基于深度学习的内镜超声胆管扫查辅助分站系统构建

目的　构建基于深度学习的人工智能内镜超声（endoscopic ultrasonography， EUS）胆管扫查辅助分站系统，以期辅助医师学习多站成像技术，提高操作水平。方法　回顾性收集武汉大学人民医院和武汉协和医院2016年5月—2020年10月522例EUS视频资料，基于视频截取图像，获得来自武汉大学人民医院的3 000张白光胃镜、31 003张超声胃镜图像和来自武汉协和医院的799张超声胃镜图像，用于EUS胆管扫查系统的模型训练和模型测试。模型包括：白光胃镜图像过滤模型，标准站图像与非标准站图像区分模型和EUS胆管扫查标准图像分站模型，用以将标准图像分为肝窗、胃窗、球窗、降窗。然后从测试集中随机抽取110张图像进行人机大赛，比较专家、高级内镜医师与人工智能模型对胆管扫查多站成像每个站点识别的准确度。结果　白光胃镜图像过滤模型准确率为100.00%（1 200/1 200），标准站图像与非标准站图像区分模型准确率为93.36%（2 938/3 147），EUS胆管扫查标准图像分站模型在内部测试集中各分类的准确率分别为肝窗97.23%（1 687/1 735），胃窗96.89%（1 681/1 735），球窗98.73%（1 713/1 735），降窗97.18%（1 686/1 735）；外部测试集中准确率分别为肝窗89.61%（716/799），胃窗92.74%（741/799），球窗90.11%（720/799），降窗92.24%（737/799）。人机大赛中，模型分站的正确率为89.09%（98/110），高于内镜医师［85.45%（94/110），74.55%（82/110），85.45%（94/110）］，接近专家水平［92.73%（102/110），90.00%（99/110）］。结论　本研究构建了一种基于深度学习的EUS胆管扫查系统，可以较为准确地实时辅助内镜医师进行标准多站扫查，提高EUS完整性及操作质量。     

计算机辅助判别超声内镜图像诊断胰腺癌的实验研究

目的观察利用数字图像处理技术提取超声内镜图像纹理特征,并运用于胰腺癌诊断的价值。方法随机选择2005年2月-2007年2月间行胰腺EUS检查的216名患者。其中胰腺癌153例,非胰腺癌患者（包括正常胰腺与慢性胰腺炎）63例,所有胰腺癌病例均经EUS-FNA细胞学检查确诊。选择EUS图像并提取纹理特征。根据最优特征组合,通过支撑向量机将病例进行自动分类为胰腺癌和非胰腺癌病例,并计算该诊断方法的敏感性、特异性和准确率。结果根据EUS图像共提取9大类,69个特征用于模式分类特征,其中类间距最大的25个特征被选取作为初始特征。将现有216例病例,随机划分为训练集和测试集,训练集108例（癌症76例,非癌症32例）、测试集108例（癌症77例,非癌症31例）,用训练集训练分类器,测试集进行测试。共进行50次随机实验,最终得出胰腺癌分类的准确性为（97．98±1．237）％,敏感性为（94．32±0．0354）％,特异性为（99．45±0．0102）％。结论数字图像处理技术与计算机辅助EUS图像判别法准确率高,无创伤性,为胰腺癌的临床诊断提供了一个新的、有价值的研究方向。     

