In a case study of Phoenix Healthcare Group' s role in Beijing Mentougou District Hospital's reform,this paper systematically reviewed the PPP application in public medical service,revealing the problems found in the reform including concept deviations for public benefit,and sustainability of the business pattern.On such basis,the authors recommended from the perspectives of the government,hospital and society.Their recommendations cover policies on improving the supporting policies,reconstruction of the hospital mechanisms of motivation,income distribution and supervision,fostering efficient private medical enterprises and the third-party appraisers.These recommendations aim at improving PPP model in the reform of public hospitals in China.

Objective:To evaluate the value of MAGiC STIR sequence inassessing bone marrow edema (BME) of the sacroiliac joint (SIJ) in ankylosing spondylitis (AS) patients.Methods:Twenty two AS patients from the Affiliated Fifth Hospital of Sun Yat-sen University from September 2019 to January 2020 underwent MR scanning, consisting of T 2WI fat-suppressed (FS) sequence and MAGiC sequence. Each SIJ was divided into four quadrants for analysis on oblique coronal MR image. Two radiologists [middle-higher seniority levels (radiologist 1 and radiologist 2) reviewed image sets of two sequences [T 2WI FS and MAGiC short T 1 inversion recovery (STIR) sequence] independently at separate times, using Hermann-Braun scoring system for degree of BME and summing numbers of quadrants based on BME findings. Disagreements were resolved to reach an agreement for final results.Assessments were repeated two weeks later by radiologist 2. Intra-class correlation coefficients (ICC) were calculated to assess the inter-and intra-observer agreement for the degree of BME and numbers of positive BME quadrants in the MAGiC STIR sequence, respectively. The differences between the two sequences for the degree of BME and numbers of positive BME quadrants were analyzed with Wilcoxon signed-rank test and Chi-square test, respectively. Taking the results of T 2WI FS sequence as a reference standard, the ROC curve was drawn, sensitivity, specificity, positive predictive value and negative predictive value in the detection of BME were calculated for the MAGiC STIR sequence. Results:Nineteen AS patients including 38 sacroiliac joints were enrolled in our study, being divided into 152 quadrants. Inter-and intra-observer agreement for the degree of BME and inter-observer agreement for numbers of positive BME quadrants in the MAGiC STIR sequence were all good (ICC=0.936, 95%CI 0.912-0.953, P<0.001; ICC=0.910, 95%CI 0.878-0.934, P<0.001; ICC=0.876, 95%CI 0.833-0.909, P<0.001). The difference between the two sequences for the degree of BME in quadrants was statistically significant ( Z=-3.132, P=0.002). But there was no statistically significant difference between sequences for numbers of quadrants in detecting BME (χ2=0.244, P=0.622). Area under the ROC curve of MAGiC STIR sequence in diagnosis of BME was 0.920 ( P<0.001). Sensitivity, specificity, positive predictive value and negative predictive value for the MAGiC STIR sequence in the detection of BME were 82.0%, 95.1%, 89.1% and 91.5%, respectively. Conclusion:MAGiC STIR sequence can be used to evaluate bone marrow edema of SIJ in AS patients with high sensitivity and specificity.

OBJECTIVE: To develop and validate radiomics models based on non-enhanced magnetic resonance (MR) imaging for differentiating chondrosarcoma from enchondroma. METHODS: We retrospectively evaluated a total of 68 patients (including 27 with chondrosarcoma and 41 with enchondroma), who were randomly divided into training group (=46) and validation group (=22). Radiomics features were extracted from TWI and TWI-FS sequences of the whole tumor by two radiologists independently and selected by Low Variance, Univariate feature selection, and least absolute shrinkage and selection operator (LASSO). Radiomics models were constructed by multivariate logistic regression analysis based on the features from TWI and TWI-FS sequences. The receiver-operating characteristics (ROC) curve and intraclass correlation coefficient (ICC) analyses of the radiomics models and conventional MR imaging were performed to determine their diagnostic accuracy. RESULTS: The ICC value for interreader agreement of the radiomics features ranged from 0.779 to 0.923, which indicated good agreement. Ten and 11 features were selected from the TWI and TWI-FS sequences to construct radiomics models, respectively. The areas under the curve (AUCs) of TWI and TWI-FS models were 0.990 and 0.925 in training group and 0.915 and 0.855 in the validation group, respectively, showing no significant differences between the two sequence-based models (>0.05). In all the cases, the AUCs of the two radiomics models based on TWI and TWI-FS sequences and conventional MR imaging were 0.955, 0.901 and 0.569, respectively, demonstrating a significantly higher diagnostic accuracy of the two sequence-based radiomics models than conventional MR imaging (<0.01). CONCLUSIONS The radiomics models based on TWI and TWI-FS non-enhanced MR imaging can be used for the differentiation of chondrosarcoma from enchondroma.
Chondroma ; Chondrosarcoma ; Humans ; Magnetic Resonance Imaging ; ROC Curve ; Retrospective Studies

ObjectiveTo analyze the value of grey-scale reversed T₁-weighted （rT₁） MRI in the detection of structural lesions of the sacroiliac joint （SIJ） in patients with axial spondyloarthritis （ax-SpA）. MethodsFifty-two ax-SpA patients who underwent both MRI and CT in our hospital within a week from February 2020 to December 2022 were retrospectively included. Both sacral and iliac side of each SIJ on oblique coronal images were divided into anterior， middle and posterior portion. Two radiologists reviewed independently three groups of MRI including T1-weighted imaging （T₁WI）， rT₁ and T₁WI + rT₁ images to evaluate the structural lesions like erosions， sclerosis and joint space changes in each of the 6 regions of the SIJ. One of the radiologist did the evaluation again one month later. CT images were scored for lesions by a third radiologist and served as the reference standard. Intra-class correlation coefficients （ICC） were calculated to test the inter- and intra-reader agreement for the assessment of SIJ lesions. A Friedman test was performed to compare the lesion results of MRI and CT image findings. We examined the diagnostic performance ［accuracy， sensitivity （SE） and specificity］ of different groups of MRI in the detection of lesions by using diagnostic test. A McNemar test was used to compare the differences of three groups of MRI findings. ResultsCT showed erosions in 71 joints， sclerosis in 65 and joint space changes in 53. Good inter-and intra-reader agreements were found in three groups of MRI images for the assessment of lesions， with the best agreement in T₁WI + rT₁. There were no difference between T₁WI + rT₁ and CT for the assessment of all lesions， nor between rT₁ and CT for the assessment of erosions and joint space changes （P>0.05）. T₁WI + rT₁ yielded better accuracy and SE than T₁WI in detection of all lesions （Accuracy _erosions： 90.3% vs 76. 9%； SE _erosions： 91.6% vs 76.1%； Accuracy _sclerosis： 89.4% vs 80.8%； SE _sclerosis： 84.6% vs 73.9%； Accuracy _{joint space changes}： 86.5% vs 73.1%； SE _{joint space changes}： 84.9% vs 60.4%； P<0.05）. rT₁ yielded better accuracy and SE than T₁WI in detection of erosions and joint space changes （Accuracy _erosions： 87.5% vs 76.9%； SE _erosions： 88.7% vs 76.1%； Accuracy _{joint space changes}： 85.6% vs 73.1%； SE _{joint space changes}： 83.0% vs 60.4%； P<0.05）. ConclusionsIn the detection of SIJ structural lesions in ax-SpA， rT₁ improves the diagnostic performance and T₁WI + rT₁ is more superior to others.