Neuro-Fuzzy Prediction of Spine Loads in Response to Ri*

产品说明：与职业相关的下背部疾病（LBD）仍然是造成工作日损失的主要原因，也是当今行业面临的成本最高的职业安全和健康问题。已经确定的是，大多数与职业相关的LBD风险与手工材料处理活动以及工作场所的心理社会影响有关（美国国家科学院，2001年）。此外，个人因素也会影响风险。然而，我们对LBD风险的这些不同方面的风险特征的能力却知之甚少。最近的文献表明，这些风险维度中的一个共同联系涉及躯干肌肉共激活的增加，这可能导致脊柱负荷增加和随后的LBD。肌电图（EMG）辅助模型提供了唯一的手段，以准确评估和量化的影响，在躯干肌肉共激活脊髓负荷后的变化。然而，在大多数工业条件下收集EMG是不切实际的。这项工作的目的是开发一种脊柱负荷评估系统（SLAS），该系统能够评估躯干肌肉共激活模式和随后的脊柱负荷，以应对多个风险维度。该系统将允许人们准确地估计由于物理工作场所因素、心理社会因素和个人因素而导致的脊柱负荷，但不需要使用EMG。这一目标将通过开发混合神经模糊引擎（HNFE）来实现。这个引擎将作为一个系统人工“大脑”，能够综合有关工作场所的信息，并评估躯干肌肉组织的行为。该引擎将与一个成熟的生物驱动的动态生物力学模型的躯干接口。通过这种方式，我们将能够准确地预测脊柱负荷在工作场所响应于各种风险因素的尺寸，而不需要在工作场所收集EMG数据。总的来说，辅助计划会带来多项好处。首先，它可用于评估脊柱结构损伤的风险，这些损伤被认为是与工作场所相关的工作尺寸的函数，从而导致腰痛。因此，该模型将立即应用于工作场所设计。其次，该系统将提供关于风险的各个方面如何协同影响肌肉骨骼系统的见解。最后，它将促进有关稳定性和共活性的进一步研究。

DESCRIPTION: Occupationally-related low back disorders (LBDs) continue to be the leading cause of lost work days and the most costly occupational safety and health problem facing industry today. It has been well established that most occupationally-related LBD risk is associated with manual materials handling activities as well as psychosocial influences in the workplace (National Academy of Sciences, 2001). In addition, individual factors can influence risk. However, our ability to characterize risk associated with these various dimensions of LBD risk has been rather poorly understood. Recent literature indicates that a common link within each of these risk dimensions involves increases in trunk muscle coactivation that can lead to increased spine loading and subsequent LBD. Electromyographic (EMG)-assisted models provide the only means to accurately assess and quantify the effect of changes in trunk muscle coactivation upon spinal loading. However, the collection of EMG under most industrial conditions is impractical. The objective of this work is to develop a Spine Loading Assessment System (SLAS) that has the capacity to assess trunk muscle coactivation patterns and subsequent spine loading in response to multiple risk dimensions. This system would permit one to accurately estimate spine loading as a result of physical workplace factors, psychosocial factors, and individual factors but would not require the use of EMG. This objective will be achieved through the development of a Hybrid Neuro-Fuzzy Engine (HNFE). This engine would act as a system artificial "brain" able to synthesize information about the workplace and assess how the trunk musculature would behave. The engine will interface with a well-developed biologically-driven dynamic biomechanical model of the trunk. In this manner, we will be able to accurately predict spine loading in the workplace in response to various risk factor dimensions without the need to collect EMG data in the workplace. Collectively, the SLAS will have several benefits. First, it can be used to assess the risk of spine structure damage believed to contribute to low back pain as a function of work dimensions commonly associated with the workplace. Hence, this model will have immediate applications to workplace designs. Second, the system will provide insights as to how the various dimensions of risk synergistically impact the musculoskeletal system. Finally, it will facilitate further investigations regarding stability and coactivity.