# Pilot Project 1

> **NIH NIH U54** · JOHNS HOPKINS UNIVERSITY · 2024 · $23,259

## Abstract

Project Summary – Pilot Project
Prostate cancer is a significant disease in men, particularly among African Americans who face higher
incidence and mortality rates than Caucasians. Late detection contributes to increased mortality, with current
screening methods having limitations. Our proposed solution involves developing a non-invasive imaging
modality utilizing molecular CEST MRI on standard 3T MRI scanners, combined with a machine learning-
based metabolic imaging processing pipeline. This aims to provide a precise, cost-effective tool for detecting
and prognosing prostate cancer in underserved communities.
Molecular MRI, though challenging, can be enhanced using machine learning superresolution techniques. By
simulating metabolic imaging in organ-scale vascular networks, it is possible to obtain high resolution images
for training a superresolution neural network. We hypothesize that CEST MRI with subject-specific metabolism
simulation will enable detection of metabolic activity at small scales, which will improve prediction of cancer
progression. Our study aligns with the Howard-Hopkins Partnership priorities, addressing the urgent need for
accurate detection and prediction of cancer aggressiveness in African American men, ultimately aiming to
reduce mortality rates and disparities.
The specific aims of our study include: 1) using dynamic CEST MRI to analyze energy metabolism variations in
aggressive prostate tumors, mapping glucose uptake in PSMA-positive and negative cancers; 2) developing a
superresolution machine learning model for improved spatial resolution in metabolic imaging using subject-
specific metabolism simulation; and 3) validating patient-specific metabolic/perfusion models generated from
magnetic resonance angiography (MRA) and anatomic images in a small pilot project and assessing the
quality of superresolution results through image similarity metrics and ratings.

## Key facts

- **NIH application ID:** 11012040
- **Project number:** 1U54CA295336-01
- **Recipient organization:** JOHNS HOPKINS UNIVERSITY
- **Principal Investigator:** Curtiland Deville
- **Activity code:** U54 (R01, R21, SBIR, etc.)
- **Funding institute:** NIH
- **Fiscal year:** 2024
- **Award amount:** $23,259
- **Award type:** 1
- **Project period:** 2024-09-20 → 2029-08-31

## Primary source

NIH RePORTER: https://reporter.nih.gov/project-details/11012040

## Citation

> US National Institutes of Health, RePORTER application 11012040, Pilot Project 1 (1U54CA295336-01). Retrieved via AI Analytics 2026-05-26 from https://api.ai-analytics.org/grant/nih/11012040. Licensed CC0.

---

*[NIH grants dataset](/datasets/nih-grants) · CC0 1.0*