Videos can be an effective way to deliver contextualized, just-in-time medical information for patient education. However, video analysis, from topic identification and retrieval to extraction and analysis of medical information and understandability from a patient perspective are extremely challenging tasks. This study utilizes data analysis methods to retrieve medical information from YouTube videos concerning colonoscopy to manage health conditions. We first use the YouTube Data API to collect metadata of desired videos on select search keywords and use Google Video Intelligence API to analyze texts, frames and objects data. Then we annotate the YouTube video materials on medical information, video understandability annotation and recommendation. We develop a bidirectional long short-term memory (BLSTM) model to identify medical terms in videos and build three classifiers to group videos based on the level of encoded medical information, video understandability level and whether the videos are recommended. Our study provides healthcare practitioners and patients with guidelines for generating new educational video content and enabling management of health conditions.

翻译：视频可作为传递情境化、即时性医疗信息的有效患者教育工具。然而，从主题识别与检索到患者视角下医疗信息的提取分析及可理解性评估，视频分析仍面临极大挑战。本研究采用数据分析方法，从YouTube视频中检索与结肠镜相关的医疗信息以管理健康状况。我们首先利用YouTube数据API收集选定搜索关键词下目标视频的元数据，并采用Google视频智能API分析文本、画面帧及对象数据；随后对YouTube视频材料进行医疗信息标注、视频可理解性标注及推荐标注。通过构建双向长短期记忆（BLSTM）模型识别视频中的医学术语，并建立三个分类器分别对视频的医疗信息编码程度、视频可理解性水平及是否被推荐进行分组。本研究为医疗从业者及患者生成新型教育视频内容、实现健康状况管理提供了指导方针。