Classification of asymptomatic and osteoarthritic knee gait patterns using gait analysis via deterministic learning
Gait measures have received increasing attention in the evaluation of patients with knee osteoarthritis (OA). Comprehending gait parameters is an essential requirement for studying the causes of knee disorders. The aim of this work is to develop a new method to distinguish between asymptomatic (AS) and osteoarthritic knee gait patterns using gait analysis via deterministic learning. Spatiotemporal parameters and three-dimensional knee joint rotations and translations are measured and compared in 19 patients with knee OA and 28 AS control subjects during level walking. The classification approach consists of two stages: a training stage and a classification stage. In the training stage, gait features representing gait dynamics, including knee rotations and translations, are derived from the kinematic data of the knees in six-degree-of-freedom. Gait dynamics underlying gait patterns of AS control subjects and patients with knee OA are locally accurately approximated by radial basis function (RBF) neural networks. The obtained knowledge of approximated gait dynamics is stored in constant RBF networks. Gait patterns of AS control subjects and patients with knee OA constitute a training set. In the classification stage, a bank of dynamical estimators is constructed for all the training gait patterns. Prior knowledge of gait dynamics represented by the constant RBF networks is embedded in the estimators. By comparing the set of estimators with a test knee OA gait pattern to be classified, a set of classification errors are generated. The average \(L_1\) norms of the errors are taken as the classification measure between the dynamics of the training gait patterns and the dynamics of the test knee OA gait pattern according to the smallest error principle. Finally, experiments are carried out to demonstrate that the proposed method can effectively separate the gait patterns between the groups of AS control subjects and patients with knee OA. By using the two-fold cross-validation and leave-one-out cross-validation styles, the correct classification rates for knee OA gait patterns are reported to be 95.7 and 97.9%, respectively.
KeywordsGait analysis Knee osteoarthritis (OA) Movement disorders Neural networks
This work was supported by the National Natural Science Foundation of China (Grant nos. 61773194, 61304084), by the Natural Science Foundation of Fujian Province of China (Grant no. 2018J01542) and by the Program for New Century Excellent Talents in Fujian Province University.
Compliance with ethical standards
Conflict of interest
Authors declare that they have no conflict of interest.
- Chang AH, Moisio KC, Chmiel JS, Eckstein F, Guermazi A, Prasad PV, Zhang Y, Almagor O, Belisle L, Hayes K, Sharma L (2015) External knee adduction and flexion moments during gait and medial tibiofemoral disease progression in knee osteoarthritis. Osteoarthr Cartil 23(7):1099–1306CrossRefGoogle Scholar
- Forman MD, Malamet R, Kaplan D (1983) A survey of osteoarthritis of the knee in the elderly. J Rheumatol 10(2):282–287Google Scholar
- Murary MP, Drought AB, Kory RC (1967) Walking pattern of movement. Am J Med 46(1):290–332Google Scholar
- Wang C, Hill DJ (2009) Deterministic learning theory for identification, recognition and control. CRC Press, Boca RatonGoogle Scholar