深度学习技术对胃肠道间质瘤与平滑肌瘤超声内镜图像的鉴别诊断价值

目的　尝试构建基于深度学习技术的胃肠道间质瘤（gastrointestinal stromal tumors,GISTs）与平滑肌瘤（leiomyomas,LM）超声内镜图像分类模型,并验证其鉴别诊断价值。方法　回顾性纳入2014年10月至2021年10月在苏州大学附属第二医院接受超声内镜检查且经外科手术或内镜下切除后病理确诊的69例GISTs和73例LM病例,每例病例选取1张清晰且有典型病变的超声内镜图片,利用留出法将每种疾病图片按训练集图片数比验证集图片数为8∶2的比例分入训练集和验证集,最终由113张（55张GISTs和58张LM）超声内镜图片组成训练集,由29张（14张GISTs和15张LM）超声内镜图片组成验证集,训练集用于对深度学习模型进行训练与优化,验证集用于对分类模型进行验证,主要观察指标包括鉴别诊断的灵敏度、特异度、阳性预测值、阴性预测值和准确率。结果　利用Resnet 34网络结构建立的分类模型对GISTs与LM进行鉴别诊断的准确率趋于0.89,较Resnet 50网络结构建立的分类模型（0.81）的分类性能更佳。基于Resnet 34网络结构构建的分类模型对验证集中超声内镜图片进行鉴别诊断的灵敏度、特异度、阳性预测值、阴性预测值和准确率分别为85.71%（12/14）（95%CI：67.38%~100.00%）、93.33%（14/15）（95%CI：80.71%~100.00%）、92.31%（12/13）（95%CI：77.82%~100.00%）、87.50%（14/16）（95%CI：71.30%~100.00%）和89.66%（26/29）（95%CI：78.57%~100.00%）。结论　深度学习技术用于GISTs与LM超声内镜图像的鉴别诊断是可行的,可为临床医师对两者的鉴别提供辅助诊断意见。基于Resnet 34网络结构建立的深度学习模型对GISTs与LM超声内镜图像进行鉴别诊断的准确性较高。     

线上交互式培训在提高内镜医师胰腺内镜超声引导下细针抽吸术细胞病理诊断能力中的价值

目的　开发线上交互式细胞病理读片培训方案,并评估该方案在提高内镜医师胰腺内镜超声引导下细针抽吸术（endoscopic ultrasound?guided fine?needle aspiration,EUS?FNA）细胞病理诊断能力中的价值。方法　纳入2018年8月—2019年8月在南京鼓楼医院消化内科行EUS?FNA的194例胰腺实性占位患者的细胞刷片,从中共采集5 500张细胞病理图片,由高年资细胞病理医师对每张图片中的细胞类别进行标注,用于搭建线上交互式细胞病理读片培训学习和测试平台。5名无病理基础的内镜医师受邀参与本培训,比较培训前后内镜医师鉴别诊断癌和非癌的敏感度、特异度、阳性预测值和阴性预测值,评价线上交互式细胞病理读片培训方案在提高内镜医师细胞病理诊断能力中的作用。结果　本研究成功搭建供内镜医师线上学习和测试的交互式细胞病理读片培训平台。培训前内镜医师诊断癌和非癌的敏感度、特异度、阳性预测值、阴性预测值及准确率分别为0.55（95%CI：0.53~0.58）、0.32（95%CI：0.30~0.35）、0.43（95%CI：0.41~0.45）、0.44（95%CI：0.41~0.47）和0.43（95%CI：0.42~0.45）。经过培训,以上指标分别为0.96（95%CI：0.95~0.97）、0.70（95%CI：0.68~0.73）、0.74（95%CI：0.72~0.76）、0.95（95%CI：0.94~0.96）和0.81（95%CI：0.80~0.83）,较培训前均得到显著提升（P<0.001）。结论　线上交互式细胞病理读片培训方案可提高内镜医师对胰腺细胞病理的认识水平和诊断能力,有助于内镜医师在EUS?FNA过程中实施快速现场评估,提高EUS?FNA的诊断效能。     

