A vision-based analysis system for gait recognition in patients with Parkinson’s disease

Abstract

Recognition of specific Parkinsonian gait patterns is helpful in the diagnosis of Parkinson’s disease (PD). However, there are few computer-aided methods to identify the specific gait patterns of PD. We propose a vision-based diagnostic system to aid in recognition of the gait patterns of Parkinson’s disease. The proposed system utilizes an algorithm combining principal component analysis (PCA) with linear discriminant analysis (LDA). This scheme not only addresses the high data dimensionality problem during image processing but also distinguishes different gait categories simultaneously. The feasibility of the proposed system for the recognition of PD gait was tested by using gait videos of PD and normal subjects. The efficiency of feature extraction using PCA and LDA coefficients are also compared. Experimental results showed that LDA had a recognition rate for Parkinsonian gait of 95.49%, which is higher than the conventional PCA feature extraction method. The proposed system is a promising aid in identifying the gait of Parkinson’s disease patients and can discriminate the gait patterns of PD patients and normal people with a very high classification rate.

Introduction

Most of the current methods used for evaluating Parkinson’s disease (PD) rely heavily on human expertise, e.g., the use of the unified Parkinson disease rating scale (UPDRS) (Martı´n et al., 2004). UPDRS is a rating tool that follows the longitudinal course of PD. It is composed of 5 separate categories including mentation, behavior, mood, activities of daily living and motor examinations, all evaluated by interview. Some sections require multiple grades assigned to each extremity.

The analysis of gait characteristics, as documented by Knutsson (1972), shows that PD patients exhibit large gait variability. Compared with normal people, PD patients’ walking speed is slower, duration of gait cycle is longer, stride length is shorter, and amplitude of range of movement of joints is decreased. These specific gait patterns (e.g. festinating gait, freezing gait) are widely accepted as a prominent feature of PD (McDowell, 1971). However, since posture and gait movement can vary from person to person, the evaluation of Parkinsonian gait tends to be subjective and depends greatly on the experience and judgment of the clinician (Blin et al., 1990, Lubik et al., 2006, Melnick et al., 2002, Shan et al., 2001, Salarian et al., 2004, Sofuwa et al., 2005, Stern et al., 1983, Vokaer et al., 2003).

In addition to evidence-based practice, therapists also use objective, quantitative methods to improve diagnosis of Parkinsonian gait. As a result, engineering-oriented machine learning-based methods have attracted more and more attention in this field (Engin et al., 2007, Fahrenberg et al., 1997, Makikawa and Iizumi, 1995, Sekine et al., 2002, Veltink et al., 1996). Many previous studies have used dc and ac accelerometers to assess gait patterns. They classified the accelerometer signals into different types of walking and correlated them with energy consumption. Nevertheless, those methods often used a number of sensors, causing patient discomfort.

Vision-based gait analysis systems avoid this problem. Since these systems require no physical contact, they are more comfortable and acceptable to the patients. Vision-based gait analysis is divided into two main categories, model-based and holistic. Model based approaches fit their model to the image data (Cunado et al., 1999, Yam et al., 2002). These image processing systems use markers on the body and record several steps of the patient. An average of three or more walks is then computed. The temporal characteristics of gait, e.g., stride length, width, cadence and velocity are measured (Melnick et al., 2002). In one study (Cunado et al., 1999) the gait signature was extracted using a Fourier series to describe the motion of the leg and temporally correlate this leg motion to determine the dynamic model from a sequence of images.

Holistic methods (Huang et al., 1999a, Huang et al., 1999b, Little and Boyd, 1998) extract posture cues by preserving the silhouettes of people when walking, derive statistical information directly from the gait image and attempt to correlate various features for biometric authentication. The holistic approach has a high human identification rate. For instance, Murase and Sakai (1996) captured the complete gait images of people which they then subtracted and matched using spatial–temporal correlation.

To reduce the dimensionality of image data, there are many linear transformation approaches that can be used. These methods are usually expressed as y = W^Tx, where x and y are the original and the dimensionality-reduced image vectors, respectively, and W is a linear projection matrix such that y becomes discriminative so as to aid separation of different classes of image sequences.

Many types of optimization criteria can be used to determine an appropriate W, such as maximizing the variance, non-Gaussianity for independancy, negentropy, or the ratio of between- and within-class variations (Hyvarinen et al., 2001, Liu and Wechsler, 1998, Murase and Sakai, 1996). Among them, principal component analysis (PCA) is well-known and widely used (Polat & Güneş, 2007). PCA focuses on computing eigenvectors that account for the largest variance of the data selected, but these directions do not necessarily provide the best separation of gait classes. On the other hand, the ratio of between- and within-class variations (Fisher’s linear discriminant criterion) appears to be an especially valid index since it allows simultaneous balancing between the maximization and minimization of the between- and within-class variations. Based on Fisher’s linear discriminant criterion, linear discriminant analysis (LDA) then produces a linear projection matrix, which greatly enhances classification.

The aim of this paper is to discriminate PD patients from normal subjects using a vision-based gait analysis approach. The scheme utilizes the holistic image of subjects, and extracts and reduces the feature space by using PCA and LDA. The meaning of the obtained LDA transformation matrix (reduced to a vector in our case) is not only treated as a black box but is also used to describe the posture information of PD patients in a numerical way.