Related Experiment Video
Updated: Jul 13, 2025
09:30
Quantitative Magnetic Resonance Imaging of Skeletal Muscle Disease
Published on: December 18, 2016
19.6K
Highly accelerated knee magnetic resonance imaging using deep neural network (DNN)-based reconstruction: prospective,
Joohee Lee1, Min Jung2, Jiwoo Park1
1Department of Radiology, Research Institute of Radiological Science, and Center for Clinical Imaging Data Science (CCIDS), Yonsei University College of Medicine, 50-1 Yonsei-ro, Seodaemun-gu, Seoul, 03722, Korea.
Scientific Reports
|October 12, 2023
Summary
Deep neural network (DNN) reconstruction accelerates 2D fast spin-echo (FSE) knee MRI scans by 41%, maintaining comparable image quality and diagnostic performance. This DNN-based approach shows promise for efficient clinical MRI practices.
Area of Science:
- Radiology
- Medical Imaging
- Artificial Intelligence
Background:
- Accelerated MRI techniques are crucial for improving patient throughput and comfort.
- Deep neural networks (DNNs) offer potential for enhancing image reconstruction in MRI.
- Fast spin-echo (FSE) sequences are commonly used for knee imaging.
Purpose of the Study:
- To evaluate the performance of a deep neural network (DNN)-based reconstruction for accelerated 2D fast spin-echo (FSE) knee MRI.
- To assess the impact of DNN reconstruction on scan time, image quality, and diagnostic performance.
- To determine the feasibility of DNN-accelerated knee MRI across multiple vendors.
Main Methods:
- Prospective, multi-reader, multi-vendor study involving 45 subjects.
- Acquisition of conventional and accelerated 2D FSE knee MRI sequences on 3T scanners.
- Reconstruction of accelerated FSE images using DNN software (FSE-DNN).
- Evaluation of quantitative metrics (SNR, CNR) and diagnostic performance by musculoskeletal radiologists.
Main Results:
- Accelerated FSE-DNN reduced scan times by an average of 41.0%.
- FSE-DNN demonstrated significantly improved signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) (p < 0.001).
- Overall image quality and lesion detection performance were comparable to conventional FSE (p > 0.05).
- Good inter-reader agreement was observed for FSE-DNN compared to conventional FSE (R² = 0.76, p < 0.001).
Conclusions:
- DNN-based reconstruction is effective for accelerating 2D FSE knee MRI on multi-vendor platforms.
- The technique significantly reduces scan time while maintaining diagnostic image quality.
- DNN-accelerated knee MRI holds potential for clinical implementation, enhancing efficiency and patient experience.
