Spiral Optimization Algorithm with Neurocomputing and Fuzzy Logic Based Human Gait Analysis for Osteoarthritis Prediction

Due to the latest developments in smart sensors, advanced computational competencies of devices across layers have become possible to design intelligent decision-making tools in automation and robotics. Human gait analysis (HGA) becomes a hot research topic among computer vision (CV) communities, which can be used in the healthcare sector for decision-making. The advances in deep learning (DL) and neurocomputing models can be employed to design robust HGA models for disease prediction. The study presents a Spiral Optimization Algorithm with Neurocomputing and Fuzzy Logic Based Human Gait Analysis for Osteoarthritis Prediction (SPONCFL-HGA) technique. The proposed SPONCFL-HGA technique aims to examine and classify human gaits for osteoarthritis prediction. The proposed SPONCFL-HGA technique employs a spiral optimization algorithm (SOA) with the SqueezeNet model for the extraction of features effectively. In addition, the adaptive fuzzy inference system (ANFIS) was utilized for the classification of human gaits. Finally, the wild horse optimization (WHO) methodology can be applied for tuning the parameters included in the ANFIS. A wide range of experiments are implemented, and the experimental outcomes are scrutinized under various measures. The simulation outcome demonstrates the significant performance of the SPONCFL-HGA methodology over other recent approaches.