Unsupervised machine learning framework for discriminating major variants of concern during COVID-19.

Due to the high mutation rate of the virus, the COVID-19 pandemic evolved rapidly. Certain variants of the virus, such as Delta and Omicron emerged with altered viral properties leading to severe transmission and death rates. These variants burdened the medical systems worldwide with a major impact to travel, productivity, and the world economy. Unsupervised machine learning methods have the ability to compress, characterize, and visualize unlabelled data. This paper presents a framework that utilizes unsupervised machine learning methods to discriminate and visualize the associations between major COVID-19 variants based on their genome sequences. These methods comprise a combination of selected dimensionality reduction and clustering techniques. The framework processes the RNA sequences by performing a k-mer analysis on the data and further visualises and compares the results using selected dimensionality reduction methods that include principal component analysis (PCA), t-distributed stochastic neighbour embedding (t-SNE), and uniform manifold approximation projection (UMAP). Our framework also employs agglomerative hierarchical clustering to visualize the mutational differences among major variants of concern and country-wise mutational differences for selected variants (Delta and Omicron) using dendrograms. We also provide country-wise mutational differences for selected variants via dendrograms. We find that the proposed framework can effectively distinguish between the major variants and has the potential to identify emerging variants in the future.

Severe Acute Respiratory Infection Surveillance during the Initial Phase of the COVID-19 Outbreak in North India: A Comparison of COVID-19 to Other SARI Causes.

INTRODUCTION: World Health Organization proposes severe acute respiratory infection (SARI) case definition for coronavirus disease 2019 (COVID-19) surveillance; however, early differentiation between SARI etiologies remains challenging. We aimed to investigate the spectrum and outcome of SARI and compare COVID-19 to non-COVID-19 causes. PATIENTS AND METHODS: A prospective cohort study was conducted between March 15, 2020, to August 15, 2020, at an adult medical emergency in North India. SARI was diagnosed using a "modified" case definition-febrile respiratory symptoms or radiographic evidence of pneumonia or acute respiratory distress syndrome of ≤14 days duration, along with a need for hospitalization and in the absence of an alternative etiology that fully explains the illness. COVID-19 was diagnosed with reverse transcription-polymerase chain reaction testing. RESULTS: In total, 95/212 (44.8%) cases had COVID-19. Community-acquired pneumonia (n = 57), exacerbation of chronic lung disease (n = 11), heart failure (n = 11), tropical febrile illnesses (n = 10), and influenza A (n = 5) were common non-COVID-19 causes. No between-group differences were apparent in age ≥60 years, comorbidities, oxygenation, leukocytosis, lymphopenia, acute physiology and chronic health evaluation (APACHE)-II score, CURB-65 score, and ventilator requirement at 24-hour. Bilateral lung distribution and middle-lower zones involvement in radiography predicted COVID-19. The median hospital stay was longer with COVID-19 (12 versus 5 days, p = 0.000); however, mortality was similar (31.6% versus 28.2%, p = 0.593). Independent mortality predictors were higher mean APACHE II in COVID-19 and early ventilator requirement in non-COVID-19 cases. CONCLUSIONS: COVID-19 has similar severity and mortality as non-COVID-19 SARI but requires an extended hospital stay. Including radiography in the SARI definition might improve COVID-19 surveillance. HOW TO CITE THIS ARTICLE: Pannu AK, Kumar M, Singh P, Shaji A, Ghosh A, Behera A, et al. Severe Acute Respiratory Infection Surveillance during the Initial Phase of the COVID-19 Outbreak in North India: A Comparison of COVID-19 to Other SARI Causes. Indian J Crit Care Med 2021;25(7):761-767.