Artificial intelligence-enabled rapid diagnosis of patients with COVID-19

Imaging Diagnostics and Pathology in SARS-CoV-2-Related Diseases

In December 2019, physicians reported numerous patients showing pneumonia of unknown origin in the Chinese region of Wuhan. Following the spreading of the infection over the world, The World Health Organization (WHO) on 11 March 2020 declared the novel severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) outbreak a global pandemic. The scientific community is exerting an extraordinary effort to elucidate all aspects related to SARS-CoV-2, such as the structure, ultrastructure, invasion mechanisms, replication mechanisms, or drugs for treatment, mainly through in vitro studies. Thus, the clinical in vivo data can provide a test bench for new discoveries in the field of SARS-CoV-2, finding new solutions to fight the current pandemic. During this dramatic situation, the normal scientific protocols for the development of new diagnostic procedures or drugs are frequently not completely applied in order to speed up these processes. In this context, interdisciplinarity is fundamental. Specifically, a great contribution can be provided by the association and interpretation of data derived from medical disciplines based on the study of images, such as radiology, nuclear medicine, and pathology. Therefore, here, we highlighted the most recent histopathological and imaging data concerning the SARS-CoV-2 infection in lung and other human organs such as the kidney, heart, and vascular system. In addition, we evaluated the possible matches among data of radiology, nuclear medicine, and pathology departments in order to support the intense scientific work to address the SARS-CoV-2 pandemic. In this regard, the development of artificial intelligence algorithms that are capable of correlating these clinical data with the new scientific discoveries concerning SARS-CoV-2 might be the keystone to get out of the pandemic.

Discrimination of pulmonary ground-glass opacity changes in COVID-19 and non-COVID-19 patients using CT radiomics analysis

PURPOSE

The coronavirus disease 2019 (COVID-19) has evolved into a worldwide pandemic. CT although sensitive in detecting changes suffers from poor specificity in discrimination from other causes of ground glass opacities (GGOs). We aimed to develop and validate a CT-based radiomics model to differentiate COVID-19 from other causes of pulmonary GGOs.

METHODS

We retrospectively included COVID-19 patients between 24/01/2020 and 31/03/2020 as case group and patients with pulmonary GGOs between 04/02/2012 and 31/03/2020 as a control group. Radiomics features were extracted from contoured GGOs by PyRadiomics. The least absolute shrinkage and selection operator method was used to establish the radiomics model. We assessed the performance using the area under the curve of the receiver operating characteristic curve (AUC).

RESULTS

A total of 301 patients (age mean ± SD: 64 ± 15 years; male: 52.8 %) from three hospitals were enrolled, including 33 COVID-19 patients in the case group and 268 patients with malignancies or pneumonia in the control group. Thirteen radiomics features out of 474 were selected to build the model. This model achieved an AUC of 0.905, accuracy of 89.5 %, sensitivity of 83.3 %, specificity of 90.0 % in the testing set.

CONCLUSION

We developed a noninvasive radiomics model based on CT imaging for the diagnosis of COVID-19 based on GGO lesions, which could be a promising supplementary tool for improving specificity for COVID-19 in a population confounded by ground glass opacity changes from other etiologies.

Progress in the diagnosis and treatment of COVID-19 and the role of surgeons in the front line of the pandemic

The current struggle to control and contain COVID-19 is critical and surgeons are on the front line in the fight against this virus. Surgeons, and other medical workers in the field of surgery, have a solid foundation and experience in medical treatment and intensive care, and an understanding of the support of respiratory, circulatory, digestive, and other systemic organs. Furthermore, the operative standards of aseptic technique in their daily work enable surgeons to adapt to the working environment in infected areas. As surgeons in the anti-pandemic front line in China, we describe our experience with the diagnosis and treatment of COVID-19 in this country and how the work of surgeons is unfolding during the pandemic.

Artificial intelligence in ophthalmology during COVID-19 and in the post COVID-19 era

PURPOSE OF REVIEW

To highlight artificial intelligence applications in ophthalmology during the COVID-19 pandemic that can be used to: describe ocular findings and changes correlated with COVID-19; extract information from scholarly articles on SARS-CoV-2 and COVID-19 specific to ophthalmology; and implement efficient patient triage and telemedicine care.

