Comparison between Classical and Intelligent Identification Systems for Classification of Gait Events

Abstract

Gait event detection is important for diagnosis and evaluation. This is a challenging endeavor due to subjectivity, high amount of data, among other problems. ANFIS (Artificial Neural Fuzzy Inference Systems), ARX (Autoregressive Models with Exogenous Variables), OE (Output Error models), NARX (Nonlinear Autoregressive Models with Exogenous Variables) and models based on NN (neural networks) were developed in order to detect gait events without the problems mentioned. The objective was to compare developed models' performance and determinate the most suitable model for gait events detection. Knee joint angle, heel foot switch and toe foot switch during normal walking in a treadmill were collected from a healthy volunteer. Gait events were classified by three experts in human motion. Experts' mean classification was obtained and all models were trained and tested with the collected data and experts' mean classification. Fit percentage was obtained to evaluate models performance. Fit percentages were: ANFIS: 79.49%, ARX: 68.8%, OE: 71.39%, NARX: 88.59%, NNARX: 67.66%, NNRARX: 68.25% and NNARMAX: 54.71%. NARX had the best performance for gait events classification. For ARX and OE, previous filtering is needed. NN's models showed the best performance for high frequency components. ANFIS and NARX were able to integrate criteria from three experts for gait analysis. NARX and ANFIS are suitable for gait event identification. Test with additional subjects is needed.

Distinct Feature Extraction for Video-Based Gait Phase Classification

“…Knee joint angle and foot switches are acquired as features to detect gait events in [30]. Classification fit percentages comparison between artificial neural fuzzy inference systems (ANFIS), autoregressive models with exogenous variables (ARX), output error models (OE), NARX and other NN-based models, demonstrate that the best model is NARX with a 88.59% fit rate.…”

“…Regardless of the motion capture system used, both the quality of acquired representative joint movements of interest and effectiveness of distinct feature selection for specific pathological classification, have a significant impact on gait assessment methods [30]. This is especially critical for systems relying on inexpensive motion capture systems, which need to demonstrate a comparable level of accuracy to clinical systems.…”

Gait phase classification for in-home gait assessment

“…In [22], artificial neural fuzzy inference systems (ANFIS), autoregressive models with exogenous variables (ARX), output error models (OE), NARX and other NN-based models are compared for gait event detection. Goniometer and foot switches are placed on volunteers leg and footwear to measure the knee flexion/extension angle with foot switches as features.…”

“…Gait phase analysis is an assessment method used for gait diagnosis and evaluation popularity [22]. Joint movements of interest are typically used to represent gait motion.…”

“…The quality of gait assessment not only depends on the accurate motion representation of each joint of interest [22], but also on accurate feature selection and classification from the joint trajectories. Therefore, in this paper, we propose exploiting the trajectories of joints of interest from motion analysis to generate additional features that can be used to better characterise gait, thus improving the quality of gait analysis.…”