Conventional treatment methods make even the most basic healthcare issues more complicated, which in turn increases the number of parties involved. Classical computing lacks the speed and accuracy needed for effective stakeholder collaboration in COVID-19 healthcare solutions, such as patients, insurance agents, healthcare practitioners, pharmaceutical suppliers, etc. The research uses organizational information processing theory (OIPT) to examine how quantum computing which is applications of artificial intelligence (AI) could transform the healthcare business, creating a more sustainable and less burdened system. The study of quantum computing (QC) has the potential to bring about “quantum leaps,” which might have unforeseen consequences for healthcare. The discovery of new medications, the personalization of medicinal treatments, and the acceleration of DNA sequencing are just a few of the many possible applications of this method. The potential of QC to transform compute-intensive healthcare tasks like drug-discovery, personalized-medicine, DNA-sequencing, medical-imaging, and operational-optimization is the primary focus of this survey paper, which offers the first comprehensive analysis of QCs diverse capabilities in improving healthcare systems. After a thorough literature study, we created taxonomies on the healthcare QC paradigm’s history and supporting technologies, applications, needs, architectures, security, outstanding questions, and future research prospects. We hope that by conducting this survey, researchers with varying levels of experience in quantum computing and healthcare will better understand the state of the art, assess opportunities and threats, and make informed decisions as they develop novel architectures and applications for this emerging field.

Human Action Recognition is considered to be a critical problem and it is always a challenging issue in computer vision applications, especially video surveillance applications. State-of-the-art classifiers introduced to solve the problem are computationally expensive to train and require very large amounts of data. In this paper, we solve the problems of low data and resource availability in surveillance datasets by employing transfer learning and fine-tuning the Inflated 3D CNN model and the SlowFast Network model to automatically extract features from surveillance videos in the SPHAR dataset for classification into respective action classes. This approach works well to process the spatio-temporal nature of videos. Fine-tuning is carried out in the networks by replacing the last classification (dense) layer as per the available number of classes in the constructed new dataset. We ultimately compare the performance of both fine-tuned networks by taking accuracy as the metric, and find that the I3D model performs better for our use-case.