STTR Phase II: Real-time Analysis and Feedback during Colonoscopy to improve Quality

这个小企业技术转让（STTR）第二阶段项目将为内窥镜医生开发一种辅助软件工具，以便实时反馈结肠和潜在息肉区域检查的客观质量。该技术旨在减少结肠镜检查期间的息肉漏诊率。结肠镜检查有助于结直肠癌相关死亡的显著下降。 然而，最近的数据表明，即使是大息肉和癌症的检测也存在显着的漏诊率。 研究表明，息肉检测率可能与手术的退出阶段的持续时间有关，而癌症漏诊率与执行手术的内窥镜医生的身份有关。该工具将在实时结肠镜检查期间提供视频流分析和反馈，由新型中间件软件组成，以确保在经济实惠的工作站上高性能执行视频分析，并且是通用的，可重新配置的，具有支持时间约束视频分析的新任务分配。 实时反馈的客观度量来源于真实的时间分析，其将解决由模糊帧、粪便、相机移动模式以及阑尾口和息肉区域引起的复杂性。 从开发这种结肠镜检查工具中获得的技术见解可能适用于使用其他医学领域（如支气管镜检查、膀胱镜检查、关节镜检查和腹腔镜检查）生成的视频进行质量控制的新研究。如果在大规模日常医疗环境中成功实施，通过真实的时间客观质量评估，将为接受结肠镜检查程序的患者提供更高的护理质量，这在目前是不可行的。 在美国每年进行超过1400万次结肠镜检查。 这一辅助工具将促进高质量的检查，同时完成文件。 这将意味着内窥镜医生将能够花费更多的时间进行结肠镜检查，而花费更少的时间进行记录。 医院/诊所管理人员将能够更有效地运行内窥镜设施。 保险公司可能会受益于更好的文件和更低的成本，因为随着遗漏息肉和早期癌症的频率下降，需要广泛癌症治疗的患者将减少。该辅助工具还有望通过在内窥镜新手的教学和培训或经验丰富的内窥镜医生的继续教育期间提供自动反馈，为医学教育、研究和实践做出贡献。

This Small Business Technology Transfer (STTR) Phase II project will develop an assistive software tool for endoscopists to have real-time feedback of objective quality for colon and potential polyp region examinations. The technology is aimed to reduce polyp miss rates during colonoscopy. Colonoscopy has contributed to a marked decline in colorectal cancer related deaths. However, recent data suggest that there is a significant miss rate for the detection of even large polyps and cancers. Studies suggest that polyp detection rate may be related to the duration of the withdrawal phase of the procedure and cancer miss rate is related to the identity of the endoscopist performing the procedure. This tool, which will provide video stream analysis and feedback during live colonoscopy, is made up of novel middleware software to ensure high performance execution of video analysis on an affordable workstation, and are generic, reconfigurable with new task allocation that support time-constraint video analysis. Objective metrics for real-time feedback are derived from real time analysis that will address complexities arising from blurry frames, stool, camera movement patterns, and regions of appendiceal orifice and polyps. Technical insights learned from development of this tool for colonoscopy may be applicable to new research on quality control using videos generated in other areas of medicine, such as bronchoscopy, cystoscopy, arthroscopy and laparoscopy.The broader/commercial impact of this project, if successfully implemented in large-scale day-to-day medical settings, will be higher quality of care for patients undergoing colonoscopy procedures with real time objective quality assessment, which is currently not feasible. Over 14 million colonoscopies are performed annually in the US. This assistive tool will stimulate high quality inspection, while documentation is done. That will mean that endoscopists will be able to spend more time on performing the colonoscopy and less time on documentation. Hospital/clinic administrators will be able to run endoscopic facilities more efficiently. Insurance companies may benefit from better documentation and lower costs since fewer patients will require extensive cancer treatment as the frequency of missed polyps and early cancers declines. This assistive tool is also expected to contribute to medical education, research, and practice by providing automated feedback during teaching and training of novice endoscopists or continuing education for experienced endoscopists.