STTR Phase I: Smartphone-based Biomechanical Analysis for Job Risk Assessment

这项小型企业技术转移（STTR）第一阶段项目的更广泛影响/商业潜力是预防与工作相关的肌肉骨骼疾病（WMSDs），这是美国劳动力中导致疼痛、痛苦和残疾的主要原因。这项拟议中的技术仅通过处理智能手机拍摄的视频来估计施加在人体上的力和3D运动，而不需要在工人或物体上安装传感器。这一技术进步将为安全专业人员提供快速、简单和负担得起的人体工程学风险评估工具，而不会妨碍工作现场的活动，从而改善潜在患者的预后。在受WMSDs影响的众多行业中，直接目标是伤害率较高的制造业和分销业。这项小型企业技术转移（STTR）第一阶段项目旨在克服目前在工作现场进行人体工程学风险评估时所存在的侵入性、耗时和繁琐的力测量所带来的障碍。拟议的技术通过智能手机拍摄的视频来估计施加在身体关键部位（如颈部、肩部、背部和膝盖）上的力及其运动，而不会干扰工人正在进行的工作，从而使人体工程学风险评估无创、快速和简单。然而，对于在真实工作现场的应用，频繁的遮挡（遮挡工人身体的物体或结构）和复杂的工作环境对从视频中估计力和运动构成了重大挑战，该技术中使用的算法必须进行改进，并使用各种运动和力数据进行测试。该项目旨在通过新的数据增强方法，改进和优化力估计模型，以及广泛的实验室和实际工作现场数据收集和测试来解决这些技术挑战。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

The broader impact/commercial potential of this Small Business Technology Transfer (STTR) Phase I project is to prevent work-related musculoskeletal disorders (WMSDs), a leading cause of pain, suffering, and disability in the US workforce. The proposed technology estimates forces exerted on the human body and 3D motions solely by processing videos captured with a smartphone, and without needing to attach sensors to the workers or objects. This technological advancement will provide safety professionals with a rapid, easy and affordable ergonomic risk assessment tool without hindering activity at job sites, improving outcomes for potential patients. Among many industries that suffer from WMSDs, immediate targets are manufacturing and distribution that have high injury rates. This Small Business Technology Transfer (STTR) Phase I project aims to overcome the current hurdles posed by invasive, time-consuming, and cumbersome force measurement for ergonomic risk assessment at job sites. The proposed technology estimates forces exerted on key body parts (e.g., neck, shoulder, back, and knees) and their motions through videos captured with a smartphone and without interfering with workers’ ongoing work, thereby making ergonomic risk assessment non-invasive, rapid and easy. However, for application at real job sites where frequent occlusions (objects or structures obscuring the worker’s body) and complicated work environments pose a significant challenge to estimating forces and motions from videos, the algorithms used in the technology have to be improved and tested with a diverse range of motion and force data. This project aims to address these technical challenges through a new data augmentation approach, refinement and optimization of force estimation models, and extensive lab and real job site data collection and testing.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.