摘要
一位新闻记者-机器人与机器学习的工作人员新闻编辑每日新闻-肿瘤学的新研究-乳腺癌是一篇报道的主题。据《新闻周刊》编辑在上海报道，“乳腺癌最常见的转移途径是乳腺淋巴网络，术前对腋窝淋巴结(ALN)负荷进行ACCU率评估，可以避免不必要的腋窝手术，从而预防手术并发症。”我们的新闻记者引用了上海公立医院临床中心的研究，“在本研究中，”我们的目的是建立一个结合乳腺特异性γ射线图像(BSGI)特征和超声参数的非侵入性预测模型，以评估腋窝淋巴结状态。建立了2012年至2021年间接受手术的乳腺癌患者队列(Trainin G组包括235例患者的1104个超声图像和940个BSGI图像。以99例患者的568幅超声图像和296幅BSGI图像为样本,在训练集中训练6种机器学习(ML)方法和递归特征剔除方法,建立了一个STRONG预测模型,并在此基础上建立了预测模型。我们建立了一个便于临床医师获取的线性预测器,用接收器工作特性(ROC)和校准曲线分别验证了模型的性能和评价模型的临床效果。在最优模型的基础上选择淋巴彩色多普勒血流显像分级和一个BSGI特征(腋窝肿块状态)，在测试集中，支持向量机模型的预测能力最好(AUC=0.794，灵敏度=0.641，特异性=0.8，PV=0.676，NPV=0.774，准确度=0.737)。ROC结果表明，该模型可以从BSGI特征中获益。
Abstract
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News – New research on Oncology - Breast Canc er is the subject of a report. According to news reporting out of Shanghai, Peop le’s Republic of China, by NewsRx editors, research stated, “The most common rou te of breast cancer metastasis is through the mammary lymphatic network. An accu rate assessment of the axillary lymph node (ALN) burden before surgery can avoid unnecessary axillary surgery, consequently preventing surgical complications.” Our news journalists obtained a quote from the research from Shanghai Public Hea lth Clinical Center, “In this study, we aimed to develop a non-invasive predicti on model incorporating breast specific gamma image (BSGI) features and ultrasono graphic parameters to assess axillary lymph node status. Cohorts of breast cance r patients who underwent surgery between 2012 and 2021 were created (The trainin g set included 1104 ultrasound images and 940 BSGI images from 235 patients, the test set included 568 ultrasound images and 296 BSGI images from 99 patients) f or the development of the prediction model. six machine learning (ML) methods an d recursive feature elimination were trained in the training set to create a str ong prediction model. Based on the best-performing model, we created an online c alculator that can make a linear predictor in patients easily accessible to clin icians. The receiver operating characteristic (ROC) and calibration curve are us ed to verify the model performance respectively and evaluate the clinical effect iveness of the model. Six ultrasonographic parameters (transverse diameter of tu mour, longitudinal diameter of tumour, lymphatic echogenicity, transverse diamet er of lymph nodes, longitudinal diameter of lymph nodes, lymphatic color Doppler flow imaging grade) and one BSGI features (axillary mass status) were selected based on the best-performing model. In the test set, the support vector machines ’ model showed the best predictive ability (AUC = 0.794, sensitivity = 0.641, sp ecificity = 0.8, PPV = 0.676, NPV = 0.774 and accuracy = 0.737). The result in ROC showed the model could benefit from incorporating BSGI fea ture.”
NETL
NSTL