Predicting the necessity of oxygen therapy in the early stage of COVID-19 using machine learning

Predicting Severe Respiratory Failure in Patients with COVID-19: A Machine Learning Approach

Background/Objectives: Studies attempting to predict the development of severe respiratory failure in patients with a COVID-19 infection using machine learning algorithms have yielded different results due to differences in variable selection. We aimed to predict the development of severe respiratory failure, defined as the need for high-flow oxygen support, continuous positive airway pressure, or mechanical ventilation, in patients with COVID-19, using machine learning algorithms to identify the most important variables in achieving this prediction. Methods: This retrospective, cross-sectional study included COVID-19 patients with mild respiratory failure (mostly receiving oxygen through a mask or nasal cannula). We used XGBoost, support vector machines, multi-layer perceptron, k-nearest neighbor, random forests, decision trees, logistic regression, and naïve Bayes methods to accurately predict severe respiratory failure in these patients. Results: A total of 320 patients (62.1% male; average age, 54.67 ± 15.82 years) were included in this study. During the follow-ups of these cases, 114 patients (35.6%) required high-level oxygen support, 67 (20.9%) required intensive care unit admission, and 43 (13.4%) died. The machine learning algorithms with the highest accuracy values were XGBoost, support vector machines, k-nearest neighbor, logistic regression, and multi-layer perceptron (0.7395, 0.7395, 0.7291, 0.7187, and 0.75, respectively). The method that obtained the highest ROC-AUC value was logistic regression (ROC-AUC = 0.7274). The best predictors of severe respiratory failure were a low lymphocyte count, a high computed tomography score in the right and left upper lung zones, an elevated neutrophil count, a small decrease in CRP levels on the third day of admission, a high Charlson comorbidity index score, and a high serum procalcitonin level. Conclusions: The development of severe respiratory failure in patients with COVID-19 could be successfully predicted using machine learning methods, especially logistic regression, and the best predictors of severe respiratory failure were the lymphocyte count and the degree of upper lung zone involvement.

Machine and deep learning methods for clinical outcome prediction based on physiological data of COVID-19 patients: a scoping review

INTRODUCTION

Since the beginning of the COVID-19 pandemic, numerous machine and deep learning (MDL) methods have been proposed in the literature to analyze patient physiological data. The objective of this review is to summarize various aspects of these methods and assess their practical utility for predicting various clinical outcomes.

METHODS

We searched PubMed, Scopus, and Cochrane Library, screened and selected the studies matching the inclusion criteria. The clinical analysis focused on the characteristics of the patient cohorts in the studies included in this review, the specific tasks in the context of the COVID-19 pandemic that machine and deep learning methods were used for, and their practical limitations. The technical analysis focused on the details of specific MDL methods and their performance.

RESULTS

Analysis of the 48 selected studies revealed that the majority (∼54 %) of them examined the application of MDL methods for the prediction of survival/mortality-related patient outcomes, while a smaller fraction (∼13 %) of studies also examined applications to the prediction of patients' physiological outcomes and hospital resource utilization. 21 % of the studies examined the application of MDL methods to multiple clinical tasks. Machine and deep learning methods have been shown to be effective at predicting several outcomes of COVID-19 patients, such as disease severity, complications, intensive care unit (ICU) transfer, and mortality. MDL methods also achieved high accuracy in predicting the required number of ICU beds and ventilators.

CONCLUSION

Machine and deep learning methods have been shown to be valuable tools for predicting disease severity, organ dysfunction and failure, patient outcomes, and hospital resource utilization during the COVID-19 pandemic. The discovered knowledge and our conclusions and recommendations can also be useful to healthcare professionals and artificial intelligence researchers in managing future pandemics.

O2 supplementation disambiguation in clinical narratives to support retrospective COVID-19 studies

BACKGROUND

Oxygen saturation, a key indicator of COVID-19 severity, poses challenges, especially in cases of silent hypoxemia. Electronic health records (EHRs) often contain supplemental oxygen information within clinical narratives. Streamlining patient identification based on oxygen levels is crucial for COVID-19 research, underscoring the need for automated classifiers in discharge summaries to ease the manual review burden on physicians.

METHOD

We analysed text lines extracted from anonymised COVID-19 patient discharge summaries in German to perform a binary classification task, differentiating patients who received oxygen supplementation and those who did not. Various machine learning (ML) algorithms, including classical ML to deep learning (DL) models, were compared. Classifier decisions were explained using Local Interpretable Model-agnostic Explanations (LIME), which visualize the model decisions.

RESULT

Classical ML to DL models achieved comparable performance in classification, with an F-measure varying between 0.942 and 0.955, whereas the classical ML approaches were faster. Visualisation of embedding representation of input data reveals notable variations in the encoding patterns between classic and DL encoders. Furthermore, LIME explanations provide insights into the most relevant features at token level that contribute to these observed differences.

CONCLUSION

Despite a general tendency towards deep learning, these use cases show that classical approaches yield comparable results at lower computational cost. Model prediction explanations using LIME in textual and visual layouts provided a qualitative explanation for the model performance.

Oxygen Requirement in Overweight/Obese Kidney Transplant Recipients with COVID-19: An Observational Cohort Study

INTRODUCTION

Obesity is one of the components of the cardiometabolic syndrome that contributes to COVID-19 progression and mortality. Immunosuppressed individuals are at greater risk of the COVID-19 burden. Therefore, we sought to investigate the impact of the combination of overweight/obesity and kidney transplant on oxygen (O2) requirements in the COVID-19 setting.