人工智能对内镜医师染色放大内镜下胃癌识别能力的影响研究

目的评估人工智能（artificial intelligence,AI）辅助胃癌诊断系统在实时染色放大内镜视频中对内镜医师识别胃癌能力的影响。方法回顾性收集2017年3月—2020年1月武汉大学人民医院和公开数据集中的早期胃癌和非癌染色放大内镜图片作为训练集和独立测试集,其中训练集包括4 667张图片（1 950张早期胃癌和2 717张非癌）,测试集包括1 539张图片（483张早期胃癌和1 056张非癌）。利用深度学习进行模型训练。前瞻性收集2020年6月9日—2020年11月17日来自北京大学肿瘤医院和武汉大学人民医院的100例患者的染色放大内镜视频（包含38例癌和62例非癌）作为视频测试集。纳入来自另外4家医院的4名不同年资内镜医师,分2次（无或有AI辅助）对视频测试集进行诊断,评估AI对内镜医师判断胃癌能力的影响。结果无AI辅助时,内镜医师诊断视频测试集中胃癌的准确率、敏感度和特异度分别为81.00%±4.30%、71.05%±9.67%和87.10%±10.88%;在AI辅助下,内镜医师辨认胃癌的准确率、敏感度和特异度分别为86.50%±2.06%、84.87%±11.07%和87.50%±4.47%,诊断准确率（P=0.302）和敏感度（P=0.180）较无AI辅助时均有提升。AI在视频测试集中辨认胃癌的准确率为88.00%（88/100）,敏感度为97.37%（37/38）,特异度为82.26%（51/62）,AI的敏感度高于内镜医师平均水平（P=0.002）。结论AI辅助诊断系统是染色放大内镜模式下辅助诊断胃癌的有效工具,可提高内镜医师对胃癌的诊断能力。它能实时提醒内镜医师关注高风险区域,以降低漏诊率。     

基于深度学习的人工智能技术在早期胃癌诊断中的应用

目的 构建和验证一个用于早期胃癌自动识别的深度学习模型,旨在提高早期胃癌的识别和诊断水平。 方法 从长海医院消化内镜中心数据库选取2014年5月至2016年12月期间5 159张胃镜图像,其中包括早期胃癌1 000张,良性病变及正常图像4 159张。首先选取4 449张图像（其中早期胃癌图像768张,其他良性病变及正常图像3 681张）用于深度学习模型的训练。然后将剩余的710张图像用于模型的验证,同时再交给4名内镜医师进行诊断。最后统计相关结果。 结果 深度学习模型用于早期胃癌诊断的准确率89.4％（635／710）、敏感度88.8％（206／232）、特异度89.7％（429／478）,每张图像的诊断时间为（0.30±0.02）s,均优于相比较的4名内镜医师。 结论 本研究构建的深度学习模型用于早期胃癌的诊断具有较高的准确率、特异度和敏感度,可在胃镜检查中辅助内镜医师进行实时诊断。     