RECENT FINDINGS

Ophthalmology has been leading in artificial intelligence and technology applications. With medical imaging analysis, pixel-annotated distinguishable features on COVID-19 patients may help with noninvasive diagnosis and severity outcome predictions. Using natural language processing (NLP) and data integration methods, topic modeling on more than 200 ophthalmology-related articles on COVID-19 can summarize ocular manifestations, viral transmission, treatment strategies, and patient care and practice management. Artificial intelligence for telemedicine applications can address the high demand, prioritize and triage patients, as well as improve at home-monitoring devices and secure data transfers.

SUMMARY

COVID-19 is significantly impacting the way we are delivering healthcare. Given the already successful implementation of artificial intelligence applications and telemedicine in ophthalmology, we expect that these systems will be embraced more as tools for research, education, and patient care.

COVID-19: a brief update for radiologists

Coronavirus disease 2019 (COVID-19), which is caused by a new coronavirus-severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2)-is a pandemic with major impacts on the health care sector, and a broad view of the disease is of fundamental importance for any radiologist. The purpose of this review is to address the main clinical and imaging aspects of COVID-19, as well as guidelines for requesting and using imaging methods; measures to protect patients and health care professionals; systems for quantifying pulmonary findings and preparing integrated reports; and the main innovations that have emerged during this pandemic.

Review of radiographic findings in COVID-19

The purpose of this study is to review the published literature for the range of radiographic findings present in patients suffering from coronavirus disease 2019 infection. This novel corona virus is currently the cause of a worldwide pandemic. Pulmonary symptoms and signs dominate the clinical picture and radiologists are called upon to evaluate chest radiographs (CXR) and computed tomography (CT) images to assess for infiltrates and to define their extent, distribution and progression. Multiple studies attempt to characterize the disease course by looking at the timing of imaging relative to the onset of symptoms. In general, plain CXR show bilateral disease with a tendency toward the lung periphery and have an appearance most consistent with viral pneumonia. Chest CT images are most notable for showing bilateral and peripheral ground glass and consolidated opacities and are marked by an absence of concomitant pulmonary nodules, cavitation, adenopathy and pleural effusions. Published literature mentioning organ systems aside from pulmonary manifestations are relatively less common, yet present and are addressed in this review. Similarly, publications focusing on imaging modalities aside from CXR and chest CT are sparse in this evolving crisis and are likewise addressed in this review. The role of imaging is examined as it is currently being debated in the medical community, which is not at all surprising considering the highly infectious nature of Severe Acute Respiratory Syndrome coronavirus 2.

The Model for Early COvid-19 Recognition (MECOR) Score: A Proof-of-Concept for a Simple and Low-Cost Tool to Recognize a Possible Viral Etiology in Community-Acquired Pneumonia Patients during COVID-19 Outbreak

This study aims to assess the peripheral blood cell count "signature" of Severe Acute Respiratory Syndrome-Coronavirus 2 (SARS-CoV-2) to discriminate promptly between COronaVIrus Disease 19 (COVID-19) and community-acquired pneumonia (CAP). We designed a retrospective case-control study, enrolling 525 patients (283 COVID-19 and 242 with CAP). All patients had a fever and at least one of the following signs: cough, chest pain, or dyspnea. We excluded patients treated with immunosuppressants, steroids, or affected by diseases known to modify blood cell count. COVID-19 patients showed a significant reduction in white blood cells (neutrophils, lymphocytes, monocytes, eosinophils) and platelets. We studied these parameters univariately, combined the significant ones in a multivariate model (AUROC 0.86, Nagelkerke PSEUDO-R2 0.5, Hosmer-Lemeshow p-value 0.9) and examined its discriminative performance in an internally-randomized validation cohort (AUROC 0.84). The cut-off selected according to Youden's Index (-0.13) showed a sensitivity of 84% and a specificity of 72% in the training cohort, and a sensitivity of 88% and a specificity of 73% in the validation cohort. In addition, we determined the probability of having COVID-19 pneumonia for each Model for possible Early COvid-19 Recognition (MECOR) Score value. In conclusion, our model could provide a simple, rapid, and cheap tool for prompt COVID-19 diagnostic triage in patients with CAP. The actual effectiveness should be evaluated in further, prospective studies also involving COVID-19 patients with negative nasopharyngeal swabs.

