Evaluation of methods based on conventional videography for detection of gait events

Abstract

People with lower limb amputations often have gait patterns that are different from non-amputees. Spatiotemporal parameters are commonly used to characterise amputee gait. In order to compute some of these parameters it is necessary to detect the initial contact (IC) of the foot with the floor and the end of contact or foot off (FO). A variety of methods have been proposed for detection of gait events, however, to be included in clinical settings in Argentina, the system needs to be portable, low cost and easy to use. Conventional videocameras are often used clinically as a tool for the analysis of walking. However, the characteristics of conventional video cameras, including slow sampling frequency, raise questions regarding their suitability for the detection of gait events. The objective of this pilot study was to evaluate the accuracy and reliability of two videographic methods for detection of IC and FO in non-amputees and amputees. Five non-amputees subjects and two subjects with unilateral lower limb amputation participated in the study. Two methods were used for event detection: a totally visual performed manually by raters and a method that included software tools to aid the raters (semi-automatic method). The results of detection were compared with those derived from the force platform. The reliability of the methods was evaluated by 10 raters (5 experienced and 5 inexperienced in the use of gait analysis tools) and the intra-class correlation coefficient (ICC) was then calculated. The methods showed mean differences lower than 60 ms when compared with event determination using force platforms. ICC values were greater than 0.90, indicating high reliability. However, the results from the raters in detection of FO suggest the need for appropriate training for the semiautomatic method. To conclude, the results of this study suggest that both methods could be used in clinical settings, if proper training is provided.