I-Corps: Fall prediction and prevention by an artificial compound eye camera system

该 I-Corps 项目更广泛的影响/商业潜力是预测和避免因各种与年龄相关的慢性疾病（如糖尿病、阿尔茨海默病和其他痴呆症、帕金森病和心力衰竭）而导致高跌倒风险的老年人跌倒和跌倒相关的伤害/死亡。跌倒是造成 65 岁及以上老年人意外伤害的主要原因，也是 65 岁及以上老年人死亡的第七大原因，仅在美国每年就有约 300 万人次到急诊室就诊。预防跌倒不仅可以改善整体健康状况并挽救生命，而且还可以通过减少医疗和急诊科就诊的费用来造福社会。具体来说，所提议技术的成功商业化可以为所有老年护理利益相关者提供价值，例如医院、护理人员（包括长期护理机构和家庭成员）、政府和私人保险公司以及与年龄相关的学术机构和辅助设备制造商。该I-Corps项目将进一步开发人工复眼摄像头（ACECam）系统，该系统可以从跌倒风险较高的老年人的日常生活中提取与临床相关的步态和移动数据，例如步态速度和从坐到站的时间。通过将创新的复眼光学器件与现成的图像传感器相集成，ACECam技术可以利用基于卷积神经网络开发的临床相关步态和移动数据算法，显着降低运动跟踪和动作识别的计算成本。此外，ACECam 不需要当前步态/运动跟踪可穿戴设备中常用的可穿戴传感器系统，因此对于老年人的日常生活来说并不引人注目。 ACECam 获取的原始数据包含有限的视觉信息，因此将保护用户的视觉隐私。该奖项反映了 NSF 的法定使命，并通过使用基金会的智力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 I-Corps project is to predict and avert falls and fall-related injuries/deaths of older adults with high risks of fall due to various age-related chronic diseases such as diabetes, Alzheimer's disease, and other dementias, Parkinson's disease and heart failure. Falls are a leading cause of unintentional injuries and seventh leading cause of mortality among older people aged 65 and over, which leads to approximately 3 million emergency department visits per year in U.S. alone. Prevention of falls will not only improve the overall health outcomes and save lives, but also will benefit society by reducing the expenses in medical treatment and emergency department visits. Specifically, the successful commercialization of the proposed technology could provide value to all senior care stakeholders such as hospitals, caregivers including long-term care facilities and family members, government and private insurers and age-related academic institutions and assisted device manufacturers.This I-Corps project will further develop an artificial compound eye camera (ACECam) system that can extract clinically-relevant gait and mobility data, such as gait speed and sit-to-stand time, from the daily life of older adults who can have higher risk of falls. By integration of innovative compound eye optics with off-the-shelf image sensors, the ACECam technology can significantly reduce the computational costs of motion tracking and action recognition with the algorithms for clinical-relevant gait and mobility data developed based on convolutional neural network. Additionally, ACECam is unobtrusive for the daily life of older adults by not requiring wearable sensor systems that are commonly applied in current gait/motion tracking wearables. The raw data acquired by ACECam contains limited visual information and so will protect visual privacy of the users.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.