基于超声影像组学的乳腺癌预测模型的诊断性能研究

目的评估并比较基于不同机器学习算法建立的乳腺癌超声影像组学预测模型的诊断性能。 方法回顾性收集2017年1月至2019年4月就诊皖南医学院第一附属医院、有明确病理结果的乳腺肿块病例828例，以2018年8月31日为节点将其分为训练集（526例）和验证集（302例），提取肿块的超声影像组学特征并进行特征筛选，运用k最近邻（kNN）、逻辑回归（LR）、朴素贝叶斯（NB）、随机森林（RF）和支持向量机（SVM）5种机器学习算法分别建立预测模型，使用重复交叉验证方法做内部验证，计算比较各模型的敏感度、特异度、阳性预测值（PPV）和阴性预测值（NPV），并实施外部验证，绘制ROC曲线并比较ROC曲线下面积（AUC）以评价模型的鉴别诊断性能，绘制校准曲线评价模型校准度。 结果从提取的109个影像组学特征中筛选出19个特征建立了5种机器学习算法模型。在内部验证中，5种模型的敏感度、特异度、PPV、NPV比较，总体差异均有统计学意义（P均＜0.001）。LR模型的特异度、PPV、NPV中位数分别为0.769、0.816、0.778，3项指标均高于其他4种模型；敏感度中位数为0.824，高于kNN、RF和SVM模型。此外，SVM模型的特异度、PPV、NPV中位数分别为0.706、0.774、0.759，虽均低于LR模型，但均高于其他3种模型。在外部验证中，LR、SVM、RF、kNN和NB的AUC依次为0.890、0.832、0.821、0.746和0.703，其中LR与SVM的AUC差异有统计学意义（P=0.012）；此外，各模型在校准性能上表现并不一致，LR和SVM模型的校准曲线显示乳腺癌实际概率与预测概率之间的一致性较好。 结论以超声影像组学特征为基础，运用不同机器学习算法建立的乳腺癌超声预测模型，均表现出较高的诊断性能，其中LR模型表现最为突出；选择合适的机器学习算法有助于进一步提高预测模型的诊断性能，提供更加准确的量化预测结果。     

护士"互联网+护理服务"参与行为意愿量表的编制及信效度检验

目的编制护士"互联网+护理服务"参与行为意愿量表,并进行信效度检验。方法以计划行为理论为框架,采用文献回顾、半结构式访谈法形成量表的条目池;通过专家咨询、项目分析对量表条目进行筛选,初步形成护士"互联网+护理服务"参与行为意愿量表。采用方便抽样法,于2020年10—12月选取江苏省6所医院(2所三级医院、2所二级医院、2所一级医院)的825名护士进行正式调查,检验问卷的信效度。结果护士"互联网+护理服务"参与行为意愿量表包括3个分量表(参与行为态度、主观行为规范、感知行为控制)、8个维度、17个条目。总量表的Cronbach'sα系数为0.927、重测信度系数为0.805。护士"互联网+护理服务"参与行为意愿与参与行为的相关系数为0.743(P<0.01)。因子分析提取出8个公因子,累积方差贡献率为72.33%。结论护士"互联网+护理服务"参与行为意愿量表具有较好的信效度,可用于护士"互联网+护理服务"参与行为意愿的评估。     

