【摘要】 目的 对比研究人工智能压缩感知单激发液体衰减反转恢复（ACS-SS-FLAIR）技术与并行成像液体衰减反转恢复（PI-FLAIR）技术在头颅MR检查中的成像特点。方法 在40例行头颅MR检查中同时应用2种技术行T₂FLAIR序列成像,ACS-SS组使用单激发液体衰减反转恢复（SS-FLAIR）序列,PI组使用常规PI-FLAIR序列,对2种技术的图像质量、病灶显示和成像时间进行对比分析。结果 ACS-SS组总体图像质量评分与PI组无显著差异（3.20±0.61 vs 3.47±0.78,P>0.05）,ACS-SS组图像信噪比（SNR）低于PI组,但图像伪影显著改善（3.87±0.40 vs 3.33±0.86,P<0.001）,ACS-SS组检出病灶493/505枚,PI组检出病灶496/505枚,2组病灶显示能力无统计学差异（P>0.05）;ACS-SS组总扫描时间为1 320 s,较PI组4 760 s显著缩短。结论 ACS-SS-FLAIR技术用于头颅MR检查可显著缩短检查时间并减少图像伪影,有助于配合欠佳患者或儿童的颅脑MR检查。更多还原

【Abstract】 Objective To compare and research the cranial MR imaging characteristics of artificial intelligence compressed sensing single-shot fluid attenuated inversion recovery（ACS-SS-FLAIR） technology and parallel imaging fluid attenuated inversion recovery（PI-FLAIR） technology.Methods Forty cases of cranial MR T₂ FLAIR imaging were performed using two technical sequences.The ACS-SS group used single-shot fluid attenuated inversion recovery（SS-FLAIR） sequence,and the PI group used PI-FLAIR sequence.The image quality,lesion display,and the imaging time of the two technical methods were compared and analyzed.Results There was no significant difference in the overall image quality score between the ACS-SS group and the PI group（3.20±0.61 vs 3.47±0.78,P>0.05）.The signal-to-noise ratio（SNR） of the ACS-SS group was inferior to that of the PI group,but the image artifact was significantly improved（3.87±0.40 vs 3.33±0.86,P<0.001）.The lesions detected in the ACS-SS group and the PI group were 493/505and 496/505,respectively,showed no statistical difference in the lesion display ability between two groups（P>0.05）.The total scanning time of the ACS-SS group was 1 320 s,significantly shorter than that of the PI group（4 760 s）.Conclusion The ACS-SS-FLAIR technique in cranial MR can significantly shorten the examination time and reduce image artifacts,which is helpful for cranial MR in patients with poor cooperation or children.更多还原