基于人工智能的自动内镜下病灶尺寸测量系统（含视频）

目的　开发一个基于人工智能的自动内镜下病灶尺寸测量系统,并测试其实时测量白光内镜下病灶尺寸的能力。方法　测量系统由3个模型组成：首先由模型1识别视频的连续图片中有无活检钳,有钳者标记钳叶轮廓;随后由模型2对有钳图片进行分类,分为张钳图片和未张钳图片;与此同时,模型3识别视频的连续图片中有无病灶,有病灶者标记边界;最后系统根据活检钳钳叶轮廓与病灶边界的像素对比,实时计算出病灶尺寸。数据集1由回顾性收集的武汉大学人民医院2017年1月1日—2019年11月30日4 835张图片组成,用于模型的训练和验证;数据集2由前瞻性收集的武汉大学人民医院内镜中心2019年12月1日—2020年6月4日检查拍摄的图片组成,用于测试模型分割活检钳边界和病灶边界的能力;数据集3由151个模拟病灶的302张图片组成,每个模拟病灶包括活检钳倾斜角度较大（与病灶垂直线夹角45°）和倾斜角度较小（与病灶垂直线夹角10°）情况下的图片各1张,用于测试模型在活检钳不同状态下测量病灶尺寸的能力;数据集4为视频测试集,由前瞻性收集的武汉大学人民医院内镜中心2019年8月5日—2020年9月4日检查拍摄的视频组成。以内镜医师复核后结果或内镜手术病理作为金标准,观察模型1识别有无活检钳的准确率、模型2分类活检钳状态（张钳或未张钳）的准确率和模型3识别有无病灶的准确率,用交并比（intersection over union,IoU）评价模型1的活检钳钳叶分割效果和模型3的病灶分割效果,用绝对误差和相对误差评价系统的病灶尺寸测量能力。结果　（1）数据集2共纳入1 252张图片,有钳图片821张（其中张钳图片401张、未张钳图片420张）、无钳图片431张;包含病灶图片640张、不包含病灶图片612张。模型1判断无钳图片433张（430张准确）、有钳图片819张（818张准确）,识别有无活检钳的准确率为99.68%（1 248/1 252）,以818张模型1准确判断有钳图片的数据统计模型1的活检钳钳叶分割效果,平均IoU为0.91（95%CI：0.90～0.92）。使用模型1准确判断的818张有钳图片评价模型2的活检钳状态分类准确率,模型2判断张钳图片384张（382张准确）、未张钳图片434张（416张准确）,模型2的活检钳状态分类准确率为97.56%（798/818）。模型3判断包含病灶图片654张（626张准确）、不包含病灶图片598张（584张准确）,识别有无病灶的准确率为96.65%（1 210/1 252）,以626张模型3准确判断有病灶图片的数据统计模型3的病灶分割效果,平均IoU为0.86（95%CI：0.85～0.87）。（2）数据集3中：活检钳倾斜角度较小状态下系统病灶尺寸测量的平均绝对误差为0.17 mm（95%CI：0.08～0.28 mm）,平均相对误差为3.77%（95%CI：0.00%～10.85%）;活检钳倾斜角度较大状态下系统病灶尺寸测量的平均绝对误差为0.17 mm（95%CI：0.09～0.26 mm）,平均相对误差为4.02%（95%CI：2.90%～5.14%）。（3）数据集4共纳入59例患者的59个内镜检查视频的780张图片,系统病灶尺寸测量的平均绝对误差为0.24 mm（95%CI：0.00～0.67 mm）,平均相对误差为9.74%（95%CI：0.00%～29.83%）。结论　基于人工智能的自动内镜下病灶尺寸测量系统可以实现内镜下对病灶尺寸的准确测量,有望提高内镜医师对病灶尺寸估计的准确率。     

人工智能超声内镜胆胰识别系统有效性的交叉试验

目的探究人工智能超声内镜(artificial intelligence-endoscopic ultrasound, AI-EUS)胆胰识别系统用于辅助识别超声内镜检查术(endoscopic ultrasonography, EUS)图像的有效性。方法从武汉大学人民医院消化内科数据库前瞻性地纳入2019年12月—2020年8月期间因怀疑有胆胰系统疾病而接受EUS检查的受试者。28例受试者的28个视频用于胰腺标准站的识别;29例受试者的29个视频用于胆管标准站的识别。8名武汉大学人民医院消化内科的新手内镜医师在有AI-EUS胆胰识别系统辅助下和无辅助下, 分别阅读了57例EUS视频。比较有AI-EUS胆胰识别系统与无AI-EUS胆胰识别系统辅助时, 内镜医师对EUS胰腺和胆管标准站点识别的准确率。结果无AI-EUS辅助时, 新手内镜医师对胰腺标准站识别准确率为67.2%(903/1 344), 有AI-EUS辅助时, 准确率提高至78.4%(1 054/1 344);胆管标准站识别准确率由无辅助时的56.4%(523/928)提高至有辅助时的73.8%(685/928)。结论 AI-...     

