ABSTRACT
This article discusses about the design and deployment of a smart robotic system on university campuses for monitoring the indoor environment, health protocols, and sanitation. The designed VEX autonomous robotic system performed the following tasks: (a) moving around the university classrooms and scanning the body temperature of students and staff, as well as tracking environmental parameters in classrooms;(b) executing sanitation function by disinfecting objects in classrooms;and (c) performing security function by sending an alert signal to health and safety officer if a student or staff with fever enters the classroom, or if staff or student is not wearing face mask indoors. Particle Photon microcontrollers linked to sensors and actuators were used to detect and manage indoor environmental conditions as well as track individuals' body temperatures from a distance, with the data being stored in the ThingSpeak and Particle cloud platforms and displayed on smartphone apps. Transfer learning through MIT App Inventor's Personal Image Classifier was used to detect health protocol violations with 93.33% accuracy. The maximum distance traversed by the robot prototype was 38 meters, with an average time of 220 seconds and an average speed of 0.17 meters per second. The robot had an 88.89% success rate in following the black-lined course. This intelligent robotic system can limit staff and student exposure to infectious diseases and implement "new normal"health and safety practices on campus as post-COVID-19 precautions. © 2023 IEEE.
ABSTRACT
This paper discusses the design and implementation of an Internet of Things (IoT)-based telemedicine health monitoring system (THMS) with an early warning scoring (EWS) function that reads, assesses, and logs physiological parameters of a patient such as body temperature, oxygen saturation level, systemic arterial pressure, breathing patterns, pulse (heart) rate, supplemental oxygen dependency, consciousness, and pain level using Particle Photon microcontrollers interfaced with biosensors and switches. The Mandami fuzzy inference-based medical decision support system (FI-MDSS) was also developed using MATLAB to assist medical professionals in evaluating a patient's health risk and deciding on the appropriate clinical intervention. The patient's physiological measurements, EWS, and health risk category are stored on the Particle cloud and Thing Speak cloud platforms and can be accessed remotely and in real-time via the Internet. Furthermore, a RESTful application programming interface (API) was developed using GO language and PostgreSQL database to enhance data presentation and accessibility. Based on the paired samples t-tests obtained from 6 sessions with 10 trials for each vital sign per session, there were no significant differences between the clinical data obtained from the designed prototype and the commercially sold medical equipment. The mean differences between the compared samples for each physiological data were not more than 0.40, the standard deviations were less than 2.3, and the p-values were greater than 0.05. With a 96.67% accuracy, the FI-MDSS predicted health risk levels that were comparable to conventional EWS techniques such as the Modified National Early Warning Score (m-NEWS) and NEWS2, which are used in the clinical decision-making process for managing patients with COVID-19 and other infectious illnesses. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.