一种基于足压数据主成分分析的足弓形态智能检测方法

目的 根据临床中平足与高弓足量化评估的要求，提出一种基于足压数据主成分分析（principal component analysis, PCA）智能快速足弓形态检测方法，并验证其临床有效性。方法 纳入诊断为足弓异常与足弓健康的志愿者，设计研发一套便携式足弓智能检测系统。采用44×52阵列式薄膜压阻传感器，采集静态站立式足底压力分布数据，利用自行编写的PCA算法自动拟合足轴线，进行足弓诊断并生成诊断报告。将足压采集结果与现有设备进行比对，验证足压数据的准确性。对于平足、高弓足和正常3类足弓的判别算法，通过对比临床诊断验证评估准确性。结果 该系统与现有压力采集设备的测量结果具有较好的相关性，接触面积偏差低于3.2%，计算拟合的足轴线与临床定义角度偏差小于1°，且该系统能获得与临床中足弓形态诊断相符率92.6%的评估结果。结论 引入PCA对足轴线自动化拟合，实现了快速而准确提取足弓信息的目的。该方法可用于临床实践中平足与高弓足的辅助筛查，有助于开展足弓畸形程度的量化分析和病理机制的研究。

An Intelligent Arch Diagnostic Method Based on Principal Component Analysis of Plantar Pressure Distribution

Objective According to clinical demand of quantification evaluation on flat foot and high arch, an intelligent and rapid method to diagnose arch shape based on principal component analysis (PCA) of plantar pressure is proposed, and its clinic validity is tested. Methods Volunteers diagnozed as abnormal arch and healthy arch were included in this study, and a portable intelligent arch test system was designed and developed. By adopting thin-firm piezoresistive sensor array with 44 rows, 52 columns of sensing units, the system could collect plantar pressure distribution data from the subjects under static standing. Foot axis could be fitted automatically by using the self-programmed PCA, so that foot diagnosis was completed with diagnostic report. The plantar pressure results from the system were compared with those from the existing plantar pressure acquisition device, so as to verify precision of collected data. The accuracy of the diagnosis algorithm for flat foot, high arch and healthy foot was verified through comparison with clinical diagnosis. Results The result of the system had a good correlation with that of the existing plantar pressure acquisition device, the deviation of contact area acquired by the system was smaller than 3.2%, and the angle deviation of the fitted foot axis with clinically defined angel was less than 1°. The system was capable of making diagnosis on arch shape that was 92.6% consistent with the clinical diagnosis. Conclusions PCA is introduced to automatically fit foot axis to achieve the purpose of fast and accurate extraction of foot arch information. The method can be used to assist clinical diagnosis of flat foot and high arch foot, and contribute to quantative analysis on foot arch deformity and its pathogenesis study.