Deep learning applied to two-dimensional color Doppler flow imaging ultrasound images significantly improves diagnostic performance in the classification of breast masses: a multicenter study

BackgroundThe current deep learning diagnosis of breast masses is mainly reflected by the diagnosis of benign and malignant lesions. In China, breast masses are divided into four categories according to the treatment method: inflammatory masses, adenosis, benign tumors, and malignant tumors. These categorizations are important for guiding clinical treatment. In this study, we aimed to develop a convolutional neural network (CNN) for classification of these four breast mass types using ultrasound (US) images.MethodsTaking breast biopsy or pathological examinations as the reference standard, CNNs were used to establish models for the four-way classification of 3623 breast cancer patients from 13 centers. The patients were randomly divided into training and test groups (n = 1810 vs. n = 1813). Separate models were created for two-dimensional (2D) images only, 2D and color Doppler flow imaging (2D-CDFI), and 2D-CDFI and pulsed wave Doppler (2D-CDFI-PW) images. The performance of these three models was compared using sensitivity, specificity, area under receiver operating characteristic curve (AUC), positive (PPV) and negative predictive values (NPV), positive (LR+) and negative likelihood ratios (LR−), and the performance of the 2D model was further compared between masses of different sizes with above statistical indicators, between images from different hospitals with AUC, and with the performance of 37 radiologists.ResultsThe accuracies of the 2D, 2D-CDFI, and 2D-CDFI-PW models on the test set were 87.9%, 89.2%, and 88.7%, respectively. The AUCs for classification of benign tumors, malignant tumors, inflammatory masses, and adenosis were 0.90, 0.91, 0.90, and 0.89, respectively (95% confidence intervals [CIs], 0.87–0.91, 0.89–0.92, 0.87–0.91, and 0.86–0.90). The 2D-CDFI model showed better accuracy (89.2%) on the test set than the 2D (87.9%) and 2D-CDFI-PW (88.7%) models. The 2D model showed accuracy of 81.7% on breast masses ≤1 cm and 82.3% on breast masses >1 cm; there was a significant difference between the two groups (P < 0.001). The accuracy of the CNN classifications for the test set (89.2%) was significantly higher than that of all the radiologists (30%).ConclusionsThe CNN may have high accuracy for classification of US images of breast masses and perform significantly better than human radiologists.Trial registrationChictr.org, ChiCTR1900021375; http://www.chictr.org.cn/showproj.aspx?proj=33139.     

临床医学与护理学专业学生对死亡态度的调查分析

目的调查临床医学与护理学专业学生对死亡的态度,希望能有助于推动不同专业生死教育课程设计与发展。方法使用死亡态度描绘修订版量表对某西医院校临床医学专业三年级学生119名、某中医院校护理专业三年级学生116名进行调查。结果学生中有90%参加过葬礼,且过半学生阅读过死亡方面相关书籍。两个专业学生对死亡的态度均以自然接受为主。护理专业学生在死亡恐惧与死亡逃避2个维度上得分均高于临床医学生,差异有统计学意义(P0.05)。结论相较于临床医学专业学生,护理学专业学生对死亡的态度更倾向于"恐惧与逃避"。建议开展具有针对性的生死观教育,以帮助学生更加深入了解生命与死亡。     

