Elderly 3D balance and gait analysis with a simple depth-camera system

对2031年的预测显示，在加拿大，65岁以上的人口将在2031年达到近25%。这一重大的人口结构转变将产生重大影响。随着加拿大人口老龄化，医疗保健和福尔斯的间接费用下降，肌肉骨骼问题将增加。例如，对老年人最大的危害之一是意外福尔斯，62%的老年人因受伤住院是福尔斯造成的。因此，必须在老年人保健领域进行创新，以确定支持老年人的新技术。在这种情况下，该研究计划的总体长期目标是开发一种低成本，易于使用和便携式视频系统，以识别老年人平衡和步态的相关临床参数，用于临床环境中的早期跌倒风险评估和预防。这将通过考虑以下三个具体目标来实现。1. 3D重建：设计和开发简单的相机设置，以捕获足够准确的身体实时3D重建，以计算可靠的平衡/步态测量。该系统将包括放置在患者前方的单个深度相机（例如，来自微软的Kinect）和位于后方的两个平面镜，以提供“虚拟”后视图，从而在平衡或步态测试期间获得受试者的完整3D视图。2.平衡和步态分析：开发创新算法，以便从3D重建身体中识别相关临床参数。为此，将使用机器学习和形状和运动特征的统计分析。3.验证：将使用已知尺寸的3D几何对象（例如圆柱体）和先前使用金标准3D成像设备（例如激光扫描仪或光学数字化仪）扫描的人体模型来评价3D数据准确性。平衡和步态分析将通过学生志愿者在实验室模拟平衡障碍进行验证。我们的计算测量和金标准功能步态和平衡测试之间的相关性将被执行，以确定我们的方法的能力。这项研究将为老年医疗保健专业人员开发独特的基于计算机视觉的平衡和步态分析系统奠定基础。它将有助于改善临床护理，从而提高老年人的生活质量，也有助于患有其他疾病或影响其平衡和步态的损伤的患者。我们期望这种类型的成像系统将用于诊所，例如作为筛查工具，在专门设施进行更复杂的测试之前，早期检测老年人的平衡/步态问题。我们希望我们的系统相对简单，将有助于平衡/步态检查老年患者，并鼓励临床医生使用它。

Projections for 2031 show that in Canada, people over 65 will reach nearly 25% in 2031. This major demographic transformation will have important repercussions. As the Canadian population is aging, medical care and indirect costs from falls and musculoskeletal problems will increase. For instance, one of the greatest hazards for elderly is accidental falls and 62% of injury-related hospitalizations for seniors are the result of falls. It is therefore essential to innovate in the area of elder healthcare to identify new technologies to support the elderly. In this context, the general long-term objective of this program of research is to develop a low-cost, easy-to-use and portable video system to identify relevant clinical parameters in elder balance and gait for early fall risk assessment and prevention for use in a clinical setting. This will be achieved by considering the following three specific objectives. 1. 3D Reconstruction: Design and development of simple camera settings to capture real-time 3D reconstruction of the body sufficiently accurate to compute reliable balance/gait measurements. The system will consist in a single depth camera (e.g. Kinect from Microsoft) placed in front of the patient and two planar mirrors positioned behind providing “virtual” rear views to get a full 3D view of the subject during a balance or gait test. 2. Balance and Gait Analysis: Development of innovative algorithms in order to identify relevant clinical parameters from the 3D reconstructed body. To this end, machine learning and statistical analysis of shape and motion features will be used. 3. Validation: 3D data accuracy will be evaluated with 3D geometric object (e.g. cylinder) with known dimensions and with human mannequin previously scanned with a gold standard 3D imaging device (e.g. laser scanner or optical digitizer). Balance and gait analysis will be validated with student volunteers performing simulated balance impairment in laboratory. Correlations between our computed measurements and gold standard functional gait and balance tests will be performed to determine the ability of our methodology. This research will set the ground for the development of unique computer vision-based balance and gait analysis system for elderly healthcare professionals. It will contribute to improve clinical care and consequently the quality of life of elderly but also for patients suffering from other diseases or injuries affecting their balance and gait. We expect that this type of imaging systems will be used in clinics, for instance as a screening tool to detect elderly balance/gait problem early, before more sophisticated tests be conducted in specialized facilities. We hope that the relative simplicity of our system will facilitate balance/gait examinations for the elderly patient and encourage clinicians to use it.