Cognitive Analytics and Reinforcement Learning

9An Artificial Intelligent Methodology to Classify Knee Joint Disorder Using Machine Learning and Image Processing Techniques

M. Sharmila Begum¹, A. V. M. B. Aruna¹*, A. Balajee² and R. Murugan³

¹Department of Computer Science and Engineering, Periyar Maniammai Institute of Science & Technology, Deemed to be University, Thanjavur, India

²School of Computer Science and Engineering, Faculty of Engineering and Technology, Jain Deemed to be University, Bangalore, India

³School of Computer Science and IT, Jain Deemed to be University, Jayanagar, Bangalore, India

Abstract

Disorder diagnosis at an earlier stage plays a vital role in the prediction and classification of samples in clinical support systems. Machine learning methods are gaining major importance in classifying the samples from the given data. In this chapter, a hybridization of isolation forest is proposed to classify the normal and abnormal knee joint signal samples. The number of features extracted from the raw data could lead to the complexity of the model. To reduce the feature burden of the model, this chapter proposes an artificial intelligence methodology to select the abstract set of features from the given raw data, and classification is done through Hybrid Isolation Forest (HIF). The chapter consists of three phases, starting from processing the raw images and extraction of the preprocessed dataset. The second phase identifies the abstract feature set using the statistical and regressive parameters. The third ...