胰腺疾病的内镜超声检查术概述

内镜超声检查术（endoscopic ultrasonography,EUS）原理与腹部超声相似,区别只是将微型高频超声探头安装在内镜顶端,进行实时超声扫描。胰腺是腹膜后器官,没有骨性结构,普通影像学检查对诊断胰腺疾病有一定局限性。由于EUS内镜插入胃和十二指肠肠腔内,经胃壁、十二指肠壁观察邻近胰腺组织,可清晰显示胰腺实质和胰管,因此EUS对诊断胰腺疾病具有独特优势,加之内镜超声引导下细针穿刺活检（EUS—FNA）技术,EUS在胰腺疾病诊断方面发挥着不可或缺的作用。     

内镜超声诊断胰腺疾病的价值

内镜超声（endoscopic ultrasonography，EUS）在胰腺各种良恶性病变的诊断和治疗中的应用价值正被越来越多的内镜医师和消化科医师所认识。在胰腺癌的TNM分期上．Eus与CT、MRI、MRCP、PET等影像学检查互为补充．着重于对胰腺癌的浸润范围（T）和淋巴结转移（N）进行判断；在对胰腺囊性病变性质的判断上，其影像学特征较CT检查更为敏感．加之EUS引导下的细针穿刺（EUS—guided fine—needleaspiration，EUSFNA）对囊液进行各项生化（CEA、CAl99和淀粉酶）及细胞学检查．对判断囊肿的性质和决定治疗方案具有不可替代的作用。随着内镜超声引导下的Trucut活检技术（EUSguided Trucut biopsy．EUSTCB）的发明，为我们准确判断胰腺各种病变的性质提供了组织病理学的诊断。鉴于EUS及其引导下的细针穿刺，尤其是Trucut活检在国内尚未普遍开展，有必要将这些技术进行总结．以更好地指导临床运用，并使临床医师对其有更深入的了解。     

Numerical modification of the Los Angeles classification of gastroesophageal reflux disease fails to decrease observer variation

Background: We previously reported that a new endoscopic classification of gastroesophageal reflux disease, the Los Angeles classification, showed considerable observer variation depending on the experience of the endoscopist. In the present study, we evaluated some modifications of the classification to determine whether we could decrease observer variation. Methods: Fifty endoscopic photographs, each showing four images of the squamo‐columnar junction, were prospectively obtained from 50 consecutive patients with gastroesophageal reflux disease. Two groups of eight endoscopists divided by their endoscopic experience, group 1 (100–500 procedures) and group 2 (more than 500 procedures), assessed the photographs using classifications with the following modifications: (i) addition of grade O to describe healed mucosal breaks and setting grade B as more than 5 mm or 10 mm; or (ii) addition of grade O and setting grade D as 75–99% or 100% circumferential. Results: Changing the definition of grade B or grade D did not increase the kappa values for either group of observers. Conclusions: These modifications of the Los Angeles classification were unable to decrease observer variation.     

Role of endoscopic ultrasound in diagnosis and therapy of pancreatic adenocarcinoma

Since its advent more than 20 years ago, endoscopic ultrasound （EUS） has undergone evolution from an experimental to a diagnostic instrument and is now established as a therapeutic tool for endoscopists. Endoscopic ultrasound cannot accurately distinguish benign from malignant changes in the primary lesion or lymph node on imaging alone. With the introduction of the curved linear array echoendoscope in the 2990s, the indications for EUS have expanded. The curved linear array echoendoscope enables the visualization of a needle as it exits from the biopsy channel in the same plane of ultrasound imaging in real time. This allows the endoscopist to perform a whole range of interventional applications ranging from fine needle aspiration （FNA） of lesions surrounding the gastrointestinal tract to celiac plexus block and drainage of pancreatic pseudocyst. This article reviews the current role of EUS and EUS-FNA in diagnosis, staging and interventional application of solid pancreatic cancer.     