定量分析对常规超声联合超声造影定性误诊甲状腺结节的诊断价值

摘要：目的：探讨超声造影（CEUS）定量分析对常规超声联合CEUS定性诊断误诊为恶性的甲状腺结节与甲状腺乳头状癌的鉴别诊断价值。方法：选取2018年7月至2021年11月在我院超声科行甲状腺常规超声及CEUS检查，术前联合超声诊断为恶性（C-TIRADS≥4b 或 CEUS 表现为低增强和快速消退任一征像）而病理为良性的35位患者共35个结节为研究组；选择同期联合超声诊断与病理一致的甲状腺乳头状癌患者47例做为对照组。分析常规超声及CEUS特征，并将结节整体、中央部、边缘部及邻近正常甲状腺实质作为感兴趣区分别绘制时间-强度曲线（TIC）进行定量分析，比较两组结节各定量参数之间的差异。结果：研究组整体相对上升时间（△RT）以0.45为截断值，诊断恶性结节的敏感度为97.9%，特异度为85.7%，受试者工作（ROC）曲线下面积为0.96；研究组中央区△RT以0.55为截断值，诊断恶性结节的敏感度为95.7%，特异度为82.9%，ROC曲线下面积为0.87；研究组边缘区△RT以0.25为截断值，诊断恶性结节的敏感度为89.4%，特异度为77.1%，ROC曲线下面积为0.89。结论：CEUS定量分析在常规超声联合CEUS定性误诊甲状腺结节的诊断中具有一定的鉴别诊断价值。     

基于Triangle模型的2型糖尿病患者分层分级延续护理方案的构建

目的 构建2型糖尿病患者“医院-社区”一体化分层分级延续护理实践方案，为护理人员提供整体性、科学性、可操作性的分层分级延续护理实践范本。方法 以Triangle慢病分层管理模型、分级护理、生理-心理-社会医学模式为理论指导，通过文献分析初步拟定2型糖尿病患者“医院-社区”一体化分层分级延续护理实践方案，采用德尔菲法选取20名专家对初始方案进行论证，形成最终方案。结果 最终形成的方案包括实践基础、实践标准2大部分，其中实践基础包括组织架构、建立以医院为主导的糖尿病多学科团队、运行机制、开展团队同质化培训，实践标准包括适用人群、分层级别、分层评估标准、分级随访措施。2轮函询问卷回收率分别为95.2%、100%,专家权威系数分别为0.923、0.940,Kendall协调系数分别为0.205、0.243(均P<0.01)。结论 构建的2型糖尿病患者“医院-社区”一体化分层分级延续护理实践方案具有较好的可靠性、科学性及可行性，可为护理人员实施分层分级延续护理提供参考。     

半夏泻心汤对DSS诱导的溃疡性结肠炎小鼠肠道微生态及Th17/Treg细胞平衡的影响＊

目的 观察半夏泻心汤对饮用葡聚糖硫酸钠（DSS）诱导建立的溃疡性结肠炎（UC）模型小鼠Th17/Treg细胞平衡及肠道微生态的影响，从免疫失衡及菌群紊乱角度分析半夏泻心汤治疗UC的作用机制。方法 将40只C57BL/6J雄性小鼠随机分为空白组、模型组、阳性药（美沙拉嗪）组及半夏泻心汤组，除空白组外的各小鼠采用2.5% DSS水溶液自由饮用7天诱导建立UC小鼠模型。自实验第8日起，分别采用灭菌水、美沙拉嗪水溶液、半夏泻心汤水溶液给上述各组别小鼠灌胃。采用HE染色观察结肠组织病理学变化，ELISA方法检测组织炎症因子含量，流式细胞术检测Th17及Treg细胞的相对数量，16S rRNA测序技术进一步检测模型小鼠粪便中菌群的变化。结果 与空白组比较，模型组小鼠结肠损伤严重，血清炎症因子含量显著升高（P<0.01）；与模型组比较，中药干预组小鼠结肠损伤减轻，血清TNF-α、IFN-γ含量显著下降（P<0.05）。较空白组小鼠相比，模型组小鼠Th17细胞百分比显著升高（P<0.01），而Treg细胞百分比显著降低（P<0.01）。经半夏泻心汤干预后，其小鼠Th17细胞数量均显著降低（P<0.05），Treg细胞数量则显著升高（P<0.05）。同时，半夏泻心汤可升高门水平上Firmicutes、属水平上Lachnospiraceae_NK4A136_group、unclassified_f_Lachnospiraceae、Prevotellaceae_UCG-001、Alistipes；降低门水平上Actinobacteriota，科水平上Bifidobacteriaceae、Atopobiaceae、Clostridiaceae，属水平上Clostridium_sensu_stricto_1、Coriobacteriaceae_UCG-002、Bifidobacterium的相对丰度。结论 半夏泻心汤可使Treg细胞相对数量升高、Th17细胞相对数量降低，以维持Th17/Treg细胞动态平衡；同时可降低有害菌Clostridium_sensu_stricto_1、Coriobacteriaceae_UCG-002的相对丰度，升高产短链脂肪酸类菌Lachnospiraceae_NK4A136_group、Prevotellaceae_UCG-001、unclassified_f_Lachnospiraceae等菌属的相对丰度，调控肠道微生物结构及动态平衡，从而发挥治疗UC的作用。     

