ABSTRACT
P80 Figure 1This ongoing project will enable our team to identify those requiring blood tests and potential treatment. As we continue we expect an increase in patient numbers to our CNS clinics. One major limitation is limited access to regional hospital laboratory results. Another limitation is staff being available to review files, responses to letters and chasing blood tests or results.Public Health England (2017). Hepatitis C in the UK – 2017 Report.Working to Eliminate Hepatitis C as a Major Public Health Threat. London: PHE.Thursz M. (2017). The fight against hepatitis C has not yet been won: here’s what we have to do. Huffington Post;10 August 2017.Vine LJ et al. Diagnosis and management of hepatitis C. British Journal of Hospital Medicine;2015;76:11, 625–630.World Health Organization (2016). Combating Hepatitis B and C to Reach Elimination by 2030. Geneva: WHO.
ABSTRACT
The rapid spread of COVID-19 cases in recent months has strained hospital resources, making rapid and accurate triage of patients presenting to emergency departments a necessity. Machine learning techniques using clinical data such as chest X-rays have been used to predict which patients are most at risk of deterioration. We consider the task of predicting two types of patient deterioration based on chest X-rays: adverse event deterioration (i.e., transfer to the intensive care unit, intubation, or mortality) and increased oxygen requirements beyond 6 L per day. Due to the relative scarcity of COVID-19 patient data, existing solutions leverage supervised pretraining on related non-COVID images, but this is limited by the differences between the pretraining data and the target COVID-19 patient data. In this paper, we use self-supervised learning based on the momentum contrast (MoCo) method in the pretraining phase to learn more general image representations to use for downstream tasks. We present three results. The first is deterioration prediction from a single image, where our model achieves an area under receiver operating characteristic curve (AUC) of 0.742 for predicting an adverse event within 96 hours (compared to 0.703 with supervised pretraining) and an AUC of 0.765 for predicting oxygen requirements greater than 6 L a day at 24 hours (compared to 0.749 with supervised pretraining). We then propose a new transformer-based architecture that can process sequences of multiple images for prediction and show that this model can achieve an improved AUC of 0.786 for predicting an adverse event at 96 hours and an AUC of 0.848 for predicting mortalities at 96 hours. A small pilot clinical study suggested that the prediction accuracy of our model is comparable to that of experienced radiologists analyzing the same information.
Subject(s)
COVID-19 , Learning DisabilitiesABSTRACT
During the coronavirus disease 2019 (COVID-19) pandemic, rapid and accurate triage of patients at the emergency department is critical to inform decision-making. We propose a data-driven approach for automatic prediction of deterioration risk using a deep neural network that learns from chest X-ray images and a gradient boosting model that learns from routine clinical variables. Our AI prognosis system, trained using data from 3,661 patients, achieves an area under the receiver operating characteristic curve (AUC) of 0.786 (95% CI: 0.745-0.830) when predicting deterioration within 96 hours. The deep neural network extracts informative areas of chest X-ray images to assist clinicians in interpreting the predictions and performs comparably to two radiologists in a reader study. In order to verify performance in a real clinical setting, we silently deployed a preliminary version of the deep neural network at New York University Langone Health during the first wave of the pandemic, which produced accurate predictions in real-time. In summary, our findings demonstrate the potential of the proposed system for assisting front-line physicians in the triage of COVID-19 patients.
Subject(s)
COVID-19ABSTRACT
Background: There are currently no FDA-approved medications for the treatment of COVID-19. At the onset of the pandemic, off-label medication use was supported by limited or no clinical data. We sought to characterize experimental COVID-19 therapies and identify safety signals during this period. Methods: We conducted a non-interventional, multicenter, point prevalence study of patients hospitalized with suspected/confirmed COVID-19. Clinical and treatment characteristics within a 24-hour window were evaluated in a random sample of up to 30 patients per site. The primary objective was to describe COVID-19 targeted therapies. The secondary objective was to describe adverse drug reactions (ADRs). Results: A total of 352 patients from 15 US hospitals were included. Most patients were treated at academic medical centers (53.4%) or community hospitals (42.6%). Sixty-seven patients (19%) were receiving drug therapy in addition to supportive care. Drug therapies included hydroxychloroquine (69%), remdesivir (10%), and interleukin-6 inhibitors (9%). Five patients (7.5%) were receiving combination therapy. Patients with a history of asthma (14.9% vs. 7%, p=0.037) and those enrolled in clinical trials (26.9% vs. 3.2%, p<0.001) were more likely to receive therapy. Among those receiving COVID-19 therapy, eight patients (12%) experienced an ADR, and ADRs were more commonly recognized in patients enrolled in clinical trials (62.5% vs 22%, OR=5.9, p=0.028). Conclusions: While we observed high rates of supportive care for patients with COVID-19, we also found that ADRs were common among patients receiving drug therapy including in clinical trials. Comprehensive systems are needed to identify and mitigate ADRs associated with experimental COVID-19 therapies.