人工智能压缩感知技术在不同线圈下单髋关节MRI中的应用价值OA

Objective To explore the application value of artificial intelligence compressed sensing(ACS)technology in single hip joint magnetic resonance imaging(MRI)examinations under different coil conditions.Methods Thirty healthy volunteers were prospectively recruited.Multi-sequence scans of the single hip joint were performed using three protocols:a large ultra-soft coil with parallel imaging(PI)technology,a large ultra-soft coil with ACS technology,and an ultra-soft body coil with ACS technology.For the PI group,ACS group,and body coil ACS group,the signal intensity of anatomical structures(including the femoral head and gluteus medius)and background noise were measured,and the signal to noise ratio(SNR)and contrast to noise ratio(CNR)were calculated.Two radiologists conducted subjective scoring of the images in terms of overall image quality,tissue structure visualization,fat suppression effect,and image artifacts.The Friedman rank-sum test was used for statistical analysis of the subjective and objective evaluation results of the three groups.Results There was no statistically significant difference in the subjective image quality scores between the ACS group and the PI group(P>0.05).In the body coil ACS group,the overall image scores of the sagittal PDWI sequence and coronal PDWI sequence showed statistically significant differences compared with those of the PI group and ACS group(P<0.05).In terms of artifact scores,the body coil ACS group scored lower than the ACS group in the axial,sagittal,and coronal PDWI sequences,with statistically significant differences(P<0.05).In the objective data,except for the SNR of muscle in the axial PDWI sequence,the other SNR and CNR values of the ACS group were higher than those of the PI group and body coil ACS group.For the ACS group,there was no statistically significant difference in the muscle SNR and CNR of the coronal PDWI and coronal T1WI sequences compared with the PI group(P>0.05),while statistically significant differences were observed in other sequences(P<0.05).Additionally,there was no statistically significant difference between the ACS group and the body coil ACS group in the femoral head SNR,muscle SNR,and CNR of the axial PDWI sequence,as well as the femoral head SNR of the coronal PDWI sequence(P>0.05),but statistically significant differences were found in other sequences(P<0.05).The overall scanning time of the ACS group was 45.3%shorter than that of the PI group.Conclusion ACS technology offers good image quality and significantly shortened scanning time in single hip joint MRI.In some sequences,the image quality of ACS technology with an ultra-soft body coil is comparable to that with a conventional coil,so it can be used when the patient's condition is not suitable for a large ultra-soft coil.