Application of three-dimensional fluid-attenuated inversion recovery sequence using artificial intelligence-assisted compressed sensing technique in intravenous gadolinium contrast-enhanced magnetic resonance imaging of inner ear
10.12025/j.issn.1008-6358.2025.20250063
- VernacularTitle:人工智能辅助压缩感知三维液体衰减反转恢复序列在静脉内耳钆造影中的应用
- Author:
Kai LIU
1
;
Jian WANG
2
;
Huaili JIANG
3
;
Shujie ZHANG
1
;
Di WU
1
;
Xinsheng HUANG
3
;
Mengsu ZENG
1
;
Menglong ZHAO
1
Author Information
1. Department of Radiology, Zhongshan Hospital, Fudan University, Shanghai Institute of Medical Imaging, Shanghai 200032, China.
2. Department of Radiology, Juxian People’s Hospital, Rizhao 276500, Shandong, China.
3. Otolaryngology Head and Neck Surgery, Zhongshan Hospital, Fudan University, Shanghai 200032, China.
- Publication Type:Monographicreport:Multi-disciplinarydiagnosisandtreatmentofvertigo
- Keywords:
endolymphatic hydrops;
3D-FLAIR;
artificial intelligence-assisted compressed sensing;
united compressed sensing;
acquisition time
- From:
Chinese Journal of Clinical Medicine
2025;32(2):212-217
- CountryChina
- Language:Chinese
-
Abstract:
Objective To investigate the value of artificial intelligence-assisted compressed sensing (ACS) technology for intravenous gadolinium contrast-enhanced magnetic resonance imaging of the inner ear using three-dimensional fluid-attenuated inversion recovery (3D-FLAIR) sequence. Methods The patients received gadolinium contrast-enhanced magnetic resonance imaging using ACS and united compressed sensing (uCS) 3D-FLAIR at Zhongshan Hospital, Fudan University from January to November 2024 were prospectively enrolled. The repetition time was 16 000 ms, and acquisition time was 6 min 40 s and 10 min 24 s in ACS 3D-FLAIR and uCS 3D-FLAIR, respectively. The images on the two sequences were evaluated independently by two radiologists. The image quality of the two sequences was subjectively evaluated and compared. The signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) were compared between the two sequences. The grading consistencies using two sequences and between the two doctors were analyzed. Results There was no statistically difference in subjective score of image quality between the two sequences. SNR and CNR of the ACS 3D-FLAIR sequence were significantly higher than those of the uCS 3D-FLAIR sequence (P＜0.001). The kappa values of grades of cochlear and vestibular endolymphatic hydrops were 0.942 and 0.888 using two sequences (P＜0.001). The kappa values of grades of cochlear and vestibular endolymphatic hydrops using the ACS 3D-FLAIR sequence between the two doctors were 0.784 and 0.831, respectively (P＜0.001); the kappa values of grades of cochlear and vestibular endolymphatic hydrops using uCS 3D-FLAIR sequence between the two doctors were 0.725 and 0.756, respectively (P＜0.001). Conclusions ACS 3D-FLAIR could provide higher SNR and CNR than uCS 3D-FLAIR, and is more suitable for intravenous gadolinium contrast-enhanced magnetic resonance imaging of the inner ear; the endolymphatic hydrops grades using ACS 3D-FLAIR is similar to use uCS 3D-FLAIR.
