Related Experiment Video
Updated: Jun 20, 2026

MR Molecular Imaging of Prostate Cancer with a Small Molecular CLT1 Peptide Targeted Contrast Agent
Published on: September 3, 2013
Dual Low-b-value-driven U-shaped fusion GAN for synthesizing high-b-value prostate DWI
Rui Feng1, Qun Huang2, Aiping Chen3
1Jiangsu Province Engineering Research Center of Smart Wearable and Rehabilitation Devices, School of Biomedical Engineering and Informatics, Nanjing Medical University, Nanjing, 211166, China.
A novel deep learning method synthesizes high-b-value prostate diffusion-weighted imaging (DWI) from low-b-value data. This UsFGAN approach improves image quality and diagnostic accuracy for prostate cancer detection, reducing scan times.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Radiology
Background:
- High-b-value diffusion-weighted imaging (DWI) is crucial for prostate cancer detection, offering superior contrast.
- Direct acquisition of high-b-value DWI is limited by long scan times, technical demands, and image artifacts.
Purpose of the Study:
- To develop a novel deep learning method for synthesizing high-b-value prostate DWI from low-b-value data.
- To improve the efficiency and accessibility of high-quality prostate DWI for clinical use.
Main Methods:
- A dual low-b-value-driven U-shaped Fusion Generative Adversarial Network (UsFGAN) was proposed.
- The network integrates U-Net subnetworks, Swin-Transformer with residual blocks (STRB), and a hierarchical fusion network with multiple feature fusion blocks (MFFB).
- Validation was performed on a multi-center dataset of 280 subjects.
Main Results:
- The UsFGAN method demonstrated superior performance compared to state-of-the-art models (CycleGAN, Pix2Pix, DiscoGAN) with high signal-to-noise and structural similarity metrics.
- Synthesized high-b-value DWI achieved 86.3% accuracy in prostate lesion detection.
- Qualitative evaluation by radiologists confirmed synthesized images were comparable to real high-b-value scans in noise suppression, artifact reduction, and diagnostic acceptability.
Conclusions:
- UsFGAN effectively synthesizes high-quality high-b-value prostate DWI by leveraging complementary low-b-value information.
- The method shows significant potential to reduce scan time and enhance the diagnostic value of prostate MRI for disease assessment.
Related Concept Videos
¹H NMR: Complex Splitting
Splitting diagrams or splitting tree diagrams are routinely used to depict such complex couplings. While drawing splitting diagrams, the splitting with the larger coupling constant is usually applied first.
¹H NMR Signal Multiplicity: Splitting Patterns
Interpreting ¹H NMR Signal Splitting: The (n + 1) Rule
Spin–Spin Coupling: Two-Bond Coupling (Geminal Coupling)
The central atom need not be NMR-active because its electrons are affected by the electron polarization of the spin-active atoms. However, spin information is transmitted less effectively than in one-bond coupling, and 2J values are usually weaker than 1J values. The energy of...
¹³C NMR: ¹H–¹³C Decoupling
A broadband decoupling technique is used to simplify these complex, sometimes overlapping, signals. Broadband decoupling relies on a...
Urodynamic Studies: Uroflowmetry

