Unsupervised photon-counting CT reconstruction for extremity imaging
Official title: Unsupervised Deep Photon-Counting Computed Tomography Reconstruction for Human Extremity Imaging
Funded at University of Massachusetts Lowell · NIH-11541294 · Data updated 2026-11-12
A new deep-learning approach aims to improve image quality for photon-counting CT scans of arms and legs with less radiation.
Quick facts
| Grant type | R01 grant |
|---|---|
| Study type | NIH-funded research |
| Funding institution | University of Massachusetts Lowell NIH-funded |
| Lab location | 1 site (Lowell, United States) |
| Project start | July 1, 2023 |
| Project end | April 30, 2027 |
| Funding year | 2026 |
| Award (this year) | $72,748 |
| NIH project number | 3R01EB034737-04S2 |
| Project ID | NIH-11541294 on NIH RePORTER |
Who could benefit from this research
✓Could help
Individuals needing high-quality extremity CT imaging, especially where material decomposition and contrast-enhanced imaging could improve diagnosis.
✕Less likely to help
Patients not undergoing photon-counting CT or those where image quality is already optimal may see minimal benefit.
What this research studies
This project focuses on making photon-counting CT images clearer for clinical use by using unsupervised deep learning techniques. It seeks to overcome instability issues seen in some neural networks and improve reconstruction quality without large labeled datasets. The work involves advanced algorithms, big data handling, and exploration of contrast agents like bismuth and gold nanoparticles to enhance tissue and material differentiation. The research is centered on practical imaging challenges to benefit patients who need accurate extremity scans.
Why it matters
Potential benefit: If successful, patients could receive clearer CT images with potentially lower radiation and better detection of tissue differences.
How similar studies have gone: Deep-learning reconstruction has shown promise but remains challenged by instability and data requirements; this project proposes unsupervised methods to address these gaps.
Where this research is happening
-
University of Massachusetts Lowell
Lowell, United States Active
Researchers
- Principal investigator: Yu, Hengyong — University of Massachusetts Lowell
- Study coordinator: Yu, Hengyong
Related research
Other NIH-funded projects on the same topics.
-
Unsupervised Deep Photon-Counting CT Reconstruction for Extremity Imaging
A team at a university will develop unsupervised deep learning methods to improve photon-counting CT images of arms and legs, aiming for clearer images with lower radiation.
-
Gut microbiome and brain health in Puerto Ricans with Alzheimer’s risk
We’re studying how the gut microbiome may influence cognitive aging and Alzheimer’s risk in Puerto Ricans aged 21 and older.
-
Studying how Satb2 helps Gli3 control jaw development without a morphogen gradient
A study in mice is exploring how Satb2 and Gli3 work together to control jaw development in a way that may not rely on a typical signaling gradient.
-
Gut viruses, diet, and Alzheimer’s risk in Latino adults
A small study will explore whether gut viruses are linked to Alzheimer's-related outcomes in Latino adults and how diet might influence that link.
-
Regional hub for lyophilization research and equipment access
A regional center at UMass Lowell will acquire a state-of-the-art freeze dryer to speed up early-stage formulation and method development for drugs and biologics.
-
How sex hormones shape the oral microbiome and health
This NIH grant explores how sex hormones influence the community of bacteria in the mouth and how that affects oral and overall health.
About this research
- This is an active NIH-funded research project — typically early-stage science, not a clinical trial accepting patient enrollment.
- Some NIH-funded labs run parallel clinical studies or seek volunteers for related work. To check, contact the principal investigator or institution listed above.
- For full project details, budget, and progress reports, visit the official NIH RePORTER page.