EUS-guided FNA diagnostic yield of malignancy in solid pancreatic masses: a benchmark for quality performance measurement

BACKGROUND: The diagnostic yield of EUS-guided FNA (EUS-FNA) of solid pancreatic masses is a potential benchmark for EUS-FNA quality, because the majority of EUS-FNA of solid pancreatic masses should be diagnostic for malignancy. OBJECTIVES: To determine the cytologic diagnostic rate of malignancy in EUS-FNA of solid pancreatic masses and to determine if variability exists among endoscopists and centers. DESIGN: Multicenter retrospective study. PATIENTS: EUS centers provided cytology reports for all EUS-FNAs of solid, noncystic, >or=10-mm-diameter, solid pancreatic masses during a 1-year period. MAIN OUTCOME MEASUREMENT: Cytology diagnostic of pancreatic malignancy. RESULTS: A total of 1075 patients underwent EUS-FNA at 21 centers (81% academic) with 41 endoscopists. The median number of EUS-FNA of solid pancreatic masses performed during the year per center was 46 (range, 4-177) and per endoscopist was 19 (range, 1-97). The mean mass dimensions were 32 x 27 mm, with 73% located in the head. The mean number of passes was 3.5. Of the centers, 90% used immediate cytologic evaluation. The overall diagnostic rate of malignancy was 71%, 95% confidence interval 0.69%-0.74%, with 5% suspicious for malignancy, 6% atypical cells, and 18% negative for malignancy. The median diagnostic rate per center was 78% (range, 39%-93%; 1st quartile, 61%) and per endoscopist was 75% (range, 0%-100%; 1st quartile, 52%). LIMITATIONS: Retrospective study, participation bias, and varying chronic pancreatitis prevalence. CONCLUSIONS: (1) EUS-FNA cytology was diagnostic of malignancy in 71% of solid pancreatic masses and (2) endoscopists with a final cytologic diagnosis rate of malignancy for EUS-FNA of solid masses that was less than 52% were in the lowest quartile and should evaluate reasons for their low yield.     

Sedation practices in gastrointestinal endoscopy—A survey from southern India

Gastrointestinal endoscopies can cause an unpleasant experience for the patient. In India, most endoscopists follow a common institutional policy for sedation. The aim of this study was to analyze the sedation practices in various endoscopy centers across southern India. Data were collected with the help of a structured questionnaire given to a senior endoscopist of the center. Data from the completed questionnaire were later analyzed. Data were obtained from 19 centers across southern India. All endoscopy suites had central oxygen supply and emergency cart. A defibrillator was available in 12 centers (63.2%). Common criteria followed for administering sedation included therapeutic procedures (84.2%), patients who requested sedation (63.2%), children (63.2%), high-risk procedures (57.9%), and uncooperative patients (57.9%). Monitoring methods included pulse oximetry alone in six centers (31.6%), pulse oximetry with blood pressure monitoring in five centers (26.3%), and pulse oximetry, blood pressure, and electrocardiography (ECG) monitoring in eight centers (42.1%). For advanced procedures like endoscopic ultrasonography (EUS) and endoscopic retrograde cholangiopancreatography (ERCP), sedation was universally used. An anesthesiologist was available in the endoscopy suite in eight centers (42.1%). Five endoscopists administered propofol sedation without anesthesiologist’s presence (26.3%). Thirteen centers had a written protocol for pre-procedure risk assessment (68.4%). A dedicated post-procedure observation area was available in seventeen centers (89.5%). Seven centers followed a written post-sedation discharge protocol (36.8%). Significant variations exist in the practice of sedation among endoscopists in southern India. There is an urgent need to formulate guidelines by endoscopy societies for ensuring better patient outcomes in endoscopy.     

A new endoscopic ultrasonography image processing method to evaluate the prognosis for pancreatic cancer treated with interstitial brachytherapy