Biomarkers of COVID-19 and technologies to combat SARS-CoV-2

Due to the unprecedented public health crisis caused by COVID-19, our first contribution to the newly launching journal, Advances in Biomarker Sciences and Technology, has abruptly diverted to focus on the current pandemic. As the number of new COVID-19 cases and deaths continue to rise steadily around the world, the common goal of healthcare providers, scientists, and government officials worldwide has been to identify the best way to detect the novel coronavirus, named SARS-CoV-2, and to treat the viral infection - COVID-19. Accurate detection, timely diagnosis, effective treatment, and future prevention are the vital keys to management of COVID-19, and can help curb the viral spread. Traditionally, biomarkers play a pivotal role in the early detection of disease etiology, diagnosis, treatment and prognosis. To assist myriad ongoing investigations and innovations, we developed this current article to overview known and emerging biomarkers for SARS-CoV-2 detection, COVID-19 diagnostics, treatment and prognosis, and ongoing work to identify and develop more biomarkers for new drugs and vaccines. Moreover, biomarkers of socio-psychological stress, the high-technology quest for new virtual drug screening, and digital applications are described.

Federated Learning of Electronic Health Records Improves Mortality Prediction in Patients Hospitalized with COVID-19

Machine learning (ML) models require large datasets which may be siloed across different healthcare institutions. Using federated learning, a ML technique that avoids locally aggregating raw clinical data across multiple institutions, we predict mortality within seven days in hospitalized COVID-19 patients. Patient data was collected from Electronic Health Records (EHRs) from five hospitals within the Mount Sinai Health System (MSHS). Logistic Regression with L1 regularization (LASSO) and Multilayer Perceptron (MLP) models were trained using local data at each site, a pooled model with combined data from all five sites, and a federated model that only shared parameters with a central aggregator. Both the federated LASSO and federated MLP models performed better than their local model counterparts at four hospitals. The federated MLP model also outperformed the federated LASSO model at all hospitals. Federated learning shows promise in COVID-19 EHR data to develop robust predictive models without compromising patient privacy.

Emerging Technologies for Use in the Study, Diagnosis, and Treatment of Patients with COVID-19

INTRODUCTION

The COVID-19 pandemic has caused an unprecedented health and economic worldwide crisis. Innovative solutions are imperative given limited resources and immediate need for medical supplies, healthcare support and treatments.

AIM

The purpose of this review is to summarize emerging technologies being implemented in the study, diagnosis, and treatment of COVID-19.

RESULTS

Key focus areas include the applications of artificial intelligence, the use of Big Data and Internet of Things, the importance of mathematical modeling for predictions, utilization of technology for community screening, the use of nanotechnology for treatment and vaccine development, the utility of telemedicine, the implementation of 3D-printing to manage new demands and the potential of robotics.

CONCLUSION

The review concludes by highlighting the need for collaboration in the scientific community with open sharing of knowledge, tools, and expertise.

Application of artificial intelligence in hepatology: Minireview

With the rapid advancements in computer science, artificial intelligence (AI) has become an intrinsic part of our daily life and clinical practices. The concepts of AI, such as machine learning, deep learning, and big data, are extensively used in clinical and basic research. In this review, we searched for the articles in PubMed and summarized recent developments of AI concerning hepatology while focusing on the diagnosis and risk assessment of liver diseases. Ultrasound is widely conducted for the routine surveillance of hepatocellular carcinoma along with tumor markers. Computer-aided diagnosis is useful in the detection of tumors and characterization of space-occupying lesions. The prognosis of hepatocellular carcinoma can be estimated via AI using large-scale and high-quality training datasets. The prevalence of nonalcoholic fatty liver disease is increasing worldwide and pivotal concern in the field is who will progress and develop hepatocellular carcinoma. Most AI studies require a large dataset, including laboratory or radiological findings and outcome data. AI will be useful in reducing medical errors, supporting clinical decisions, and predicting clinical outcomes. Thus, cooperation between AI and humans is expected to improve healthcare.

