An Artificial Intelligence System Focused on COVID-19 Pandemic: Results and Impacts

被引:0
作者
Pocho, Claudia Lopes [1 ]
de Carvalho, Marcelo [1 ]
Maciel Silveira, Carlos Frederico [2 ]
Marques, Ricardo Andre [1 ]
Rodrigues da Silva Quirino, Ana Claudia [1 ]
Werdine Machado, Luiz Humberto [1 ]
Pineiro, Marcelo Fernandez [1 ]
机构
[1] FURNAS Centrais Eletr SA, Rio De Janeiro, Brazil
[2] Cyberlabs Prod & Serv Tecnol SA, Rio De Janeiro, Brazil
来源
PROCEEDINGS OF THE 3RD EUROPEAN CONFERENCE ON THE IMPACT OF ARTIFICIAL INTELLIGENCE AND ROBOTICS (ECIAIR 2021) | 2021年
关键词
COVID-19; artificial intelligence; cloud computing; employee's COVID-19 risk score; self-assessment; facial recognition; LGPD law compliance; anti-ransomware;
D O I
10.34190/EAIR.21.040
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Artificial intelligence (AI) is rapidly becoming fundamental in our society, reaching from industry to our homes. AI benefits for human being lives are undoubtful, especially in its application on Health and Safety at Work. Specifically, when there is a high degree of uncertainty in preventive measures and protocols definition, such as the case of COVID-19, AI may enlighten and anticipate decisions saving lives. Therefore, this paper aims to discuss the success factors and the results of an AI application related to Covid-19 prevention in workplaces, named CyberLabs KeyApp. It presents a case study in the Brazilian energy sector applied to approximately 3000 employees located in more than 30 different cities. The system was composed by an application including Control Facial Recognition when employees reach the electric power plants, substations, and offices, conjugated with temperature control; a daily COVID-19 self-assessment form fulfilment gathering personal habits related to Covid-19 furthering the artificial intelligence processing; and surveillance cameras for agglomerations detecting on facilities. As a result, there is an AI COVID score calculation that classifies the employee's risk into four categories. Medical staff is notified with warns reporting each employee with High or Very High classification. The four categories are listed: Low - user pre-access to the facility is granted. Medium - user pre-access to the facility is granted and AI prescribes individual COVID avoidance good practices that are informed to the user by notification; High - user preaccess to the facility is conditioned to a prior individual medical staff evaluation and medical staff is notified for proceed to contact with the user; and Very High - user pre-access to the facility is not granted and the AI application notifies medical staff to contact immediately with the user. A set of operational, tactical, and strategic panels are available for managers and doctors for daily consulting. Through these AI dashboards, the variables - habits, symptoms, and preventive measures - that influences employees' Covid high risk are detected. Innovative Protocols were elaborated to diminish this risk rates as well as specific communication campaigns. Consequently, the company has reached the lowest Covid-19 rates among the Brazilian public energy sector institutions.
引用
收藏
页码:229 / 237
页数:9
相关论文
共 10 条
  • [1] Abu-Nasser BS, 2017, Int J Eng, V1, P218
  • [2] Classical and modern face recognition approaches: a complete review
    Ali, Waqar
    Tian, Wenhong
    Din, Salah Ud
    Iradukunda, Desire
    Khan, Abdullah Aman
    [J]. MULTIMEDIA TOOLS AND APPLICATIONS, 2021, 80 (03) : 4825 - 4880
  • [3] DeCaprio D, 2020, arXiv
  • [4] ArcFace: Additive Angular Margin Loss for Deep Face Recognition
    Deng, Jiankang
    Guo, Jia
    Xue, Niannan
    Zafeiriou, Stefanos
    [J]. 2019 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2019), 2019, : 4685 - 4694
  • [5] FDA (U.S. Food and Drugs Administration), 2021, THERMAL IMAGING SYST
  • [6] Hollister M, 2021, COVID 19 CAN HELP BU
  • [7] Application of Artificial Intelligence in COVID-19 Pandemic: Bibliometric Analysis
    Islam, Md. Mohaimenul
    Poly, Tahmina Nasrin
    Alsinglawi, Belal
    Lin, Li-Fong
    Chien, Shuo-Chen
    Liu, Ju-Chi
    Jian, Wen-Shan
    [J]. HEALTHCARE, 2021, 9 (04)
  • [8] 3DPC-Net: 3D Point Cloud Network for Face Anti-spoofing
    Li, Xuan
    Wan, Jun
    Jin, Yi
    Liu, Ajian
    Guo, Guodong
    Li, Stan Z.
    [J]. IEEE/IAPR INTERNATIONAL JOINT CONFERENCE ON BIOMETRICS (IJCB 2020), 2020,
  • [9] Obeidat S, 2021, ARTIF INTELL
  • [10] The role of artificial intelligence in healthcare: a structured literature review
    Secinaro, Silvana
    Calandra, Davide
    Secinaro, Aurelio
    Muthurangu, Vivek
    Biancone, Paolo
    [J]. BMC MEDICAL INFORMATICS AND DECISION MAKING, 2021, 21 (01)