Adaptive time-frequency analysis of knee joint vibroarthrographic signals for noninvasive screening of articular cartilage pathology

Vibroarthrographic (VAG) signals emitted by human knee joints are nonstatio nary and multicomponent in nature; time-frequency distributions (TFD's) pro vide powerful means to analyze such signals. The objective of this paper is to construct adaptive TFD's of VAG signals suitable for feature extraction . An adaptive TFD was constructed by minimum cross-entropy optimization of the TFD obtained by the matching pursuit decomposition algorithm. Parameter s of VAG signals such as energy, energy spread, frequency, and frequency sp read were extracted from their adaptive TFD's, The parameters carry informa tion about the combined TF dynamics of the signals, The mean and standard d eviation of the parameters were computed, and each VAG signal was represent ed by a set of just six features. Statistical pattern classification experi ments based on logistic regression analysis of the parameters showed an ove rall normal/abnormal screening accuracy of 68.9% with 90 VAG signals (51 no rmals and 39 abnormals), and a higher accuracy of 77.5% with a database of 71 signals with 51 normals and 20 abnormals of a specific type of patellofe moral disorder. The proposed method of VAG signal analysis is independent o f joint angle and clinical information, and shows good potential for noninv asive diagnosis and monitoring of patellofemoral disorders such as chondrom alacia patella.