherostenosis

Understanding pancreatic endocrine and exocrine loss in pre-type 1 diabetes

Prospective Evaluation of Chronic Pancreatitis for Epidemiologic and Translational Studies (The PROCEED Study)

Diabetes RElated to Acute Pancreatitis and Its Mechanisms (DREAM) An Observational Cohort Study from the Type 1 Diabetes in Acute Pancreatitis Consortium (T1DAPC)

This project aims to develop advanced AI models to identify imaging biomarkers for Alzheimer’s disease using multimodal data from multicenter cohorts and the UF-ADIR dataset. By combining brain imaging, laboratory, and cognitive data, the AI system will detect patterns associated with disease while providing interpretable insights for clinicians. The approach emphasizes robust, reproducible results across different scanners and settings, supporting broader clinical application and improved understanding of Alzheimer’s disease.

To evaluate the diagnostic performance of multiparametric lung MRI with diffusion-weighted imaging (DWI) in the characterization of Lung-RADS 4 lesions by (1) determining its sensitivity and specificity in differentiating malignant from benign lesions, (2) assessing whether DWI alone provides comparable accuracy to the full MRI protocol, and (3) comparing its diagnostic performance to PET/CT.

To assess the feasibility and diagnostic yield of simultaneously detecting thoracic aortic pathology on LDCT performed for lung cancer screening, thereby broadening preventive benefits without additional radiation or cost.

This retrospective study provides real-world evidence on how advanced deep-learning reconstructions (PIQE) plus retrospective CT perfusion might improve diagnostic accuracy and reduce downstream costs compared to standard reconstructions by leveraging existing CCTA datasets.

The primary objective of this study is to assess the difference in the nodule detection rate and the percentage of lung cancer diagnosed through nodule route between the pre and post deployment period.  Secondary objective is to assess whether AI can aid in detecting more early-stage lung cancer. Exploratory objectives include summarizing the reason behind patients dropping out of nodule clinic pathway.

Using the infrastructure at the Biomedical Radiology Research and Artificial Intelligence Navigation (Br²AIn) Lab, our role is to advise and integrate the classification model within the radiological workflow, including integration with a clinical grade viewer(s) (commonly referred to as PACS).

Development of Precision Imaging Network™ (PIN) as a single enterprise-wide AI platform, the development, clinical validation, testing and adoption of the Third Party Applications.

Deep Learning Spectral is a new technology offed with Canon CT systems. To maximize the benefits of spectral CT radiologist training is key. There is need to understand the new types of images produced by Spectral CT and to know where and when to use these new images.

The purpose of this project is to evaluate the capabilities of Canon’s Deep Learning Spectral CT technology and Canon’s Deep Learning Reconstruction.

Grant for the acquisition of the NPIX system at the University of Florida, to be accessible to students, researchers, and industry partners, and positions the university to be the only one worldwide equipped with such capability.

To develop and clinically validate an AI algorithm that rapidly identifies acute pulmonary embolism, a critical competing risk in patients undergoing lung imaging for cancer evaluation.

To develop and openly share annotated thoracic CT datasets and AI algorithms that advance early detection and characterization of idiopathic pulmonary fibrosis and related interstitial lung diseases, conditions that often mimic or coexist with lung cancer.

The hypothesis is that AI based Computer Aided Detection (CAD) device can help physicians detect more nodules on CXR and this can potentially increase the number of cancers diagnosed.