The 2021 SIIM-FISABIO-RSNA Machine Learning COVID-19 Challenge: Annotation and Standard Exam Classification of COVID-19 Chest Radiographs.

We describe the curation, annotation methodology, and characteristics of the dataset used in an artificial intelligence challenge for detection and localization of COVID-19 on chest radiographs. The chest radiographs were annotated by an international group of radiologists into four mutually exclusive categories, including "typical," "indeterminate," and "atypical appearance" for COVID-19, or "negative for pneumonia," adapted from previously published guidelines, and bounding boxes were placed on airspace opacities. This dataset and respective annotations are available to researchers for academic and noncommercial use.

EventScore: An Automated Real-time Early Warning Score for Clinical Events

Early prediction of patients at risk of clinical deterioration can help physicians intervene and alter their clinical course towards better outcomes. In addition to the accuracy requirement, early warning systems must make the predictions early enough to give physicians enough time to intervene. Interpretability is also one of the challenges when building such systems since being able to justify the reasoning behind model decisions is desirable in clinical practice. In this work, we built an interpretable model for the early prediction of various adverse clinical events indicative of clinical deterioration. The model is evaluated on two datasets and four clinical events. The first dataset is collected in a predominantly COVID-19 positive population at Stony Brook Hospital. The second dataset is the MIMIC III dataset. The model was trained to provide early warning scores for ventilation, ICU transfer, and mortality prediction tasks on the Stony Brook Hospital dataset and to predict mortality and the need for vasopressors on the MIMIC III dataset. Our model first separates each feature into multiple ranges and then uses logistic regression with lasso penalization to select the subset of ranges for each feature. The model training is completely automated and doesn't require expert knowledge like other early warning scores. We compare our model to the Modified Early Warning Score (MEWS) and quick SOFA (qSOFA), commonly used in hospitals. We show that our model outperforms these models in the area under the receiver operating characteristic curve (AUROC) while having a similar or better median detection time on all clinical events, even when using fewer features. Unlike MEWS and qSOFA, our model can be entirely automated without requiring any manually recorded features. We also show that discretization improves model performance by comparing our model to a baseline logistic regression model. © 2022 IEEE.

Clinical outcome prediction in COVID-19 using self-supervised vision transformer representations

Automated analysis of chest imaging in coronavirus disease (COVID-19) has mostly been performed on smaller datasets leading to overfitting and poor generalizability. Training of deep neural networks on large datasets requires data labels. This is not always available and can be expensive to obtain. Self-supervision is being increasingly used in various medical imaging tasks to leverage large amount of unlabeled data during pretraining. Our proposed approach pretrains a vision transformer to perform two self-supervision tasks - image reconstruction and contrastive learning on a Chest Xray (CXR) dataset. In the process, we generate more robust image embeddings. The reconstruction module models visual semantics within the lung fields by reconstructing the input image through a mechanism which mimics denoising and autoencoding. On the other hand, the constrastive learning module learns the concept of similarity between two texture representations. After pretraining, the vision transformer is used as a feature extractor towards a clinical outcome prediction task on our target dataset. The pretraining multi-kaggle dataset comprises 27499 CXR scans while our target dataset contains 530 images. Specifically, our framework predicts ventilation and mortality outcomes for COVID-19 positive patients using baseline CXR. We compare our method against a baseline approach using pretrained ResNet50 features. Experimental results demonstrate that our proposed approach outperforms the supervised method. © 2022 SPIE.

Clinical Efficacy of Inhaled Nitric Oxide in Preventing the Progression of Moderate to Severe COVID-19 and Its Correlation to Viral Clearance: Results of a Pilot Study

Hypoxic patients with coronavirus disease 2019 (COVID-19) are at high risk of adverse outcomes. Inhaled nitric oxide (iNO) has shown anti-viral and immunomodulatory effects in vitro. However, in vivo evidence of efficacy in hypoxic COVID-19 is sparse. This open label feasibility study was conducted at a single referral center in South India and evaluated the effectiveness of repurposed iNO in improving clinical outcomes in COVID-19 and its correlation with viral clearance. We recruited hypoxemic COVID-19 patients and allocated them into treatment (iNO) and control groups (1:1). Viral clearance on day 5 favored the treatment group (100% vs 72%, P < 0.01). The speed of viral clearance as adjudged by normalized longitudinal cycle threshold (Ct) values was positively impacted in the treatment group. The proportion of patients who attained clinical improvement, defined as a ≥2-point change on the World Health Organization ordinal scale, was higher in the iNO cohort (n = 11, 79%) as compared to the control group (n = 4, 36%) (odds ratio 6.42, 95% confidence interval 1.09-37.73, P = 0.032). The proportion of patients progressing to mechanical ventilation in the control group (4/11) was significantly higher than in the treatment group (0/14). The all-cause 28-day mortality was significantly different among the study arms, with 36% (4/11) of the patients dying in the control group while none died in the treatment group. The numbers needed to treat to prevent an additional poor outcome of death was estimated to be 2.8. Our study demonstrates the putative role of repurposed iNO in hypoxemic COVID-19 patients and calls for extended validation. Copyright © 2021 the Author(s). Published by Wolters Kluwer Health, Inc.

