Objective To observe the value of whole-tumor apparent diffusion coefficient(ADC)histogram for identifying histopathological grade of alveolar soft part sarcoma(ASPS).Methods Forty-three ASPS patients,including 27 cases of high-grade ASPS(high-grade group)and 16 cases of low-grade ASPS(low-grade group)were retrospectively enrolled.Patients'survival data were collected,MRI manifestations of ASPS were recorded,and the whole-tumor ADC histogram parameters were obtained and compared between groups.The correlations of whole-tumor ADC histogram parameters being different between groups with tumors'histological grading were analyzed,and the efficacy of whole-tumor ADC histogram parameters for identifying high-grade and low-grade ASPS were assessed.Results The 5-year survival rate of ASPS patients in low-grade group was 82.05％,higher than that(51.28％)in high-grade group(P＜0.05).The percentage of distant metastasis,tumor≥5 cm,as well as of tumors with features such as peritumoral edema and intra-tumoral septum in high-grade group were all higher than in low-grade group(all P＜0.05).The 5th,10th,25th,50th,75th,90th and the mean values of ADC in high-grade group were all lower than those in low-grade group(all P＜0.05)and negatively correlated with pathohistological grade of ASPS(from-0.547 to-0.385,all P＜0.05).The aeras under the receiver operating characteristic curves of the above parameters ranged from 0.734 to 0.822,which were fairly good for identifying high-grade and low-grade ASPS.Conclusion Whole-tumor ADC histogram parameters could be used to effectively identify high-grade and low-grade ASPS.

Objective:To explore the value of contrast enhancement T1 fluid-attenuated inversion recovery sequence (CE-T1FLAIR) based on modulated flip angle technique in refocused imaging with extended echo train (MATRIX) in detecting metastases.Methods:One hundred and seventy-six patients with pathologically diagnosed malignant tumors and brain metastases accepted enhanced 3.0T MRI scan in Department of Medical Imaging, He'nan Provincial Cancer Hospital from October 2023 to February 2024 were enrolled. Lianying's intelligent brain metastasis AI-assisted detection system and sequences of MATRIX CE-T1FLAIR, 3D GRE_fsp CE-T1FLAIR and FSE CE-T1FLAIR were used to detect the brain metastasis lesions, respectively. Length of the lesions was measured according to Lianying's intelligent brain metastasis AI-assisted detection system, and all lesions were divided into 3 categories: <3 mm, 3-10 mm, and >10 mm. Differences in detection rate in brain metastases of different lengths and locations among the 3 sequences were compared.Results:Detection rates of MATRIX CE-T1FLAIR, 3D GRE_fsp CE-T1FLAIR, and FSE CE-T1FLAIR in brain metastases were 99.67%, 90.52%, and 71.02%, which were decreased successively, with significant differences ( P<0.05). Detection rates of MATRIX CE-T1FLAIR, 3D GRE_fsp CE-T1FLAIR and FSE CE-T1FLAIR in brain metastases with length<3 mm (99.24%, 79.95% and 46.45%) or length of 3-10 mm (100%, 98.19% and 87.53%) were decreased successively, with significant differences ( P<0.05). Detection rates of MATRIX CE-T1FLAIR (100%, 80.56% and 64.24%), 3D GRE_fsp CE-T1FLAIR (100%, 97.25% and 76.11%), and FSE CE-T1FLAIR (100%, 91.18% and 70.59%) in metastases at the superficial area of the brain convexity, gray-white matter junction area, and cerebellum were decreased successively, with significant differences ( P<0.05). Detection rates of FSE CE-T1FLAIR in brain metastases in the basal ganglia and brainstem (69.33% and 50%) were significantly lower than those of MATRIX CE-T1FLAIR and 3D GRE_fsp CE-T1FLAIR (97.33% and 92.86%; 88% and 78.57%, P<0.05). Conclusion:MATRIX CE-T1FLAIR sequence is better than 3D GRE_fsp CE-T1FLAIR and FSE CE-T1FLAIR sequences in detecting brain metastases, especially for metastases with length<10 mm and metastases located at the superficial area of the brain convexity, gray-white matter junction area and cerebellum.