Executive Summary : | Swarm intelligence algorithm is a sub-group of Artificial Intelligence, which is become popular in the combined intelligence nature of self-organization and has been extensively used in various applications. Recently, several swarm intelligence algorithms, namely Particle Swarm Optimization, Colony Optimization etc., have been developed to solve complex optimization problems. Fractional Calculus (FC) is a well-known quantitative analysis in the branch of mathematical analysis. These mathematical operations have been applied in both derivatives as well as integral functions to non-integer orders. Recently, the FC methods have been widely used in the field of biomedical signal and image applications. These methods deliver an efficient characterization of the practical approach in the real world than conventional integer order frameworks. Moreover, the non-integer derivatives, as well as integrals, allow modelling the system in precise memory representation and essential properties of all kinds of heterogeneous processes. The incorporation of fractional calculus and swarm intelligence algorithm can improve the computational capability of the system and give rise to techniques with higher generalization as well as learning capability. In this proposal, fractional order-based swarm intelligence algorithms are used to optimize the feature subsets to attain the global convergence properties using biceps brachii muscle electric signals. Further, the incorporation of conventional neural networks with fractional order-based swarm intelligence algorithms is highly suitable for inter-modal as well as uneven search space. This proposal necessitates computer-assisted analytical tools for the investigation of biceps brachii muscle electric signals such as extraction of relevant features, fractional order based swarm intelligence algorithms, and categorization of abnormalities in the EMG signals. |