Objective

To investigate the effects of compressed sensing acceleration factors on the quality of nasopharyngeal rapid spin echo diffusion-weighted imaging (TSE-DWI)

Methods and Data

A total of 25 patients with nasopharyngeal carcinoma and 25 healthy volunteers (group A and group B) were successively recruited. All volunteers underwent plane echo imaging diffuse-weighted imaging (EPI-DWI) with scanning time of 35s and TSE-DWI combined with traditional parallel acquisition technology (SENSE) or (CS), with AF of 2.6 for SENSE. CS AF is 2.6, 3.6, 5.6, 7.6, The corresponding scanning time was 146s, 146s, 114s, 89s and 75s, respectively. Group A and group B were in EPI-DEI and TSE-DWI On the image b=1000s/mm², the cerebellar hemisphere at the level of pharyngeal crypt, the base of occipital bone, the inner wing, the maximum level of normal turbinate and nasopharyngeal carcinoma lesions were selected, and the ROI was defined to measure the performance diffusion coefficient (ADC) and signal strength (SI), and the background signal strength (air) was taken as the standard deviation. Signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) between lesions and normal tissues were calculated. Firedman was used to examine the differences in ADC values, SNR, CNR and subjective scores of TSE-DWI in groups A and B with different AF cerebellar hemispheres, occipcipal base, and nasopharyngeal carcinoma lesions. If the difference was statistically significant, the Wilcoxon test was used for pairwise comparison. Then, the optimal AF was selected and compared with EPI

Result

The image quality subjective scores of the two groups of observers were consistent (Kappa values 0.782, 0.871, 0.853), and the magnetic susceptibility pseudo-film scores, geometric deformation degree scores, SNR and CNR of TSE-DWI images were superior to EPI-DWI images (P values < 0.001). The contrast and ADC values of tumor foci in TSE-DWI images were not significantly different from those in EPI-DWI images (P value ≥0. 001). When AF of TSE-DWI group was compared with that of EPI-DWI group, when AF increased to 5.6, SNR, CNR and image score of cerebellar hemisphere, base of occiput and nasopharyngeal carcinoma lesions showed significant differences compared with SENSE2.6. When TSE-DWI CS3.6 was compared with EPI-DWI, The SNR and CNR of cerebellar hemisphere and occipital base showed statistical difference and higher score.

Conclusions

Compared with traditional EPI-DWI sequences, TSE-DWI sequences can effectively improve the nasopharyngeal geometric deformation artifacts. The scanning time of TSE-DWI combined with CS was gradually shortened with the increase of AF. When CS is 3.6, the scanning time of SENSE2.6 is 23.25% shorter than that of conventional SENSE2.6. CS3.6 is clinically recommended for nasopharyngeal TSE-DWI sequence scanning for patients with nasopharyngeal space occupying

Acknowledgements

No acknowledgement found.