key: cord-0043645-z9kwedr7
authors: Pietroiusti, Antonio; Uva, Antonio Emmanuele; Cascella, Giuseppe Leonardo; Toschi, Nicola
title: COVID-19: contact and gesture monitoring using PROUD Technology
date: 2020-05-12
journal: Occup Med (Lond)
DOI: 10.1093/occmed/kqaa083
sha: 78de0cf09070b1fa0424ad0170fcf9c9869f48af
doc_id: 43645
cord_uid: z9kwedr7

nan

The main pathways by which COVID-19 can spread are droplets (i.e. spreading by inhalation) and surfaces on which the virus has been deposited [1] , where hand-face contact after touching a contaminated surface may cause transmission. This surface-mediated pathway is important in the workplace in general, and among healthcare workers in particular, especially during the pre-symptomatic stage and in asymptomatic infection [2] . For these reasons, handwashing is paramount [3] . However, hand contamination at work may occur through a number of basic activities (e.g. opening doors or operating light switches), potentially resulting in an almost unlimited requirement for handwashing.

The main factor in surface-mediated spread is selftouching of the peri-oral and peri-nasal area; without this, infection cannot occur. However, suggesting avoiding face-touching is simple in theory, but difficult in practice. We all touch our faces many times a day-a social and cultural habit acquired and engrained early in life. While the severe consequences of the COVID-19 pandemic may generate enough motivation to modify these habits, such deep-rooted habits will probably not change without monitoring and guidance.

We are currently expanding a technology intended to track people's position and behaviour using the PROUD (Pandemic Reduction by indoor & OUtDoor tracking) framework. PROUD is currently providing: (i) proximity time and occurrence (distance less than, e.g., 1 m between individuals); (ii) presence, time and occurrence of users in critical areas (e.g. toilets or disinfecting areas) or in other high-risk areas (e.g. COVID hubs). PROUD is currently being expanded to provide: (a) time and occurrence for key events (e.g. hand-face self-contact, handshaking, handwashing), (b) real-time zoning of interpersonal distance [4] : intimate (1-46 cm), personal (47-122 cm), social (1.2-3.7 m) and public space (>3.7 m). This will be based on sensor interaction between wearable devices (e.g. a ring or a bracelet) and a sensorized environment (e.g. beacons or depth cameras). Downstream from tracking, a simple signal (sound or device vibration) will alert the worker of risky behaviour (e.g. facetouching) or risk accumulation (e.g. frequent violation of distance zones, below-average handwashing). It is hoped that, similar to step-counting, providing this information could induce a virtuous cycle.

Anonymous data will be archived for centralized predictive analysis (e.g. identifying a critical threshold beyond which infection risk is significantly increased) and policy drafting by management and, possibly, regulators. It is also likely that significant gesturemovement patterns are still hidden in the data provided by wearables, mobiles and sensorized environments. PROUD includes artificial intelligence-based dataharvesting strategies in order to extract further reliable, effective key performance indicators to be used as alerts or in decision support systems. The framework may also assist in managing possible future epidemics, including annual influenza.

Aerosol and surface stability of SARS-CoV-2 as compared with SARS-CoV-1

Asymptomatic transmission, the Achilles' heel of current strategies to control Covid-19

Consensus of Chinese experts on protection of skin and mucous membrane barrier for health-care workers fighting against coronavirus disease 2019

A system for the notation of proxemic behavior

Nicola Toschi e-mail: pietroiu@uniroma2.it