Abstract

COVID-19 is currently on the rage all over the world and has become a pandemic. To efficiently handle it, accurate diagnosis and prompt reporting are essential. The AI-Enabled Real-time Biomedical System (AIRBiS) research project aims to develop a system that handles diagnosis using chest X-ray images. The project is divided into UI, network, software and hardware. This work focuses on the hardware, which uses CNN technology to create a model that determines the presence of pneumonia. This CNN model is designed on an FPGA to speed up diagnostic results. The FPGA increases the flexibility of circuit design, allowing us to optimize the computational processing during data transfer and CNN implementation, reducing the diagnostic measurement time for a single image.

Highlights

  • At the dawn of 2020, the coronavirus disease (COVID19), which is induced by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) [1] became a global health crisis

  • The shortage and reliability issues associated with early testing kits needed for precise diagnosis, and the lack of coordinated systems needed for the quick response have prompted the need for an alternative method of diagnosing and responding to the crisis. the AI-Enabled Real-time Biomedical System (AIRBiS) [4] project

  • We use 5216 training sets and 624 test images, and the convolutional neural network (CNN) model was trained in grayscale

Read more

Summary

Introduction

At the dawn of 2020, the coronavirus disease (COVID19), which is induced by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) [1] became a global health crisis. The disease continues to spread and as of December 8, 2020, the number of confirmed global cases have risen to 66,422,058, and deaths to 1,532,418 as confirmed by the world health organization (WHO) [2]. The study in [8] proposed a conceptual framework to screen for COVID-19 by scanning chest CT images for pneumonia types between COVID-19 and intestinal lung disease. In [9], the authors proposed a weakly supervised Deep learning framework to detect the probability of COVID-19 using 3D CT volumes. The authors in [10] proposed a 3D densely connected convolutional neural network (CNN) to classify COVID-19 patients as either

AIRBiS Overview
Hardware Acceleration of AIRBiS CNN
Optimization by Pipelining
Optimization by Function combination
Optimization of circuit
Evaluation Methodology
Evaluation Results
Discussion
Conclusion

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call

Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.