Detection of COVID-19 Infection from Routine Blood Exams with Machine Learning: A Feasibility Study

The COVID-19 pandemia due to the SARS-CoV-2 coronavirus, in its first 4 months since its outbreak, has to date reached more than 200 countries worldwide with more than 2 million confirmed cases (probably a much higher number of infected), and almost 200,000 deaths. Amplification of viral RNA by (real time) reverse transcription polymerase chain reaction (rRT-PCR) is the current gold standard test for confirmation of infection, although it presents known shortcomings: long turnaround times (3-4 hours to generate results), potential shortage of reagents, false-negative rates as large as 15-20%, the need for certified laboratories, expensive equipment and trained personnel. Thus there is a need for alternative, faster, less expensive and more accessible tests. We developed two machine learning classification models using hematochemical values from routine blood exams (namely: white blood cells counts, and the platelets, CRP, AST, ALT, GGT, ALP, LDH plasma levels) drawn from 279 patients who, after being admitted to the San Raffaele Hospital (Milan, Italy) emergency-room with COVID-19 symptoms, were screened with the rRT-PCR test performed on respiratory tract specimens. Of these patients, 177 resulted positive, whereas 102 received a negative response. We have developed two machine learning models, to discriminate between patients who are either positive or negative to the SARS-CoV-2: their accuracy ranges between 82% and 86%, and sensitivity between 92% e 95%, so comparably well with respect to the gold standard. We also developed an interpretable Decision Tree model as a simple decision aid for clinician interpreting blood tests (even off-line) for COVID-19 suspect cases. This study demonstrated the feasibility and clinical soundness of using blood tests analysis and machine learning as an alternative to rRT-PCR for identifying COVID-19 positive patients. This is especially useful in those countries, like developing ones, suffering from shortages of rRT-PCR reagents and specialized laboratories. We made available a Web-based tool for clinical reference and evaluation (This tool is available at https://covid19-blood-ml.herokuapp.com/ ).

[Seroprevalence and SARS-CoV-2 testing in healthcare occupations]

The SARS-CoV‑2 causes a disease spectrum that includes asymptomatic and mildly symptomatic infections with subclinical manifestations but which can nevertheless still be potentially contagious. Evidence from SARS-CoV‑2 infected macaque monkeys and from studies with seasonal coronaviruses suggests that the infection is likely to produce an immunity that is protective for a certain period of time. Available test methods enable a high degree of reliability, e.g. if high-quality serological methods are combined. Although individual test results have to be interpreted with caution, serosurveillance in a tertiary eye care center and large eye research institute can reduce anxiety and provide clarity regarding the actual number of (unreported) SARS-CoV‑2 infections.

Detection of COVID-19 Infection from Routine Blood Exams with Machine Learning: A Feasibility Study

The COVID-19 pandemia due to the SARS-CoV-2 coronavirus, in its first 4 months since its outbreak, has to date reached more than 200 countries worldwide with more than 2 million confirmed cases (probably a much higher number of infected), and almost 200,000 deaths. Amplification of viral RNA by (real time) reverse transcription polymerase chain reaction (rRT-PCR) is the current gold standard test for confirmation of infection, although it presents known shortcomings: long turnaround times (3-4 hours to generate results), potential shortage of reagents, false-negative rates as large as 15-20%, the need for certified laboratories, expensive equipment and trained personnel. Thus there is a need for alternative, faster, less expensive and more accessible tests. We developed two machine learning classification models using hematochemical values from routine blood exams (namely: white blood cells counts, and the platelets, CRP, AST, ALT, GGT, ALP, LDH plasma levels) drawn from 279 patients who, after being admitted to the San Raffaele Hospital (Milan, Italy) emergency-room with COVID-19 symptoms, were screened with the rRT-PCR test performed on respiratory tract specimens. Of these patients, 177 resulted positive, whereas 102 received a negative response. We have developed two machine learning models, to discriminate between patients who are either positive or negative to the SARS-CoV-2: their accuracy ranges between 82% and 86%, and sensitivity between 92% e 95%, so comparably well with respect to the gold standard. We also developed an interpretable Decision Tree model as a simple decision aid for clinician interpreting blood tests (even off-line) for COVID-19 suspect cases. This study demonstrated the feasibility and clinical soundness of using blood tests analysis and machine learning as an alternative to rRT-PCR for identifying COVID-19 positive patients. This is especially useful in those countries, like developing ones, suffering from shortages of rRT-PCR reagents and specialized laboratories. We made available a Web-based tool for clinical reference and evaluation (This tool is available at https://covid19-blood-ml.herokuapp.com/ ).

