摘要
由一名新闻记者-机器人与机器学习的工作人员新闻编辑每日新闻-外科手术新研究-内窥镜是一篇报道的主题。根据NewsRx记者在印度新德里的新闻报道，研究表明，“目前缺乏视频胶囊内窥镜(VCE)期间清洁度评估的自动评分系统。本研究的重点是开发一种方法，根据最新评分，即韩国-加拿大(KODA)，自动评估VCE框架内的清洁度。”新闻记者从ECE部门获得了一句研究的引文：“最初，开发了一个名为Artificial Intellige NCE-Koda(AI-KODA)Score的易于使用的移动应用程序，以收集28个患者胶囊视频的多标签图像数据集。三名读者（胃肠病学研究员）接受了阅读VCE的培训，以重复的方式对该数据集进行评分，并实时自动保存标签，检查评分者之间和评分者内部的相关性，然后将所开发的数据集随机分成训练:验证:测试比70:20:10和60:20:20，然后使用10种机器学习和2种深度学习算法对三种多标签分类任务进行综合基准测试和评估，结果表明，可靠性估计总体良好。总体上，随机森林分类器在基于机器学习的分类任务中获得了最佳的评价指标，其次是Adaboost、Kneighbors和Gaussian Naive Bayes。Deep Le Arning算法在仅使用VM标签的情况下优于基于机器学习的分类任务。深入分析表明，该方法在清洁度评估方面具有节省时间的潜力，并且便于研究和临床使用。
Abstract
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News – New research on Surgical Procedures - Endoscopy is the subject of a report. According to news reporting from New Delhi , India, by NewsRx journalists, research stated, “An automated scoring system fo r cleanliness assessment during video capsule endoscopy (VCE) is presently lacki ng. The present study focused on developing an approach to automatically assess the cleanliness in VCE frames as per the latest scoring i.e., Korea-Canada (KODA ).” The news correspondents obtained a quote from the research from the Department o f ECE, “Initially, an easy-to-use mobile application called artificial intellige nce-KODA (AI-KODA) score was developed to collect a multi-label image dataset of twenty-eight patient capsule videos. Three readers (gastroenterology fellows), who had been trained in reading VCE, rated this dataset in a duplicate manner. T he labels were saved automatically in real-time. Inter-rater and intra-rater rel iability were checked. The developed dataset was then randomly split into train: validate:test ratio of 70:20:10 and 60:20:20. It was followed by a comprehensive benchmarking and evaluation of three multi-label classification tasks using ten machine learning and two deep learning algorithms. Reliability estimation was f ound to be overall good among the three readers. Overall, random forest classifi er achieved the best evaluation metrics, followed by Adaboost, KNeighbours, and Gaussian naive bayes in the machine learning-based classification tasks. Deep le arning algorithms outperformed the machine learning-based classification tasks f or only VM labels. Thorough analysis indicates that the proposed approach has th e potential to save time in cleanliness assessment and is user-friendly for rese arch and clinical use.”
NETL
NSTL