Smart healthcare system using machine learning and IoT
Raj Gusain, Sushant Shekhar, Anurag Vidyarthi, Rıshı Prakash, R. Gowri · 2024
The COVID-19 pandemic has emphasized the urgency of effective vaccine distribution. Conventional methods of allocation face challenges in accurately verifying recipient eligibility in real time. This chapter introduces an innovative solution by integrating embedded systems, machine learning, and the Internet of Things (IoT) to streamline vaccine delivery. Leveraging real-time Twitter data, our system aims to verify user eligibility promptly and accurately, ensuring vaccines reach rightful recipients. Our interdisciplinary approach combines to address the intricate task of verifying vaccine recipients. Twitter, as a prolific source of real-time information, proves invaluable for user verification. We collect and analyze vaccination-related tweets, focusing on user-generated content that may contain pertinent information about vaccine recipients. A bidirectional long short-term memory (LSTM) network at the center of our system is an effective means for processing sequential data, which includes text written using natural language. The model has an in-depth understanding of context and can make precise predictions about user eligibility. To enable timely decision-making in vaccine allocation, we embed the model into a microcontroller unit. This integration facilitates immediate verification based on incoming tweet data. IoT devices ensure seamless communication between the microcontroller and the external environment, expediting user verification and subsequent vaccine allocation. The model demonstrates an impressive accuracy rate of 95.2% in user verification. Moreover, the receiver operating characteristic (ROC) curve showcases an area under the curve (AUC) value of 0.94, underscoring the model&s;s discriminatory power. This research not only advances vaccine distribution methodologies but also serves as a blueprint for leveraging technology in healthcare initiatives, especially in times of crisis. Future enhancements may further refine this system, potentially revolutionizing how we approach healthcare and emergency response on a global scale.