No abstract

Background In the recent years, artificial intelligence (AI) algorithms have been used to accurately diagnose musculoskeletal diseases. However, it is not known whether the particular regions of interest (ROI) delineation method would affect the performance of the AI algorithm. Purpose The purpose of this study was to investigate the influence of ROI delineation methods on model performance and observer consistency. Methods In this retrospective analysis, ultrasound (US) measures of median nerves affected with carpal tunnel syndrome (CTS) were compared to median nerves in a control group without CTS. Two methods were used for delineation of the ROI: (1) the ROI along the hyperechoic medial edge of the median nerve but not including the epineurium (MN) (ROI1); and (2) the ROI including the hyperechoic epineurium (ROI2), respectively. The intra group correlation coefficient (ICC) was used to compare the observer consistency of ROI features (i.e. the corresponding radiomics parameters). Parameters α1 and α2 were obtained based on the ICC of ROI1 features and ROI2 features. The ROC analysis was used to determine the area under the curve (AUC) and evaluate the performance of the radiologists and network. In addition, four indices, namely sensitivity, specificity, positive prediction and negative prediction were analyzed too. Results A total of 136 wrists of 77 CTS group and 136 wrists of 74 control group were included in the study. Control group was matched to CTS group according to the age and sex. The observer consistency of ROI features delineated by the two schemes was different, and the consistency of ROI1 features was higher (α1 ˃ α2). The intra‐observer consistency was higher than the inter‐observer consistency regardless of the scheme, and the intra‐observer consistency was higher when chose scheme one. The performances of models based on the two ROI features were different, although the AUC of each model was greater than 0.8.The model performed better when the MN epineurium was included in the ROI. Among five artificial intelligence algorithms, the Forest models (model1 achieved an AUC of 0.921 in training datasets and 0.830 in testing datasets; model2 achieved an AUC of 0.967 in training datasets and 0.872 in testing datasets.) obtained the highest performance, followed by the support vector machine (SVM) models and the Logistic models. The performances of the models were significantly better than the inexperienced radiologist (Dr. B. Z. achieved an AUC of 0.702). Conclusion Different ROI delineation methods may affect the performance of the model and the consistency of observers. Model performance was better when the ROI contained the MN epineurium, and observer consistency was higher when the ROI was delineated along the hyperechoic medial border of the MN.

BACKGROUND Drug-eluting beads transarterial chemoem-bolization (DEB-TACE) has the advantages of slow and steady release, high local concentration, and low incidence of adverse drug reactions compared to the traditional TACE. DEB-TACE combined with sequentially ultrasound-guided radiofrequency ablation (RFA) therapy has strong anti-cancer effects and little side effects, but there are fewer related long-term studies until now. AIM To explore the outcome of DEB-TACE sequentially combined with RFA for patients with primary hepatocellular carcinoma (HCC). METHODS Seventy-six patients with primary HCC who underwent DEB-TACE sequentially combined with RFA were recruited. Forty patients with untreated HCC were included in Group A, and 36 patients with recurrent HCC were included in Group B. In addition, 40 patients with untreated HCC who were treated with hepatectomy were included in Group C. The serological examination, preoperative magnetic resonance imaging examination, and post-treatment computed tomography enhanced examination were performed for all patients. The efficacy was graded as complete remission (CR), partial remission (PR), stable disease and progressive disease at the 3 rd , 6 th , and 9 th . All patients were followed up for 3 years and their overall survival (OS), disease-free survival (DFS) were assessed. RESULTS The efficacy of Group A and Group C was similar ( P > 0.05), but the alanine aminotransferase, aspartate aminotransferase and total bilirubin of Group A were lower than those of Group C (all P < 0.05). The proportions of CR (32.5%), PR (37.5%) were slightly higher than Group A (CR: 27.5%, PR: 35%), but the difference was not statistically significant ( χ 2 = 0.701, P = 0.873). No operational-related deaths occurred in Group A and Group C. The OS (97.5%, 84.7%, and 66.1%) and the DFS (75.0%, 51.7%, and 35.4%) of Group A at the 1 st , 2 nd , and 3 rd year after treatment were similar with those of Group C (OS: 90.0%, 79.7%, and 63.8%; DFS: 80.0%, 59.7%, and 48.6%; P > 0.05). The OS rates in Group A and Group B (90%, 82.3%, and 66.4%) were similar ( P > 0.05). The DFS rates in Group B (50%, 31.6%, and 17.2%) were lower than that of Group A ( P = 0.013). CONCLUSION The efficacy of DEA-TACE combined with RFA for untreated HCC is similar with hepatectomy. Patients with recurrent HCC could get a longer survival time through the combined treatment.

