correspondence Research priorities for COVID-19 sensor technology To the Editor — The COVID-19 pandemic This priority-setting project involved 83 access to treatment, preserve finite resources has spurred efforts to develop sensor patients with COVID-19, family members, (in terms of staffing, facilities for quarantine technology to manage the disease1–3. Most the general public, scientists, engineers, and personal protective equipment) and of these projects have been driven by health professionals (including specialist to provide prognostic information to medical researchers, scientists and engineers clinicians from multiple disciplines, inform patient care. Second, ‘Minimizing without explicit involvement and input such as infectious diseases, diagnostic societal disruption’ was emphasized to from patients and the broader community. pathology, cardiology or cardiovascular enable a return to normal life and to reduce Here, we define sensor technology broadly diseases, respiratory medicine, geriatrics, stigma and isolation. Third, ‘Protecting to include physical, cellular and molecular emergency medicine, critical care medicine, the community’ supported the need for platforms that produce signals to identify gastroenterology, hematology, pediatrics, sensor technology that could trigger contact specific events associated with SARS-CoV-2 infection prevention and control, and tracing, establish safe environments, and/or its interaction with the host. The digital health), policy makers, industry safeguard the vulnerable, gauge individual main applications of sensor technology representatives and funders. susceptibility to COVID-19, and manage the in COVID-19 have been to detect fever We conducted an online survey to risk among healthcare workers. And finally, using infrared sensing devices and the prioritize research statements in which ‘Preparedness for the next phase of the presence of viral RNA using polymerase respondents (n = 43) rated their importance pandemic’ required sensor technology to be chain reaction (PCR) tests1. However, a using a 9-point Likert scale (7–9 indicating relevant and responsive to the development substantial proportion of individuals with ‘critical importance’). The mean score, of immunity and vaccines, and to help COVID-19 never develop fever1. PCR tests median and proportion of participants who maintain the suppression phase over the have been developed to detect SARS-CoV-2 rated the statement to be critically important long term. A detailed description with in nasopharyngeal samples, but to date they are provided in Supplementary Table 1. supporting quotations for each theme is have been expensive, resource-intensive, Research statements that had a mean and provided in Supplementary Table 3. cumbersome and relatively slow. Moreover, median of ≥7 were discussed at a consensus For each of the top research priority positive PCR tests do not imply a person is workshop, conducted using Zoom statements, the specific suggestions for still infectious and thus have not provided videoconferencing on 20 August 2020, with sensor technology (including compounds information about transmissibility or the following goals: to achieve agreement and devices) and its application are virulence1,4,5, hampering the development of on the research priorities, generate ideas summarized in Table 1. The suggestions more effective action plans in the societal, for sensor technologies and discuss of ensuring feasibility, usability and economic and public health dimensions6. facilitators and barriers to implementation. acceptability of sensor technology and Given the urgent need to better control To encourage diverse discussions, the 65 applications to address COVID-19 are the pandemic and its impact on the attendees were preassigned to six virtual outlined in Supplementary Table 4. These community, resources should be allocated breakout groups, with each group including have been identified as essential attributes in a strategic and targeted manner that patients who had been diagnosed previously for an ideal sensor for pandemics in general, takes into account community perspectives, with COVID-19 and/or family members, including accuracy, a fast response time, through an explicit consensus-based process health professionals, scientists or engineers, multiplexing capabilities, multiple sensing with equitable involvement of patients, and policymakers or funders. Each breakout modes (sensor fusion and the use of artificial the public, researchers and clinicians. group was managed by a facilitator and intelligence to detect signatures that reveal Co-production in research specifically cofacilitator who moderated the discussion infection), disposability, long shelf life, ease involving consumers or end-users is now using the workshop question guide of use, cost-effectiveness, manufacturability, widely advocated to improve the relevance, (Supplementary Table 2). All discussions and autonomy2. Particular emphasis was use and impact of the findings7,8. It requires were transcribed. We identified reasons for placed on the need for samples to be easy partnership and collaboration between the priorities, ideas for sensor technologies and safe to collect and the need for sensor researchers and the broader community from (compounds, devices, general application), devices to be non-invasive and their use the outset, beginning with priority-setting8. and the implementation of each (feasibility, regulated appropriately to ensure data There have been few research priority-setting usability, acceptability). privacy. The legal, ethical and privacy partnerships in COVID-19, with very few Of the 18 research statements, 8 had concerns surrounding the use of digital involving patients and the public, and none a mean and median score of 7 or more technology in COVID-19 are highly with a focus on sensor technology. Below, (Table 1). The top three priorities were the relevant given the need for public trust and we describe the development and outcome following: develop a point-of-care screening engagement to ensure widespread uptake1. of a process through which we identified the test for COVID-19; detect how contagious Patients in particular emphasized the shared priorities of patients, the community, a person with COVID-19 is; and identify profound impacts of COVID-19 on mental health professionals, scientists, engineers the level of immunity a person has to health as a consequence of self-isolation and and policy makers for research in sensor COVID-19. The reasons for priorities were quarantine. Specifically, patients gave high technology to address COVID-19, the summarized in four themes. First, ‘Enabling priority to the detection of immunity and reasons for their priorities, and ideas for more efficient clinical decision-making’ wanted assurance that they were no longer implementation. was driven by the need to prevent delays in contagious because families and friends were 144 NATURE BIOTECHNOLOGY | VOL 39 | FEBRUARY 2021 | 144–153 | www.nature.com/naturebiotechnology correspondence Table 1 | Suggestions and ideas for sensor technology to address COVID-19 Statement Ideas for compounds, devices and general application Develop a point-of-care (instant) • Target a different type of sample or organ (other than blood, nasal or throat swab, or temperature), for example, screening test for COVID-19 urine (non-aerosol-generating procedure) • Detect other compounds or chemicals that the body generates due to infection and that may be used as a signature of disease • Detect viral components • Use microfluidic technology: capture particles on a microfluidic system on a chip; they go to the optic sensor and the optic sensor functions as a screener • Develop wearable non-intrusive devices for healthcare workers that capture temperature and other clinical parameters, with data captured in a central service for monitoring, and use artificial intelligence to analyze the data • Create a device used in the toilet to measure microbiome, with data connected to a phone or other device; sample viral loads in sewage • Support rapid transmission of data packets • F or individual screening, offer conventional methods such as a strip and a reader, with results transmitted to a central location • Develop a dipstick test using saliva • Analyze physiological or biomarker responses (similarly to browsers that assess whether people are robots) with artificial intelligence machine learning • Use a microphone to detect characteristics of breathing • Provide a device to monitor hypoxia • Develop a device to monitor the ability to smell, olfactory type of sensing • Provide a device to monitor heart rate • Use sensors to detect inflammation in the mucosal tract • Develop a mask with breath analysis to signal infection Detect how infectious a person with • Develop face masks with sensors (for example, one that changes color if a person is contagious) a virus is • Use sensors to measure breathing, cough, inflammation • Develop a tongue swab to determine viral load on site • Develop cell cultures with cell types that are very susceptible infection by the virus, and expose them to infected people to measure transmissibility • Measure the aerosol and droplet release that come from talking, sneezing, coughing, perspiration; quantify the particle release from a person and their interaction with other people nearby • Use digital imaging processing to detect particle exchange Identify the level of immunity a • Develop a saliva test that uses spike protein as a capture medium for immunoglobulin