ABSTRACT
Deep learning based models on the edge devices have received considerable attention as a promising means to handle a variety of AI applications. However, deploying the deep learning models in the production environment with efficient inference on the edge devices is still a challenging task due to computation and memory constraints. This paper proposes a framework for the service robot named GuardBot powered by Jetson Xavier NX and presents a real-world case study of deploying the optimized face mask recognition application with real-time inference on the edge device. It assists the robot to detect whether people are wearing a mask to guard against COVID-19 and gives a polite voice reminder to wear the mask. Our framework contains dual-stage architecture based on convolutional neural networks with three main modules that employ (1) MTCNN for face detection, (2) our proposed CNN model and seven transfer learning based custom models which are Inception-v3, VGG16, denseNet121, resNet50, NASNetMobile, XceptionNet, MobileNet-v2 for face mask classification, (3) TensorRT for optimization of all the models to speedup inference on the Jetson Xavier NX. Our study carries out several analysis based on the models' performance in terms of their frames per second, execution time and images per second. It also evaluates the accuracy, precision, recall & F1-score and makes the comparison of all models before and after optimization with a main focus on high throughput and low latency. Finally, the framework is deployed on a mobile robot to perform experiments in both outdoor and multi-floor indoor environments with patrolling and non-patrolling modes. Compared to other state-of-the-art models, our proposed CNN model for face mask recognition based on the classification obtains 94.5%, 95.9% and 94.28% accuracy on training, validation and testing datasets respectively which is better than MobileNet-v2, Xception and InceptionNet-v3 while it achieves highest throughput and lowest latency than all other models after optimization at different precision levels.
ABSTRACT
Introduction: The global pandemic of the coronavirus 2019 disease (COVID-19) is caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). In addition to respiratory failures, COVID-19 patients exhibited cardiac complications. Studies observed the direct infection and replication of SARS-CoV2 in human pluripotent stem cell-derived cardiomyocytes (hPSC-CMs) accompanied by cytopathic effects. However, the underlying mechanisms of SARS-CoV-2-mediated CM death remain poorly understood. In addition, the therapeutic potential of remdesivir (RDV) on CMs has yet to be answered. Methods and Results: We confirmed that SARS-CoV-2 is infectious to and effectively replicates in hPSC-CMs and is cytopathic to hPSC-CMs. We also found that RDV effectively inhibited viral replication at a concentration of 50 nM. RNA-seq analyses demonstrated that expression of immune responsive genes was elevated in SARS-CoV-2 infected hPSC-CMs. Immunostaining and an ELISA assay further revealed formation of inflammasomes and secretion of inflammasome-mediated cytokines, such as IL-1β, IL-18, and IL-6 in SARS-CoV-2 infected hPSC-CMs. RNA-seq analyses showed gene profile changes in SARS-CoV-2 infected hPSC-CMs corroborating with activation of inflammatory signals and cell death pathways. While gene profiles of 0.1 μM RDV-treated SARS-CoV-2-infected hPSC-CMs showed reversal of such changes, a high dose (10 μM) RDV-treated CoV-2-infected hPSC-CMs showed changes in 44% of genes expressed compared to non-RDVtreated CoV2-infected hPSC-CMs. Among those, expression of protein stability related genes, such as genes associated with autophagy and protein ubiquitination increased while expression of antiviral responsive genes decreased. In addition, a high dose of RDV inhibited expression of mitochondrial genes, particularly MitoComplex I and V compositions, which are related to energy production. Conclusions: This study demonstrates that SARS-CoV2 induced inflammasome in hPSC-CMs, which can underlie cardiac damage in addition to direct cytopathic effects. In addition, RDV can reduce inflammasome when introduced early after SARS-CoV2 infection while a high-dose can aggravate cytopathic effects by potential toxicity to mitochondria.