COVID-19 as a Vascular Disease: Lesson Learned from Imaging and Blood Biomarkers

COVID-19, a disease initially thought to be prominently an interstitial pneumonia with varying degrees of severity, can be considered a vascular disease with regards to serious complications and causes of mortality. Quite recently, blood clots have emerged as the common factor unifying many of the symptoms initially attributed without an explanation to COVID-19. Cardiovascular biomarkers and particularly, D-dimer and troponin appear to be very powerful prognostic markers, signaling the need for earlier and more aggressive interventions and treatments in order to avoid and/or minimize arterial/venous thromboembolism and myocardial infarct. The ultrasound imaging patterns at both the lung and peripheral vascular level can also be very useful weapons that have the advantage of being able to monitor longitudinally the clinical picture, something that real-time PCR/nasopharyngeal swab is not able to do and that CT can only pursue with significant radiation exposure. A lesson learned in the early phase of the COVID-19 pandemic suggests quitting and starting again with targeted imaging and blood vascular biomarkers.

A COVID-19 Risk Assessment Decision Support System for General Practitioners: Design and Development Study

BACKGROUND

The coronavirus disease (COVID-19) has become an urgent and serious global public health crisis. Community engagement is the first line of defense in the fight against infectious diseases, and general practitioners (GPs) play an important role in it. GPs are facing unique challenges from disasters and pandemics in delivering health care. However, there is still no suitable mobile management system that can help GPs collect data, dynamically assess risks, and effectively triage or follow-up with patients with COVID-19.

OBJECTIVE

The aim of this study is to design, develop, and deploy a mobile-based decision support system for COVID-19 (DDC19) to assist GPs in collecting data, assessing risk, triaging, managing, and following up with patients during the COVID-19 outbreak.

METHODS

Based on the actual scenarios and the process of patients using health care, we analyzed the key issues that need to be solved and designed the main business flowchart of DDC19. We then constructed a COVID-19 dynamic risk stratification model with high recall and clinical interpretability, which was based on a multiclass logistic regression algorithm. Finally, through a 10-fold cross-validation to quantitatively evaluate the risk stratification ability of the model, a total of 2243 clinical data consisting of 36 dimension clinical features from fever clinics were used for training and evaluation of the model.

RESULTS

DDC19 is composed of three parts: mobile terminal apps for the patient-end and GP-end, and the database system. All mobile terminal devices were wirelessly connected to the back end data center to implement request sending and data transmission. We used low risk, moderate risk, and high risk as labels, and adopted a 10-fold cross-validation method to evaluate and test the COVID-19 dynamic risk stratification model in different scenarios (different dimensions of personal clinical data accessible at an earlier stage). The data set dimensions were (2243, 15) when only using the data of patients' demographic information, clinical symptoms, and contact history; (2243, 35) when the results of blood tests were added; and (2243, 36) after obtaining the computed tomography imaging results of the patient. The average value of the three classification results of the macro-area under the curve were all above 0.71 in each scenario.

CONCLUSIONS

DCC19 is a mobile decision support system designed and developed to assist GPs in providing dynamic risk assessments for patients with suspected COVID-19 during the outbreak, and the model had a good ability to predict risk levels in any scenario it covered.

