Artificial Intelligence-Enabled Blockchain Technology and Digital Twin for Smart Hospitals

15Unveiling the Challenges and Limitations in COVID-19 Health Data Prediction with Convolutional Neural Networks: A Data Science Research Perspective

Asadi Srinivasulu^1*, Piyush Agrawal², Amit Agrawal¹ and Goddindla Sreenivasulu³

¹Department of CSE, Data Science Research Laboratory, Sree Datta Group of Educational Institutions, Hyderabad, Telangana, India

²Computer Science Department, Devi Ahilya Vishwavidyalaya, Indore, Madhya Pradesh, India

³Department of Biotechnology, Prathyusha Engineering College, Thiruvallur, Tamil Nadu, India

Abstract

The COVID-19 pandemic has instigated an urgent demand for precise and effective prediction models to assist in disease management and containment. In recent years, convolutional neural networks (CNNs) have emerged as potent instruments for analyzing health data and forecasting COVID-19–related outcomes. Nevertheless, the primary objective of this research is to systematically investigate and elucidate the challenges and limitations associated with the utilization of CNNs in COVID-19 health data prediction, providing an exhaustive perspective from the realm of data science research. The study delves into diverse data-related issues, model architecture constraints, and concerns regarding generalization that may impact the efficacy of CNNs in predicting COVID-19 outcomes. By illuminating these challenges, this research endeavors to offer guidance to researchers and practitioners in making well-informed decisions while employing CNNs for ...