Notice of Violation of IEEE Publication Principles: Affirmative Fusion Process for Improving Wearable Sensor Data Availability in Artificial Intelligence of Medical Things
Priyan Malarvizhi Kumar, Latif U. Khan, Choong Seon Hong
Abstract
Notice of Violation of IEEE Publication Principles <br><br>"Affirmative Fusion Process for Improving Wearable Sensor Data Availability in Artificial Intelligence of Medical Things," <br> by P. M. Kumar, L. U. Khan and C. S. Hong, <br>in IEEE Sensors Journal, Early Access <br><br> After careful and considered review of the content and authorship of this paper by a duly constituted expert committee, this paper has been found to be in violation of IEEE’s Publication Principles. <br><br>The submitting author, Priyan Malarvizhi Kumar, added the coauthors Latif U. Khan and Choong Seon Hong without their consent. Due to the nature of this violation, the Editor in Chief has decided the article will not be published in an issue of IEEE Sensors Journal. <br><br> <br/> Artificial Intelligence of Medical Things (AIoMT) is a hybridized outcome of Internet of Things (IoT), machine learning (ML) paradigms, and data analytics procedures for sophisticated healthcare services and applications. However, the fluctuating or lacking wearable sensors (WSs) data cause trivial computing errors that lead to incomplete diagnosis/ recommendation in healthcare applications. This article proposes a novel Affirmative Fusion Process (AFP) to enable high quality WS data with fewer fluctuations in in medical diagnosis. The proposed process assimilates sensed data with the existing datasets for avoiding discrete availability of WS data during the analysis. In this fusion process, based on the dataset inputs, the discreteness in the sensed data is identified. The discreteness is mitigated through precise replacement consideration from the existing datasets, preventing computational errors. The fusion process is monitored using simulated annealing and neural learning for output approximation and identification. The fused output with and without discreteness is identified for which annealing-based approximation is performed. In this process, the recurrence of the learning iterates is confined to identifying the final best solution. The proposed process is assessed using an activity dataset for the metrics fusion ratio, time delay, complexity, and data availability.