Abstract. Contrast-enhanced ultrasound (CEUS) and magnetic resonance imaging (MRI) were compared in evaluating the short-term effects of microwave ablation (MWA) on uterine fibroids. A total of 60 patients with uterine fibroids treated by MWA were enrolled in the experimental group during their two-year follow-up period according to the inclusion criteria. Conventional two-dimensional US, MRI and CEUS were performed to determine the volume reduction and the fibroid residue by displaying the size, echo and signal intensity of fibroids prior to and after MWA treatment. As the control group, 60 consecutive patients were recruited on their follow-up visit at least two years after MWA treatment of uterine fibroids. Significant differences were observed in the wash-in rate (WiR) of the fibroid tissue, start time difference, rise time ratio (RTR) and WiR ratio between the experimental and control groups (P<0.05). However, the WiR of fibroid vessel, total area under the curve of fibroid vessel and tissue, and rise time difference (RTD) between fibroid vessel and tissue did not display any significant differences between the two groups. Fibroids were either reduced in volume or cured by MWA therapy in patients with uterine fibroids. The reductions in volume of hypointense, isointense and hyperintense fibroids were 62.42±18.13, 53.27±10.05 and 47.43±9.56%, respectively, on T1-weighted imaging (T1WI). On T2WI, the corresponding reductions were 67.32±32.63, 59.36±19.36 and 42.63±10.37%, respectively. The higher the signal intensity on T1WI and T2WI, the lower the reduction in volume. It is indicative that different blood supply to fibroids results in different ablation. CEUS was proved to be more effective than MRI in evaluating the effects of MWA on uterine fibroids during the first postoperative year.

AIMTo explore the relationship of liver and spleen shear wave velocity in patients with liver cirrhosis combined with portal hypertension, and assess the value of liver and spleen shear wave velocity in predicting the prognosis of patients with portal hypertension.METHODSAll 67 patients with liver cirrhosis diagnosed as portal hypertension by hepatic venous pressure gradient in our hospital from June 2014 to December 2014 were enrolled into this study. The baseline information of these patients was recorded. Furthermore, 67 patients were followed-up at 20 mo after treatment, and liver and spleen shear wave velocity were measured by acoustic radiation force impulse at the 1st week, 3rd month and 9th month after treatment. Patients with favorable prognosis were assigned into the favorable prognosis group, while patients with unfavorable prognosis were assigned into the unfavorable prognosis group. The variation and difference in liver and spleen shear wave velocity in these two groups were analyzed by repeated measurement analysis of variance. Meanwhile, in order to evaluate the effect of liver and spleen shear wave velocity on the prognosis of patients with portal hypertension, Cox’s proportional hazard regression model analysis was applied. The ability of those factors in predicting the prognosis of patients with portal hypertension was calculated through receiver operating characteristic (ROC) curves.RESULTSThe liver and spleen shear wave velocity in the favorable prognosis group revealed a clear decline, while those in the unfavorable prognosis group revealed an increasing tendency at different time points. Furthermore, liver and spleen shear wave velocity was higher in the unfavorable prognosis group, compared with the favorable prognosis group; the differences were statistically significant (P < 0.05). The prognosis of patients with portal hypertension was significantly affected by spleen hardness at the 3rd month after treatment [relative risk (RR) = 3.481]. At the 9th month after treatment, the prognosis was affected by liver hardness (RR = 5.241) and spleen hardness (RR = 7.829). The differences between these two groups were statistically significant (P < 0.05). The ROC analysis revealed that the area under the curve (AUC) of spleen hardness at the 3rd month after treatment was 0.644, while the AUCs of liver and spleen hardness at the 9th month were 0.579 and 0.776, respectively. These might predict the prognosis of patients with portal hypertension.CONCLUSIONSpleen hardness at the 3rd month and liver and spleen shear wave velocity at the 9th month may be used to assess the prognosis of patients with portal hypertension. This is hoped to be used as an indicator of predicting the prognosis of patients with portal hypertension.