Machine learning enabled detection of COVID-19 pneumonia using exhaled breath analysis: a proof-of-concept study.

Detection of the severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) relies on real-time-reverse-transcriptase polymerase chain reaction (RT-PCR) on nasopharyngeal swabs. The false-negative rate of RT-PCR can be high when viral burden and infection is localized distally in the lower airways and lung parenchyma. An alternate safe, simple and accessible method for sampling the lower airways is needed to aid in the early and rapid diagnosis of COVID-19 pneumonia. In a prospective unblinded observational study, patients admitted with a positive RT-PCR and symptoms of SARS-CoV-2 infection were enrolled from three hospitals in Ontario, Canada. Healthy individuals or hospitalized patients with negative RT-PCR and without respiratory symptoms were enrolled into the control group. Breath samples were collected and analyzed by laser absorption spectroscopy (LAS) for volatile organic compounds (VOCs) and classified by machine learning (ML) approaches to identify unique LAS-spectra patterns (breathprints) for SARS-CoV-2. Of the 135 patients enrolled, 115 patients provided analyzable breath samples. Using LAS-breathprints to train ML classifier models resulted in an accuracy of 72.2%-81.7% in differentiating between SARS-CoV2 positive and negative groups. The performance was consistent across subgroups of different age, sex, body mass index, SARS-CoV-2 variants, time of disease onset and oxygen requirement. The overall performance was higher than compared to VOC-trained classifier model, which had an accuracy of 63%-74.7%. This study demonstrates that a ML-based breathprint model using LAS analysis of exhaled breath may be a valuable non-invasive method for studying the lower airways and detecting SARS-CoV-2 and other respiratory pathogens. The technology and the ML approach can be easily deployed in any setting with minimal training. This will greatly improve access and scalability to meet surge capacity; allow early and rapid detection to inform therapy; and offers great versatility in developing new classifier models quickly for future outbreaks.

White Paper on Early Critical Care Services in Low Resource Settings.

This White Paper has been formally accepted for support by the International Federation for Emergency Medicine (IFEM) and by the World Federation of Intensive and Critical Care (WFICC), put forth by a multi-specialty group of intensivists and emergency medicine providers from low- and low-middle-income countries (LMICs) and high-income countries (HiCs) with the aim of 1) defining the current state of caring for the critically ill in low-resource settings (LRS) within LMICs and 2) highlighting policy options and recommendations for improving the system-level delivery of early critical care services in LRS. LMICs have a high burden of critical illness and worse patient outcomes than HICs, hence, the focus of this White Paper is on the care of critically ill patients in the early stages of presentation in LMIC settings. In such settings, the provision of early critical care is challenged by a fragmented health system, costs, a health care workforce with limited training, and competing healthcare priorities. Early critical care services are defined as the early interventions that support vital organ function during the initial care provided to the critically ill patient-these interventions can be performed at any point of patient contact and can be delivered across diverse settings in the healthcare system and do not necessitate specialty personnel. Currently, a single "best" care delivery model likely does not exist in LMICs given the heterogeneity in local context; therefore, objective comparisons of quality, efficiency, and cost-effectiveness between varying models are difficult to establish. While limited, there is data to suggest that caring for the critically ill may be cost effective in LMICs, contrary to a widely held belief. Drawing from locally available resources and context, strengthening early critical care services in LRS will require a multi-faceted approach, including three core pillars: education, research, and policy. Education initiatives for physicians, nurses, and allied health staff that focus on protocolized emergency response training can bridge the workforce gap in the short-term; however, each country's current human resources must be evaluated to decide on the duration of training, who should be trained, and using what curriculum. Understanding the burden of critical Illness, best practices for resuscitation, and appropriate quality metrics for different early critical care services implementation models in LMICs are reliant upon strengthening the regional research capacity, therefore, standard documentation systems should be implemented to allow for registry use and quality improvement. Policy efforts at a local, national and international level to strengthen early critical care services should focus on funding the building blocks of early critical care services systems and promoting the right to access early critical care regardless of the patient's geographic or financial barriers. Additionally, national and local policies describing ethical dilemmas involving the withdrawal of life-sustaining care should be developed with broad stakeholder representation based on local cultural beliefs as well as the optimization of limited resources.

Characteristics and outcomes of patients with COVID-19 admitted to hospital and intensive care in the first phase of the pandemic in Canada: a national cohort study.

BACKGROUND: Clinical data on patients admitted to hospital with coronavirus disease 2019 (COVID-19) provide clinicians and public health officials with information to guide practice and policy. The aims of this study were to describe patients with COVID-19 admitted to hospital and intensive care, and to investigate predictors of outcome to characterize severe acute respiratory infection. METHODS: This observational cohort study used Canadian data from 32 selected hospitals included in a global multisite cohort between Jan. 24 and July 7, 2020. Adult and pediatric patients with a confirmed diagnosis of COVID-19 who received care in an intensive care unit (ICU) and a sampling of up to the first 60 patients receiving care on hospital wards were included. We performed descriptive analyses of characteristics, interventions and outcomes. The primary analyses examined in-hospital mortality, with secondary analyses of the length of hospital and ICU stay. RESULTS: Between January and July 2020, among 811 patients admitted to hospital with a diagnosis of COVID-19, the median age was 64 (interquartile range [IQR] 53-75) years, 495 (61.0%) were men, 46 (5.7%) were health care workers, 9 (1.1%) were pregnant, 26 (3.2%) were younger than 18 years and 9 (1.1%) were younger than 5 years. The median time from symptom onset to hospital admission was 7 (IQR 3-10) days. The most common symptoms on admission were fever, shortness of breath, cough and malaise. Diabetes, hypertension and cardiac, kidney and respiratory disease were the most common comorbidities. Among all patients, 328 received care in an ICU, admitted a median of 0 (IQR 0-1) days after hospital admission. Critically ill patients received treatment with invasive mechanical ventilation (88.8%), renal replacement therapy (14.9%) and extracorporeal membrane oxygenation (4.0%); 26.2% died. Among those receiving mechanical ventilation, 31.2% died. Age was an influential predictor of mortality (odds ratio per additional year of life 1.06, 95% confidence interval 1.03-1.09). INTERPRETATION: Patients admitted to hospital with COVID-19 commonly had fever, respiratory symptoms and comorbid conditions. Increasing age was associated with the development of critical illness and death; however, most critically ill patients in Canada, including those requiring mechanical ventilation, survived and were discharged from hospital.