RESUMO
As the Coronavirus Disease 2019 (COVID-19) pandemic continues to grow globally, testing to detect COVID-19 and isolating individuals who test positive remains to be the primary strategy for preventing community spread of the disease. The current gold standard method of testing for COVID-19 is the reverse transcription polymerase chain reaction (RT-PCR) test. The RT-PCR test, however, has an imperfect sensitivity (around 70%), is time-consuming and labor-intensive, and is in short supply, particularly in resource-limited countries. Therefore, automatic and accurate detection of COVID-19 using medical imaging modalities such as chest X-ray and Computed Tomography, which are more widely available and accessible, can be beneficial as an alternative diagnostic tool. We develop a novel hierarchical attention neural network model to classify chest radiography images as belonging to a person with either COVID-19, other infections, or no pneumonia (i.e., normal). We refer to this model as Artificial Intelligence for Detection of COVID-19 (AIDCOV). The hierarchical structure in AIDCOV captures the dependency of features and improves model performance while the attention mechanism makes the model interpretable and transparent. Using a publicly available dataset of 5801 chest images, we demonstrate that our model achieves a mean cross-validation accuracy of 97.8%. AIDCOV has a sensitivity of 99.3%, a specificity of 99.98%, and a positive predictive value of 99.6% in detecting COVID-19 from chest radiography images. AIDCOV can be used in conjunction with or instead of RT-PCR testing (where RT-PCR testing is unavailable) to detect and isolate individuals with COVID-19 and prevent onward transmission to the general population and healthcare workers.
RESUMO
ImportanceThe rapid spread of COVID-19 means that government and health services providers have little time to plan and design effective response policies. It is therefore important to rapidly provide accurate predictions of how vulnerable geographic regions such as counties are to the spread. ObjectiveDeveloping county level prediction around near future disease movement for COVID-19 occurrences using publicly available data. DesignOriginal Investigation; Decision Analytical Model Study for County Level COVID-19 occurrences using data from March 14-31, 2020. SettingDisease spread prediction for US counties. ParticipantsAll US county level granularity based on data fused from multiple publicly available sources inclusive of health statistics, demographics, and geographical features. Exposure(s) (for observational studies)Daily county level reported COVID-19 occurrences from March 14-31, 2020. Main Outcome(s) and Measure(s)We developed a 3-stage model to quantify, firstly the probability of COVID-19 occurrence for unaffected counties using XGBoost classifier and secondly, the number of potential occurrences of a county via XGBoost regression. Thirdly, these results are combined to compute the county level risk. This risk is then used as an estimated after-five-day-vulnerability of the county. ResultsUsing data from March 14-31, 2020, the model shows a sensitivity over 71.5% and specificity over 94%. Conclusions and RelevanceWe found that population, population density, percentage of people aged 70 or greater and prevalence of comorbidities play an important role in predicting COVID-19 occurrences. We found a positive association between affected and urban counties as well as less vulnerable and rural counties. The developed model can be used for identification of vulnerable counties and potential data discrepancies. Limited testing facilities and delayed results introduces significant variation in reported cases and produces a bias in the model. Trial RegistrationNot Applicable Key PointsO_ST_ABSQuestionC_ST_ABSWhat are key factors that define the vulnerability of counties in the US to cases of the COVID-19 virus? FindingsIn this epidemiological study based on publicly available data, we develop a model that predicts vulnerability to COVID-19 for each US county in terms of likelihood of going from no documented cases to at least one case within five days and in terms of number of occurrences of the virus. MeaningPredicting county vulnerability to COVID-19 can assist health organizations to better plan for resource and workforce needs.