AIM:To develop a fuzzy classification method to score the texture features of pancreatic cancer in endoscopic ultrasonography(EUS)images and evaluate its utility in making prognosis judgments for patients with unresectable pancreatic cancer treated by EUS-guided interstitial brachytherapy.METHODS:EUS images from our retrospective database were analyzed.The regions of interest were drawn,and texture features were extracted,selected,and scored with a fuzzy classification method using a C++program.Then,patients with unresectable pancreatic cancer were enrolled to receive EUS-guided iodine 125 radioactive seed implantation.Their fuzzy classification scores,tumor volumes,and carbohydrate antigen 199(CA199)levels before and after the brachytherapy were recorded.The association between the changes in these parameters and overall survival was analyzed statistically.RESULTS:EUS images of 153 patients with pancreatic cancer and 63 non-cancer patients were analyzed.A total of 25 consecutive patients were enrolled,and they tolerated the brachytherapy well without any complications.There was a correlation between the change in the fuzzy classification score and overall survival(Spearman test,r=0.616,P=0.001),whereas no correlation was found to be significant between the change in tumor volume(P=0.663),CA199 level(P=0.659),and overall survival.There were 15 patients with a decrease in their fuzzy classification score after brachytherapy,whereas the fuzzy classification score increased in another 10 patients.There was a significant difference in overall survival between the two groups(67 d vs 151 d,P=0.001),but not in the change of tumor volume and CA199 level.CONCLUSION:Using the fuzzy classification method to analyze EUS images of pancreatic cancer is feasible,and the method can be used to make prognosis judgments for patients with unresectable pancreatic cancer treated by interstitial brachytherapy.     

Software for automated classification of probe-based confocal laser endomicroscopy videos of colorectal polyps

AIM:To support probe-based confocal laser endomi-croscopy (pCLE) diagnosis by designing software for the automated classification of colonic polyps. METHODS:Intravenous fluorescein pCLE imaging of colorectal lesions was performed on patients under-going screening and surveillance colonoscopies, followed by polypectomies. All resected specimens were reviewed by a reference gastrointestinal pathologist blinded to pCLE information. Histopathology was used as the criterion standard for the differentiation between neoplastic and non-neoplastic lesions. The pCLE video sequences, recorded for each polyp, were analyzed off-line by 2 expert endoscopists who were blinded to the endoscopic characteristics and histopathology. These pCLE videos, along with their histopathology diagnosis, were used to train the automated classification software which is a content-based image retrieval technique followed by k-nearest neighbor classification. The performance of the off-line diagnosis of pCLE videos established by the 2 expert endoscopists was compared with that of automated pCLE software classification. All evaluations were performed using leave-one-patient- out cross-validation to avoid bias. RESULTS:Colorectal lesions (135) were imaged in 71 patients. Based on histopathology, 93 of these 135 lesions were neoplastic and 42 were non-neoplastic. The study found no statistical significance for the difference between the performance of automated pCLE software classification (accuracy 89.6%, sensitivity 92.5%, specificity 83.3%, using leave-one-patient-out cross-validation) and the performance of the off-line diagnosis of pCLE videos established by the 2 expert endoscopists (accuracy 89.6%, sensitivity 91.4%, specificity 85.7%). There was very low power (< 6%) to detect the observed differences. The 95% confidence intervals for equivalence testing were:-0.073 to 0.073 for accuracy, -0.068 to 0.089 for sensitivity and -0.18 to 0.13 for specificity. The classification software proposed in this study is not a "black box" but an informative tool based on the query by example model that produces, as intermediate results, visually similar annotated videos that are directly interpretable by the endoscopist. CONCLUSION:The proposed software for automated classification of pCLE videos of colonic polyps achieves high performance, comparable to that of off-line diagnosis of pCLE videos established by expert endoscopists.     

EUS-guided core biopsies of pancreatic solid masses using a new fork-tip needle: A multicenter prospective study

Background and aimEndoscopic ultrasound-guided sampling (EUS sampling) is a safe and effective technique. The study aim was to evaluate the presence of a histological core from pancreatic lesions using a new 25G fork-tip needle.MethodsObservational multicenter prospective and analytical study, including consecutive patients with solid pancreatic masses referred for EUS-guided sampling. At each needle pass, the endoscopist performed macroscopic on-site evaluation (MOSE). The primary outcome was the histological core procurement rates. Secondary outcomes were the evaluation of interobserver agreement between endoscopists and pathologists, adequacy of EUS samples for the diagnosis and post-procedure adverse events.Results100 patients were enrolled in 3 centers. The mean size of the lesions was 28.5 mm (SD 11.7). Final diagnoses were adenocarcinoma (68%), neuroendocrine tumor (21%), inflammatory mass/benign lesions (8.0%), and pancreatic metastasis (3.0%). The pathologists described the presence of a core in 67 samples (67.0% of patients), with poor agreement with MOSE (kappa, 0. 12; 95% CI: 0.03–0.28). The diagnostic accuracy was 93%. We observed 6% of mild adverse events.ConclusionThe new 25-gauge core needle showed good overall adequacy and a good rate of histological specimens during EUS sampling of solid pancreatic masses, with a minimum number of passes and no major complications. Clinicaltrial.gov number, NCT02946840.     