Machine Learning for Precision Breast Cancer Diagnosis and Prediction of the Nanoparticle Cellular Internalization

In the field of theranostics, diagnostic nanoparticles are designed to collect highly patient-selective disease profiles, which is then leveraged by a set of nanotherapeutics to improve the therapeutic results. Despite their early promise, high interpatient and intratumoral heterogeneities make any rational design and analysis of these theranostics platforms extremely problematic. Recent advances in deep-learning-based tools may help bridge this gap, using pattern recognition algorithms for better diagnostic precision and therapeutic outcome. Triple-negative breast cancer (TNBC) is a conundrum because of the complex molecular diversity, making its diagnosis and therapy challenging. To address these challenges, we propose a method to predict the cellular internalization of nanoparticles (NPs) against different cancer stages using artificial intelligence. Here, we demonstrate for the first time that a combination of machine-learning (ML) algorithm and characteristic cellular uptake responses for individual cancer cell types can be successfully used to classify various cancer cell types. Utilizing this approach, we can optimize the nanomaterials to get an optimum structure-internalization response for a given particle. This methodology predicted the structure-internalization response of the evaluated nanoparticles with remarkable accuracy (Q² = 0.9). We anticipate that it can reduce the effort by minimizing the number of nanoparticles that need to be tested and could be utilized as a screening tool for designing nanotherapeutics. Following this, we have proposed a diagnostic nanomaterial-based platform used to assemble a patient-specific cancer profile with the assistance of machine learning (ML). The platform is composed of eight carbon nanoparticles (CNPs) with multifarious surface chemistries that can differentiate healthy breast cells from cancerous cells and then subclassify TNBC cells vs non-TNBC cells, within the TNBC group. The artificial neural network (ANN) algorithm has been successfully used in identifying the type of cancer cells from 36 unknown cancer samples with an overall accuracy of >98%, providing potential applications in cancer diagnostics.

Considerations for development and use of AI in response to COVID-19

•

Consider a CAIO as a key role to lead the use of AI in COVID-19 response.

•

Methods are needed to manage unpredictable, unexpected, or biased data.

•

Repurposed AI offers promise for rapid availability of applications.

•

AI accuracy rates achieved in development may be lower in real-life use.

•

Diverse AI team membership is meant to attain the best system performance.

Artificial intelligence (AI) is playing a key supporting role in the fight against COVID-19 and perhaps will contribute to solutions quicker than we would otherwise achieve in many fields and applications. Since the outbreak of the pandemic, there has been an upsurge in the exploration and use of AI, and other data analytic tools, in a multitude of areas. This paper addresses some of the many considerations for managing the development and deployment of AI applications, including planning; unpredictable, unexpected, or biased results; repurposing; the importance of data; and diversity in AI team membership. We provide implications for research and for practice, according to each of the considerations. Finally we conclude that we need to plan and carefully consider the issues associated with the development and use of AI as we look for quick solutions.

Review of chest CT manifestations of COVID-19 infection

Coronavirus disease-19 (COVID-19) is a viral pandemic that started in China and has rapidly expanded worldwide. Typical clinical manifestations include fever, cough and dyspnea after an incubation period of 2-14 days. The diagnosis is based on RT-PCR test through a nasopharyngeal swab. Because of the pulmonary tropism of the virus, pneumonia is often encountered in symptomatic patients. Here, we review the pertinent clinical findings and the current published data describing chest CT findings in COVID-19 pneumonia, the diagnostic performance of CT for diagnosis, including differential diagnosis, as well the evolving role of imaging in this disease.

Prediction models for diagnosis and prognosis of covid-19: systematic review and critical appraisal

OBJECTIVE

To review and appraise the validity and usefulness of published and preprint reports of prediction models for diagnosing coronavirus disease 2019 (covid-19) in patients with suspected infection, for prognosis of patients with covid-19, and for detecting people in the general population at increased risk of covid-19 infection or being admitted to hospital with the disease.

DESIGN

Living systematic review and critical appraisal by the COVID-PRECISE (Precise Risk Estimation to optimise covid-19 Care for Infected or Suspected patients in diverse sEttings) group.

DATA SOURCES

PubMed and Embase through Ovid, up to 1 July 2020, supplemented with arXiv, medRxiv, and bioRxiv up to 5 May 2020.

STUDY SELECTION

Studies that developed or validated a multivariable covid-19 related prediction model.

