Revolutionizing Pandemic Healthcare: Mask Detection and Patient Face Recognition

During the COVID-19 pandemic, Delhi, India, faced a pressing issue where approximately 1,500 COVID-19-positive patients went missing. In public health emergencies, such as pandemics, natural disasters, or other calamities, hospitals, and medical centres experience a sudden influx of patients, and hospital management faces difficulties in keeping track of patients, especially when they need to be moved between facilities or when new temporary healthcare facilities are set up. As a consequence of these challenges, there can be an increase in missing person cases. Patients may be inadvertently separated from their families. The human face is a unique biometric system that can determine the age, gender, mood, of an individual, and even identity for verification purposes. Harbouring the power of deep learning and artificial intelligence, one of the most important applications of computer vision is Patient Identification. In this study, we have proposed a state-of-the-art patient face detection model using a twofold model that uses MTCNN short for “multi-task cascaded convolution neural network” for face detection and alignment purposes with a FaceNet Convolutional Neural Network (CNN) a renowned face embedding algorithm finally with KNN algorithm as a classifier to get an accuracy of 97.1%. Also, to ensure public safety during the pandemic we have constituted a Resnet34 model for mask detection trained on the Face Mask Detection dataset with an accuracy of 97%. This study not only addresses the immediate challenges of patient identification and safety during crises but also carries implications for broader healthcare applications. The proposed models offer promising avenues for enhancing patient care and security.