A Thermal-Optical Vision System for Identification and Localization of Symptomatic COVID-19 Patients

用于识别和定位有症状的 COVID-19 患者的热光视觉系统

• 批准号: 554496-2020
• 负责人: JanabiSharifi, Farrokh
• 金额: $ 3.64万
• 依托单位: Ryerson University
• 依托单位国家: 加拿大
• 项目类别: Alliance Grants
• 财政年份: 2020
• 资助国家: 加拿大
• 起止时间: 2020-01-01 至 2021-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=705664
• 关键词: Thermal; Optical; Vision; System; Identification

According to World Health Organization (WHO) officials, timely identification and isolation of individuals infected with coronavirus is the key to containing the spread of the virus. With a shortage of testing kits, laboratory supplies, and personnel, temperature screening remains the primary approach to identifying high risk individuals. Thermal screening is non-invasive and gives immediate results for one of the primary symptoms of COVID-19. However, current methods of COVID-19 public screening in 'high traffic' environments involve handheld temperature measurements which are considered inefficient, ineffective, and potentially risky to the screener and other acutely ill patients. Relying on only elevated skin temperatures (and not core temperatures) using distance measurement units might also give false positive classification results. It is hypothesized that fusing temperature measurements with body gait patterns (markers) would increase the accuracy of identifying symptomatic COVID -19 patients. In this project, researchers from Ryerson University are partnering with Altius Analytics Labs to enhance the scalability and efficiency of COVID-19 public screening and disease tracking. The main goal of the project is to develop, validate, and test a novel thermal-optical vision-based system to automatically identify symptomatic individuals on a large scale and track their movement through an environment. The outcome of this project will have a significant impact, protecting public health for the current pandemic and providing safeguards against a potential second wave. In addition to medical screening applications, results can be extended to the human factors industry as well. Finally, the project includes significant multidisciplinary training of highly qualified personnel (HQP) in established and emerging areas, including kinesiology, instrumentation, machine vision, signal processing, and machine learning. HQP will also have opportunities to work with the project partner and gain career relevant industrial experience.

世界卫生组织（WHO）官员表示，及时识别和隔离冠状病毒感染者是遏制病毒传播的关键。由于检测试剂盒、实验室用品和人员短缺，体温筛查仍然是识别高危人群的主要方法。热筛查是非侵入性的，可以立即得出 COVID-19 主要症状之一的结果。然而，目前在“高流量”环境下进行的 COVID-19 公共筛查方法涉及手持式温度测量，这种方法被认为效率低下、效果不佳，并且对筛查人员和其他重症患者存在潜在风险。使用距离测量单位仅依赖升高的皮肤温度（而不是核心温度）也可能会给出误报分类结果。据推测，将温度测量与身体步态模式（标记）融合将提高识别有症状的 COVID-19 患者的准确性。 在该项目中，瑞尔森大学的研究人员与 Altius Analytics Labs 合作，以提高 COVID-19 公共筛查和疾病跟踪的可扩展性和效率。该项目的主要目标是开发、验证和测试一种新型的基于热光视觉的系统，以自动大规模识别有症状的个体并跟踪他们在环境中的活动。该项目的成果将产生重大影响，保护当前大流行的公共卫生并为潜在的第二波疫情提供保障。除了医学筛查应用之外，结果还可以扩展到人为因素行业。最后，该项目包括对现有领域和新兴领域的高素质人才 (HQP) 进行重要的多学科培训，包括运动机能学、仪器仪表、机器视觉、信号处理和机器学习。 HQP 还将有机会与项目合作伙伴合作并获得职业相关的行业经验。