DATA EXTRACTION

At least two authors independently extracted data using the CHARMS (critical appraisal and data extraction for systematic reviews of prediction modelling studies) checklist; risk of bias was assessed using PROBAST (prediction model risk of bias assessment tool).

RESULTS

37 421 titles were screened, and 169 studies describing 232 prediction models were included. The review identified seven models for identifying people at risk in the general population; 118 diagnostic models for detecting covid-19 (75 were based on medical imaging, 10 to diagnose disease severity); and 107 prognostic models for predicting mortality risk, progression to severe disease, intensive care unit admission, ventilation, intubation, or length of hospital stay. The most frequent types of predictors included in the covid-19 prediction models are vital signs, age, comorbidities, and image features. Flu-like symptoms are frequently predictive in diagnostic models, while sex, C reactive protein, and lymphocyte counts are frequent prognostic factors. Reported C index estimates from the strongest form of validation available per model ranged from 0.71 to 0.99 in prediction models for the general population, from 0.65 to more than 0.99 in diagnostic models, and from 0.54 to 0.99 in prognostic models. All models were rated at high or unclear risk of bias, mostly because of non-representative selection of control patients, exclusion of patients who had not experienced the event of interest by the end of the study, high risk of model overfitting, and unclear reporting. Many models did not include a description of the target population (n=27, 12%) or care setting (n=75, 32%), and only 11 (5%) were externally validated by a calibration plot. The Jehi diagnostic model and the 4C mortality score were identified as promising models.

CONCLUSION

Prediction models for covid-19 are quickly entering the academic literature to support medical decision making at a time when they are urgently needed. This review indicates that almost all pubished prediction models are poorly reported, and at high risk of bias such that their reported predictive performance is probably optimistic. However, we have identified two (one diagnostic and one prognostic) promising models that should soon be validated in multiple cohorts, preferably through collaborative efforts and data sharing to also allow an investigation of the stability and heterogeneity in their performance across populations and settings. Details on all reviewed models are publicly available at https://www.covprecise.org/. Methodological guidance as provided in this paper should be followed because unreliable predictions could cause more harm than benefit in guiding clinical decisions. Finally, prediction model authors should adhere to the TRIPOD (transparent reporting of a multivariable prediction model for individual prognosis or diagnosis) reporting guideline.

SYSTEMATIC REVIEW REGISTRATION

Protocol https://osf.io/ehc47/, registration https://osf.io/wy245.

READERS' NOTE

This article is a living systematic review that will be updated to reflect emerging evidence. Updates may occur for up to two years from the date of original publication. This version is update 3 of the original article published on 7 April 2020 (BMJ 2020;369:m1328). Previous updates can be found as data supplements (https://www.bmj.com/content/369/bmj.m1328/related#datasupp). When citing this paper please consider adding the update number and date of access for clarity.

COVID-SAFE: An IoT-Based System for Automated Health Monitoring and Surveillance in Post-Pandemic Life

In the early months of the COVID-19 pandemic with no designated cure or vaccine, the only way to break the infection chain is self-isolation and maintaining the physical distancing. In this article, we present a potential application of the Internet of Things (IoT) in healthcare and physical distance monitoring for pandemic situations. The proposed framework consists of three parts: a lightweight and low-cost IoT node, a smartphone application (app), and fog-based Machine Learning (ML) tools for data analysis and diagnosis. The IoT node tracks health parameters, including body temperature, cough rate, respiratory rate, and blood oxygen saturation, then updates the smartphone app to display the user health conditions. The app notifies the user to maintain a physical distance of 2 m (or 6 ft), which is a key factor in controlling virus spread. In addition, a Fuzzy Mamdani system (running at the fog server) considers the environmental risk and user health conditions to predict the risk of spreading infection in real time. The environmental risk conveys from the virtual zone concept and provides updated information for different places. Two scenarios are considered for the communication between the IoT node and fog server, 4G/5G/WiFi, or LoRa, which can be selected based on environmental constraints. The required energy usage and bandwidth (BW) are compared for various event scenarios. The COVID-SAFE framework can assist in minimizing the coronavirus exposure risk.