Related Experiment Video
Updated: Jun 28, 2025
15:48
Tracking the Mammary Architectural Features and Detecting Breast Cancer with Magnetic Resonance Diffusion Tensor Imaging
Published on: December 15, 2014
22.5K
Deep learning-based compressed SENSE improved diffusion-weighted image quality and liver cancer detection: A
Ting Duan1, Zhen Zhang1, Yidi Chen1
1Department of Radiology, Functional and Molecular Imaging Key Laboratory of Sichuan Province, West China Hospital, Sichuan University, Chengdu, Sichuan 610041, China.
Magnetic Resonance Imaging
|April 11, 2024
Summary
Deep learning with Compressed SENSE diffusion-weighted imaging (DL-CS-DWI) enhances liver lesion detection in patients at risk for hepatocellular carcinoma (HCC). This advanced technique improves image quality, particularly for lesions in the hepatic dome.
Area of Science:
- Radiology and Medical Imaging
- Hepatobiliary Diseases
- Oncology
Background:
- Hepatocellular carcinoma (HCC) surveillance requires sensitive imaging techniques.
- Diffusion-weighted imaging (DWI) is crucial for detecting liver lesions.
- Accelerated MRI techniques aim to improve efficiency and image quality.
Purpose of the Study:
- To evaluate the efficacy of deep learning with Compressed SENSE DWI (DL-CS-DWI) in improving image quality and liver lesion detection.
- To compare DL-CS-DWI against conventional parallel imaging DWI (PI-DWI) and standard Compressed SENSE DWI (CS-DWI) in at-risk HCC patients.
Main Methods:
- Prospective single-center study of 67 participants at high risk for HCC.
- Comparison of three DWI sequences: PI-DWI, CS-DWI, and DL-CS-DWI, all acquired at 3.0T with gadoxetate disodium.
- Qualitative and quantitative image quality assessment by masked reviewers.
- Liver lesion detection rates evaluated by masked reviewers against consensus from non-DWI sequences.
Main Results:
- DL-CS-DWI demonstrated superior qualitative and quantitative image quality compared to PI-DWI and CS-DWI.
- DL-CS-DWI achieved higher detection rates for clinically relevant liver lesions (91.4%-93.4%) than PI-DWI (82.7%-85.8%).
- DL-CS-DWI showed significantly improved detection rates for lesions in the hepatic dome (94.8%-97.4%).
Conclusions:
- DL-CS-DWI offers improved image quality and enhanced detection of clinically relevant liver lesions in patients at high risk for HCC.
- The technique is particularly beneficial for detecting lesions in challenging locations like the hepatic dome.
- DL-CS-DWI represents a promising advancement in HCC surveillance imaging.
