Objective To compare the efficacy and safety of lauromacrogol injection sclerotherapy with ethanol injection sclerotherapy in treating simple liver cysts. Methods A total of 166 patients with simple liver cyst were randomly divided into the lauromacrogol group (study group, n=86) and the absolute alcohol group (control group, n=80). Under ultrasonographic guidance, puncture aspiration of liver cyst was carried out in all patients, which was followed by injection of lauromacrogol for patients in the study group or injection of ethanol for patients in the control group. The therapeutic effect and the side-effect were evaluated. The results were compared between the two groups. Results No serious complications such as bleeding or infection occurred in both groups. During the therapeutic course , 45 patients (56.3%) in the control group felt pain to some degree and 23 patients (28.8%) developed drunk-like symptoms, while no patient in the study group felt any obvious discomfort. One week after sclerotherapy , 20 patients (25%) in the control group complained of distending pain on the right upper abdomen, while only 9 patients (10.5%) in the study group complained of pain, and the difference was statistically significant (χ2= 6.073, P < 0.05). Six months after the treatment, the cure rate of the study group and the control group was 95.7%and 93.5%respectively, and the difference between the two groups was no significant (P > 0.05). Conclusion For the treatment of liver cysts, lauromacrogol injection is safe and effective. Therefore, this technique should be recommended in clinical practice.

Objective To construct and validate a nomogram model for predicting microvascular invasion (MVI) based on the characteristics on contrast-enhanced ultrasound (CEUS) Liver Imaging Reporting and Data System (LI-RADS) in patients with hepatocellular carcinoma (HCC). Methods A total of 262 patients with HCC who were diagnosed in Wujin Hospital Affiliated to Jiangsu University from January 2017 to July 2020 were enrolled and randomly divided into modeling group and validation group at a ratio of 1∶ 1, with 131 patients in each group. MVI was confirmed by postoperative microscopic pathological results, and there were 70 patients with MVI in the modeling group and 56 patients with MVI in the validation group. CEUS was used to evaluate LI-RADS characteristics for the two groups. The independent samples t -test was used for comparison of continuous data between the two groups, and the chi-square test was used for comparison of categorical data between the two groups. Univariate and multivariate Logistic regression analyses were used to identify the risk factors for MVI in the modeling group; the receiver operating characteristic (ROC) curve was plotted, and the area under the ROC curve (AUC) was calculated for the model in predicting MVI to evaluate the accuracy of prediction; a decision curve analysis was used to evaluate the consistency of the model, and dispersion was compared between the calibration curve and the standard curve for the model in predicting MVI. Results There were no significant differences in clinical data and CEUS findings between the modeling group and the validation group (all P > 0.05). The univariate analysis showed that compared with the MVI-negative patients, the MVI-positive patients had significant increases in serum alpha fetoprotein (AFP) level, tumor diameter, and LR-5 "late and mild washout" and LR-M "early washout" on LI-RADS, as well as a significantly higher LI-RADS grade (all P < 0.05). The multivariate analysis showed that AFP 20-400 ng/mL (odds ratio [ OR ]=2.65, P < 0.001), AFP≥400 ng/mL ( OR =3.98, P < 0.001), tumor diameter ≥30 mm ( OR =2.12, P < 0.001), and LR-M on CEUS ( OR =3.24, P < 0.001) were independent risk factors for MVI. The ROC curve analysis showed that the nomogram had an AUC of 0.867 and 0.821 in predicting MVI in the modeling group and the validation group, respectively. The nomogram model had a C-Index of 0.765 (95% confidence interval: 0.701-0.834). The calibration curves of the nomogram model were close to the standard curve in both groups. Conclusion The nomogram model based on LI-RADS obtained by CEUS in combination with AFP and tumor diameter has a good application value and can guide the preoperative screening for patients at a high risk of MVI and the development of appropriate surgical plans in clinical practice.

Objective: To study the correlation between the level of serum Chitinase-3-like protein 1 (CHI3L1) and the significant liver fibrosis and liver cirrhosis in patients with chronic liver disease, and to evaluate its diagnostic value. Methods: 165 patients with chronic liver disease were selected, liver histopathological examination was performed to detect serum CHI3L1 concentration. Four indexes of hepatic fibrosis (type III procollagen, collagen IV, laminin, hyaluronic acid), aspartate aminotransferase/platelet ratio index (APRI) and FIB-4 (fibrosis- 4) scores were based on the pathological findings of liver biopsy and compared the advantages and disadvantages of serum CHI3L1 with other methods for the diagnosis of hepatic fibrosis and liver cirrhosis. A multivariate regression analysis model was created, and receiver operating characteristic curve was analyzed. Results: The level of serum CHI3L1 increased with increase of fibrosis stage and was highest in liver cirrhosis stage. In the period of S0 to 1, the levels of S2 to 3 and S4 were 62.82 (41.40 ~ 87.20), 70.94 (48.47 to 122.60) and 141.06 (78.18 ~ 197.40), and there were statistically significant differences between the groups (P < 0.001). The area under the curve for the diagnosis of significant liver fibrosis was 0.68 (0.60 to 0.77), and 0.74 (0.65 to 0.83) for cirrhosis in CHI3L1. Multivariate regression analysis showed that CHI3L1 was an independent predictor of significant fibrosis and cirrhosis. The combined diagnostic model based on CHI3L1, collagen IV and FIB-4 scores further improved the diagnostic value. The area under the curve for the diagnosis of significant fibrosis and cirrhosis was 0.79 (0.72 to 0.86) and 0.80 (0.73 to 0.87), respectively. Conclusion CHI3L1 has a good diagnostic value in patients with chronic liver disease with significant fibrosis and liver cirrhosis. The diagnostic model in combination with other markers like Collagen IV and FIB-4 scores could further improve the diagnostic value and is worthy of further study.

Recent advancement in natural language processing (NLP) and medical imaging empowers the wide applicability of deep learning models. These developments have increased not only data understanding, but also knowledge of state-of-the-art architectures and their real-world potentials. Medical imaging researchers have recognized the limitations of only targeting images, as well as the importance of integrating multimodal inputs into medical image analysis. The lack of comprehensive surveys of the current literature, however, impedes the progress of this domain. Existing research perspectives, as well as the architectures, tasks, datasets, and performance measures examined in the present literature, are reviewed in this work, and we also provide a brief description of possible future directions in the field, aiming to provide researchers and healthcare professionals with a detailed summary of existing academic research and to provide rational insights to facilitate future research.
Humans ; Natural Language Processing ; Surveys and Questionnaires