Smart Office Automation System for Covid Prevention

Today, this coronavirus is spread all around the world. Most organizations and businesses start to think about how to continue their business in a situation like COVID-19 and their employees' health and business security. To avoid and be safe from this type of disease, there are some common rules to follow. Keeping a distance, wearing a mask, cleaning our hands, are some health guidelines from them. According to the current situation, many inventors are trying and have already given some solutions to avoid these kinds of situations aligning with health guidance' provided by WHO. With the advantage of advanced modern-day technologies and ideas, researchers started to think about how to face situations like these with the new technologies and found that many users are highly interested and motivated with automated systems. Thus, from this study, we aim to provide a fully automated office management system to prevent corona with advanced technology in combination with IoT technologies, Machine learning, Cloud technologies, and sensor technologies. Considering the security aspect, Controlling the main entrance, identifying, ensuring user's authentication before entering the building, and monitoring employee activities are very significant aspects of the study. As the result of the study, the combination of IoT technologies and Machine Learning with deep learning mechanisms have guaranteed organizational business continuity, employees' health, and security. © 2021 IEEE.

Quality of life and perceived financial implications among otorhinolaryngologists during the COVID-19 pandemic across India.

OBJECTIVE: The changing trends in medical practice, uncertainties and monetary apprehensions due to the COVID-19 pandemic may influence the sense of well-being among otorhinolaryngologists. The present study was conducted to evaluate quality of life (QOL) and perceived financial implications in otorhinolaryngologists during the COVID-19 pandemic across India. METHODS: A cross-sectional study was carried out among otorhinolaryngology specialists across India using WHOQOL-BREF and Consumer Financial Protection Bureau (CFPB) Financial Well-Being Scale questionnaires on Google Forms, which was kept open for the latter half of July 2020. RESULTS: A total of 358 responses were obtained; the response rate was 26.64%. Twenty-four percent of respondents worked exclusively in academic settings; 40.22% of specialists had over 10 years of work experience. Average monthly income in 2019 was between 1-3 lakhs Indian Rupees (INR) in 43.85%, while in 2020, 62.57% of the specialists had an average monthly income of below one lakh INR; this difference was statistically significant (p < 0.001). Mean WHOQOL-BREF scores for physical, psychological, social and environmental domains were 68.8 ± 1, 62.3 ± 0.75, 68.9 ± 1.17 and 65.8 ± 1.01, respectively; mean CFPB financial well-being scale score was 55.5 ± 0.66. QOL and financial well-being were better in otolaryngologists older than 60 years, male specialists and private consultants. CONCLUSIONS: There has been a tremendous impact on quality of life and financial well-being among otorhinolaryngologists in India during the COVID-19 pandemic. The study outcome may help otolaryngologists comprehend and perceive the extent to which it has affected their professional and personal lives, and explore ways to face and overcome the situation.

Attention-Based Multi-scale Gated Recurrent Encoder with Novel Correlation Loss for COVID-19 Progression Prediction

COVID-19 image analysis has mostly focused on diagnostic tasks using single timepoint scans acquired upon disease presentation or admission. We present a deep learning-based approach to predict lung infiltrate progression from serial chest radiographs (CXRs) of COVID-19 patients. Our method first utilizes convolutional neural networks (CNNs) for feature extraction from patches within the concerned lung zone, and also from neighboring and remote boundary regions. The framework further incorporates a multi-scale Gated Recurrent Unit (GRU) with a correlation module for effective predictions. The GRU accepts CNN feature vectors from three different areas as input and generates a fused representation. The correlation module attempts to minimize the correlation loss between hidden representations of concerned and neighboring area feature vectors, while maximizing the loss between the same from concerned and remote regions. Further, we employ an attention module over the output hidden states of each encoder timepoint to generate a context vector. This vector is used as an input to a decoder module to predict patch severity grades at a future timepoint. Finally, we ensemble the patch classification scores to calculate patient-wise grades. Specifically, our framework predicts zone-wise disease severity for a patient on a given day by learning representations from the previous temporal CXRs. Our novel multi-institutional dataset comprises sequential CXR scans from N = 93 patients. Our approach outperforms transfer learning and radiomic feature-based baseline approaches on this dataset. © 2021, Springer Nature Switzerland AG.

Chest Radiograph Disentanglement for COVID-19 Outcome Prediction

Chest radiographs (CXRs) are often the primary front-line diagnostic imaging modality. Pulmonary diseases manifest as characteristic changes in lung tissue texture rather than anatomical structure. Hence, we expect that studying changes in only lung tissue texture without the influence of possible structure variations would be advantageous for downstream prognostic and predictive modeling tasks. In this paper, we propose a generative framework, Lung Swapping Autoencoder (LSAE), that learns a factorized representation of a CXR to disentangle the tissue texture representation from the anatomic structure representation. Upon learning the disentanglement, we leverage LSAE in two applications. 1) After adapting the texture encoder in LSAE to a thoracic disease classification task on the large-scale ChestX-ray14 database (N = 112,120), we achieve a competitive result (mAUC: 79.0 % ) with unsupervised pre-training. Moreover, when compared with Inception v3 on our multi-institutional COVID-19 dataset, COVOC (N = 340), for a COVID-19 outcome prediction task (estimating need for ventilation), the texture encoder achieves 13 % less error with a 77 % smaller model size, further demonstrating the efficacy of texture representation for lung diseases. 2) We leverage the LSAE for data augmentation by generating hybrid lung images with textures and labels from the COVOC training data and lung structures from ChestX-ray14. This further improves ventilation outcome prediction on COVOC. The code is available here: https://github.com/cvlab-stonybrook/LSAE. © 2021, Springer Nature Switzerland AG.