Providing care in isolation while awaiting SARS-CoV-2 test results: Considering differential diagnoses and avoiding anchoring bias.

ABSTRACT: Isolation of confirmed or suspected coronavirus disease 2019 (COVID-19) cases is essential but, as symptoms of COVID-19 are non-specific and test results not immediately available, case identification at admission remains challenging. To inform optimization of triage algorithms, patient flow and patient care, we analyzed characteristics of patients admitted to an isolation ward, both severe acute respiratory syndrome coronavirus 2019 (SARS-CoV-2) positive patients and patients in which initial suspicion was not confirmed after appropriate testing.Data from patients with confirmed or suspected COVID-19 treated in an isolation unit were analyzed retrospectively. Symptoms, comorbidities and clinical findings were analyzed descriptively and associations between patient characteristics and final SARS-CoV-2 status were assessed using univariate regression.Eighty three patients (49 SARS-CoV-2 negative and 34 positive) were included in the final analysis. Of initially suspected COVID-19 cases, 59% proved to be SARS-CoV-2-negative. These patients had more comorbidities (Charlson Comorbidity Index median 5(interquartile range [IQR] 2.5, 7) vs 2.7(IQR 1, 4)), and higher proportion of active malignancy than patients with confirmed COVID-19 (47% vs 15%; P = .004), while immunosuppression was frequent in both patient groups (20% vs 21%; P = .984). Of SARS-CoV-2 negative patients, 31% were diagnosed with non-infectious diseases.A high proportion of patients (59%) triaged to the isolation unit were tested negative for SARS-CoV-2. Of these, many suffered from active malignancy (47%) and were immunosuppressed (20%). Non-infectious diseases were diagnosed in 31%, highlighting the need for appropriate patient flow, timely expert medical care including evaluation for differential diagnostics while providing isolation and ruling out of COVID-19 in these patients with complex underlying diseases.

Swarm Learning for decentralized and confidential clinical machine learning.

Fast and reliable detection of patients with severe and heterogeneous illnesses is a major goal of precision medicine1,2. Patients with leukaemia can be identified using machine learning on the basis of their blood transcriptomes3. However, there is an increasing divide between what is technically possible and what is allowed, because of privacy legislation4,5. Here, to facilitate the integration of any medical data from any data owner worldwide without violating privacy laws, we introduce Swarm Learning-a decentralized machine-learning approach that unites edge computing, blockchain-based peer-to-peer networking and coordination while maintaining confidentiality without the need for a central coordinator, thereby going beyond federated learning. To illustrate the feasibility of using Swarm Learning to develop disease classifiers using distributed data, we chose four use cases of heterogeneous diseases (COVID-19, tuberculosis, leukaemia and lung pathologies). With more than 16,400 blood transcriptomes derived from 127 clinical studies with non-uniform distributions of cases and controls and substantial study biases, as well as more than 95,000 chest X-ray images, we show that Swarm Learning classifiers outperform those developed at individual sites. In addition, Swarm Learning completely fulfils local confidentiality regulations by design. We believe that this approach will notably accelerate the introduction of precision medicine.