多重荧光PCR-熔解曲线法同时检测常见病原真菌方法学的建立及诊断价值

目的建立多重荧光PCR-熔解曲线法，并评价其同时检测假丝酵母菌、曲霉菌及新型隐球菌的诊断价值。 方法建立多重荧光PCR-熔解曲线法并对其检测体系进行优化。收集临床高度怀疑为侵袭性真菌感染（IFI）的各类临床样本179例，其中血液65例、深部痰液35例、尿液30例、脑脊液18例、胸腹水12例、肺泡灌洗液10例、新鲜肺组织9例。通过敏感性、特异性、重复性实验验证多重荧光PCR-熔解曲线法的检测性能，并应用受试者工作特征（ROC）曲线评价该方法的诊断效能。 结果建立的多重荧光PCR-熔解曲线法可同时检测8种常见病原真菌，包括4种假丝酵母菌（白假丝酵母菌、热带假丝酵母菌、光滑假丝酵母菌、克柔假丝酵母菌）、3种曲霉菌（烟曲霉、黄曲霉、黑曲霉）和新型隐球菌。其最低检测限为1 × 10⁴ cfu/mL，且常见细菌无扩增反应，重复性实验解链温度波动小于0.5 ℃。179例临床样本中，以培养法、镜检法、病理诊断等"金标准"方法诊断IFI阳性为112例，多重荧光PCR-熔解曲线法检测阳性为96例。多重荧光PCR-熔解曲线法同时检测假丝酵母菌、曲霉菌及新型隐球菌的敏感度为0.857，特异度为0.970，阳性预测值为0.980，阴性预测值为0.802，ROC曲线下面积为0.914，95%置信区间为0.868 ~ 0.959，P < 0.001。 结论本研究建立的多重荧光PCR-熔解曲线法可快速、准确、高通量同时检测假丝酵母菌、曲霉菌及新型隐球菌，对于IFI的早期诊断具有重要意义。     

基于交直轴电流耦合的单电流调节器永磁同步电机弱磁控制

基于交直轴电流耦合的单电流调节器弱磁控制是一种新颖的永磁同步电机弱磁策略，能够解决双电流调节器在电机高速域相互冲突而易于饱和的问题。交轴电压指令如何确定是此方法的研究重点，直接影响电机的电压利用率、效率以及负载能力。该文基于电机电压方程对电流耦合调节弱磁控制基本原理进行描述，并提出改进的控制方法。利用id-iq坐标平面上的定子电流轨迹，对现有方法的缺点和所提出方法的预期控制效果进行了分析。在小信号范围内，对所提出方法的动态控制过程和可控性进行了阐述。所提出的电流耦合调节变交轴电压弱磁控制策略，交轴电压指令根据电机工况自行调节，无需查表，且不依赖电机参数，易于实现。仿真和实验结果验证了所提方法的可行性和性能优势。     

Scientific Sessions (SS)

European Radiology -     

Early detection of cardiotoxicity by 3D speckle tracking imaging of area strain in breast cancer patients receiving chemotherapy

Eighty‐three breast cancer patients who underwent six cycles of EC chemotherapy regimen (epirubicin + cyclophosphamide) without symptoms and signs of heart disease were enrolled in the study. Three‐dimensional speckle tracking imaging technique (3D‐STI) was used to measure left ventricular global area strain (GAS), overall annular strain (GCS), overall longitudinal strain (GLS), and overall radial strain (GRS). Meanwhile, serum troponin T (Hs‐cTnT) was measured. The clinical value of each index on cardiotoxicity after chemotherapy was analyzed using the receiver operating characteristic (ROC) curve. Hs‐cTnT increased from the early stage to the end during chemotherapy, but it was still in the normal range. During the mid‐chemotherapy and the end‐chemotherapy, GAS, GLS, GCS, and E/A significantly reduced, while the changes in LVESV, LVEDV, LVEF, and GRS were not significant after chemotherapy. Pearson correlation analysis showed a significant negative correlation between GAS and anthracycline doses (r = ?.772, P < .01); GAS and Hs‐cTnT were significantly negatively correlated (P < .05). The area under the curve (AUC) of GAS, GLS, GCS, and GRS are 0.815, 0.683, 0.645, and 0.585, respectively. A GAS of ?31.5% was used as the cutoff value for diagnosing left ventricular systolic dysfunction after receiving chemotherapy. The sensitivity of the previous parameters was 81.9%, and the specificity was 80.3%. Interobserver consistency analysis showed that 3D‐STI strain parameter measurement has good repeatability. GAS has greater value in predicting early myocardial damage after anthracycline chemotherapy.