Arterial spin labeling MR image denoising and reconstruction using unsupervised deep learning

Arterial spin labeling (ASL) imaging is a powerful magnetic resonance imaging technique that allows to quantitatively measure blood perfusion non-invasively, which has great potential for assessing tissue viability in various clinical settings. However, the clinical applications of ASL are currently limited by its low signal-to-noise ratio (SNR), limited spatial resolution, and long imaging time. In this work, we propose an unsupervised deep learning-based image denoising and reconstruction framework to improve the SNR and accelerate the imaging speed of high resolution ASL imaging. The unique feature of the proposed framework is that it does not require any prior training pairs but only the subject's own anatomical prior, such as T1-weighted images, as network input. The neural network was trained from scratch in the denoising or reconstruction process, with noisy images or sparely sampled k-space data as training labels. Performance of the proposed method was evaluated using in vivo experiment data obtained from 3 healthy subjects on a 3T MR scanner, using ASL images acquired with 44-min acquisition time as the ground truth. Both qualitative and quantitative analyses demonstrate the superior performance of the proposed txtc framework over the reference methods. In summary, our proposed unsupervised deep learning-based denoising and reconstruction framework can improve the image quality and accelerate the imaging speed of ASL imaging.     

Analysis of the correlation between P27 expression and Helicobacter pylori infection in gastric cancer

We aimed to explore the correlation between P27 expression and Helicobacter pylori (H. pylori) infection in gastric cancer, so as to provide evidence for understanding the pathogenesis of gastric cancer caused by H. pylori infection. A total of 82 samples of gastric cancer tissues and 56 samples of tumor-adjacent normal tissues collected from the gastrectomy were enrolled in this study. Then, 14C-urease breathing test was carried out to evaluate the infection of H. pylori in gastric cancer tissues, the expression of P27 in the tissue samples was detected by the immunohistochemistry staining, and the correlation between the H. pylori infection and P27 expression in gastric cancer was analyzed. Of 82 gastric cancer patients, there were 53 patients with H. pylori infection (64.63%). Among the patients with highly or moderately differentiated gastric cancer, the expression of P27 was much higher than that of patients with poorly differentiated gastric cancer (p < 0.01). Besides, comparison of the P27 expression between males and females, among different age groups, tumor sizes, TNM stages, tumor infiltration degrees, or lymph node metastasis, showed no significant differences (p > 0.05). Analysis of the correlation revealed that P27 expression was negatively correlated with the infection of H. pylori (p < 0.01). Multifactorial logistics regression analysis indicated that tumor differentiation was a risk factor of P27-positive expression in gastric cancer tissues (p < 0.01). In addition, P27 expression in the gastric cancer tissues was lower than that in the tumor-adjacent normal tissues (p < 0.01). In gastric cancer patients, expression of P27 is correlated with H. pylori infection which, via downregulating P27, can cause the cancerization of gastric mucosa, and P27, for its role in the development and progression of gastric cancer, is a potential auxiliary indicator for clinical diagnosis whether gastric cancer is complicated with H. pylori infection. So, P27 is a key indicator for diagnosis, treatment, and prognostic evaluation of disease in the advanced stage.