Related Experiment Video
Updated: May 28, 2026
05:32
Retrospective Cardiac Gating with A Prototype Small-Animal X-ray Computed Tomograph
Published on: February 21, 2025
Evaluation of a super-resolution deep learning reconstruction algorithm in abdominal CT imaging-A qualitative and
Aria M Salyapongse1, Martin G Wagner1,2, Meghan G Lubner2
1Departments of Medical Physics and Radiology, University of Wisconsin-Madison, Madison, Wisconsin, USA.
Journal of Applied Clinical Medical Physics
|May 26, 2026
Summary
Precise Image Quality Engine (PIQE) deep learning reconstruction improved abdominal CT image quality over Advanced Intelligent Clear-IQ Engine (AiCE). PIQE demonstrated superior image contrast, reduced noise, and enhanced sharpness and signal-to-noise ratios.
Area of Science:
- Radiology
- Medical Imaging
- Deep Learning in Medical Imaging
Background:
- Super-resolution deep learning (DL) algorithm Precise Image Quality Engine (PIQE), initially for cardiac CT, is now applicable to abdominal CT.
- Evaluation of PIQE's performance in abdominal CT is crucial for optimizing diagnostic accuracy.
Purpose of the Study:
- To compare the objective and subjective image quality (IQ) improvements offered by PIQE versus Advanced Intelligent Clear-IQ Engine (AiCE) in abdominal CT.
- Assess the impact of different PIQE strength levels (L1 and L2) on image reconstruction.
Main Methods:
- Retrospective analysis of 69 contrast-enhanced abdominopelvic CT exams reconstructed with PIQE (L1, L2) and AiCE (L1).
- Four blinded radiologists evaluated subjective IQ using Likert scales for noise, contrast, sharpness, and artifacts.
- Objective IQ metrics including CT number, noise, signal-to-noise ratio (SNR), and contrast-to-noise ratio (CNR) were quantified in specific anatomical regions.
Main Results:
- Reader agreement for subjective assessments was fair (mean alpha = 0.20).
- PIQE L2 showed significant improvements in image contrast and noise reduction compared to AiCE L1 (p < 0.05).
- PIQE L2 achieved the highest SNR and CNR, indicating superior objective image quality (p < 0.003).
Conclusions:
- PIQE significantly enhances subjective image quality, particularly for contrast, noise, and sharpness, in abdominal CT.
- PIQE L2 reconstruction provides the best objective image quality metrics (SNR and CNR).
- The findings support the use of PIQE for improved image contrast and reduced noise in abdominal CT examinations compared to AiCE.