OBSTACLES TO WIDESPREAD USE OF ENDOSCOPIC ULTRASONOGRAPHY IN THE DIAGNOSIS OF BILIOPANCREATIC DISEASES

The population of endoscopists who can perform biliopancreatic endoscopic ultrasonography (EUS) has not been increasing as expected. To encourage young endoscopists to become more involved in biliopancreatic EUS, re‐assessment and establishment of a typology of typical EUS images obtained by transduodenal scanning is necessary. Standardization of the maneuver and interpretation of EUS images would help increase the number of skilled echoendoscopists.     

国际专家组关于超声内镜引导下胰腺囊性肿瘤消融的立场声明

背景和目的最近,一些有关胰腺囊性肿瘤(pancreatic cystic neoplasms,PCNs)治疗的指南建议已经发表,但是超声内镜(endoscopic ultrasound,EUS)引导下PCNs消融的作用在这些指南中尚未明确。本文的目的是通过提出一系列临床问题并根据可获得的最佳科学证据提供答案来探讨围绕EUS引导下PCNs消融的问题。 方法我们从亚洲EUS学组和一个国际专家组中招募了一个EUS引导下PCNs消融专家小组。创建了一个临床问题列表,并将每个问题分配给一个成员以产生一个陈述。然后在2016年10月至2017年10月的三次互联网会议上讨论了这些陈述。之后相互协商对这些陈述进行修改,直到取得协商一致意见。之后,将完整的陈述总结集体发送给所有小组成员,以对陈述的强度进行投票,对陈述进行分类,对证据进行分级。 结果制定了关于EUS引导下PCNs的23个陈述。这些陈述涉及操作步骤、操作技术、操作前和操作后管理,并发症的管理以及手术中所需能力和培训。 结论在所有内镜学会中,本共识是关于EUS引导下PCNs消融的首篇共识。有兴趣进行这项技术的临床医生应参考这些共识,未来的研究应设法解决本共识中提出的重要问题。     

Preoperative diagnosis and staging of gallbladder carcinoma by EUS

BACKGROUND: EUS has recently been shown to be efficacious for the preoperative assessment of depth of invasion of gallbladder carcinoma. This study assessed the value of EUS for determining T stage (International Union Against Cancer). METHODS: Preoperative EUS findings in 41 patients with gallbladder carcinoma were analyzed retrospectively. EUS images were classified according to the shape of the tumor and the adjacent gallbladder wall structure as follows: type A, pedunculated mass with preserved adjacent wall structures; type B, sessile and/or broad-based mass with a preserved outer hyperechoic layer of the gallbladder wall; type C, sessile and/or broad-based mass with a narrowed outer hyperechoic layer; type D, sessile and/or broad-based mass with a disrupted outer hyperechoic layer. EUS and histopathologic findings were compared, including the depth of invasion of the tumor in the resection specimen. RESULTS: The 4 categories of EUS images of gallbladder carcinoma correlated with the histologic depth of invasion and T stage. Accuracies for the EUS classification as type A corresponding to pTis, type B to pT1, type C to pT2, and type D to pT3-4 were, respectively, 100%, 75.6%, 85.3%, and 92.7%. CONCLUSIONS: Preoperative EUS imaging accurately depicts T stage of gallbladder carcinoma and allows for effective therapeutic decision making.