METHODS

Retrospective analysis of 284 kidney transplant recipients (KTRs) from March/2020 to August/2020 in a single center. We investigated the risk factors associated with O2 requirements in overweight/obese KTRs.

RESULTS

Overall, 65.1% had a BMI (body mass index) ≥ 25 kg/m2, 52.4% were male, the mean age was 53.3 ± 11 years old, 78.4% had hypertension, and 41.1% had diabetes mellitus. BMI was an independent risk factor for O2 requirements (OR = 1.07, p = 0.02) alongside age, lymphopenia, and hyponatremia. When overweight/obese KTRs were older, smokers, they presented higher levels of lactate dehydrogenase (LDH), and lower levels of estimated glomerular filtration rate (eGFR), lymphocytes, and sodium at admission, and they needed O2 more often.

CONCLUSION

Being overweight/obese is associated with greater O2 requirements in KTRs, in particular in older people and smokers, with worse kidney allograft functions, more inflammation, and lower sodium levels. Therefore, the early identification of factors that predict a worse outcome in overweight/obese KTRs affected by COVID-19 contributes to risk stratification and therapeutic decisions.

Comparative study of quality of life 9 months post-COVID-19 infection with SARS-CoV-2 of varying degrees of severity: impact of hospitalization vs. outpatient treatment

Purpose

This experimental study was conducted during the post-COVID-19 period to investigate the relationship between the quality of life 9 months after and the severity of the SARS-CoV-2 infection in two scenarios: hospitalization (with/without medical oxygen) and outpatient treatment.

Methods

We employed the EQ-5D-5L Quality of Life tests and the PSQI as a survey to evaluate respondents' quality of life 9 months after a previous SARS-CoV-2 infection of varying severity.

Results

We identified a clear difference in the quality of life of respondents, as measured on the 100-point scale of the EQ-5D-5L test, which was significantly lower 9 months after a previous SARS-CoV-2 infection for Group 1 (n = 14), respondents who had received medical attention for SARS-CoV-2 infection in a hospital with oxygen treatment, compared to those with the SARS-CoV-2 infection who were treated without oxygen treatment (Group 2) (n = 12) and those who were treated on an outpatient basis (Group 3) (n = 13) (H = 7.08 p = 0.029). There were no intergroup differences in quality of life indicators between hospitalized patients (Group 2) and groups 1 and 3. PSQI survey results showed that "mobility," "self-care," "daily activities," "pain/discomfort," and "anxiety/ depression" did not differ significantly between the groups, indicating that these factors were not associated with the severity of the SARS-CoV-2 infection. On the contrary, the respondents demonstrated significant inter-group differences (H = 7.51 p = 0.023) and the interdependence of respiratory difficulties with the severity of clinically diagnosed SARS-CoV-2 infection. This study also demonstrated significant differences in the values of sleep duration, sleep disorders, and daytime sleepiness indicators between the three groups of respondents, which indicate the influence of the severity of the infection. The PSQI test results revealed significant differences in "bedtime" (H = 6.00 p = 0.050) and "wake-up time" (H = 11.17 p = 0.004) between Groups 1 and 3 of respondents. At 9 months after COVID-19, respondents in Group 1 went to bed at a later time (pp = 0.02727) and woke up later (p = 0.003) than the respondents in Group 3.

Conclusion

This study is the first of its kind in the current literature to report on the quality of life of respondents 9 months after being diagnosed with COVID-19 and to draw comparisons between cohorts of hospitalized patients who were treated with medical oxygen vs. the cohorts of outpatient patients. The study's findings regarding post-COVID-19 quality of life indicators and their correlation with the severity of the SARS-CoV-2 infection can be used to categorize patients for targeted post-COVID-19 rehabilitation programs.

Combating Covid-19 using machine learning and deep learning: Applications, challenges, and future perspectives

COVID-19, a worldwide pandemic that has affected many people and thousands of individuals have died due to COVID-19, during the last two years. Due to the benefits of Artificial Intelligence (AI) in X-ray image interpretation, sound analysis, diagnosis, patient monitoring, and CT image identification, it has been further researched in the area of medical science during the period of COVID-19. This study has assessed the performance and investigated different machine learning (ML), deep learning (DL), and combinations of various ML, DL, and AI approaches that have been employed in recent studies with diverse data formats to combat the problems that have arisen due to the COVID-19 pandemic. Finally, this study shows the comparison among the stand-alone ML and DL-based research works regarding the COVID-19 issues with the combinations of ML, DL, and AI-based research works. After in-depth analysis and comparison, this study responds to the proposed research questions and presents the future research directions in this context. This review work will guide different research groups to develop viable applications based on ML, DL, and AI models, and will also guide healthcare institutes, researchers, and governments by showing them how these techniques can ease the process of tackling the COVID-19.

Predicting oxygen requirements in patients with coronavirus disease 2019 using an artificial intelligence-clinician model based on local non-image data

BACKGROUND

When facing unprecedented emergencies such as the coronavirus disease 2019 (COVID-19) pandemic, a predictive artificial intelligence (AI) model with real-time customized designs can be helpful for clinical decision-making support in constantly changing environments. We created models and compared the performance of AI in collaboration with a clinician and that of AI alone to predict the need for supplemental oxygen based on local, non-image data of patients with COVID-19.

MATERIALS AND METHODS

We enrolled 30 patients with COVID-19 who were aged >60 years on admission and not treated with oxygen therapy between December 1, 2020 and January 4, 2021 in this 50-bed, single-center retrospective cohort study. The outcome was requirement for oxygen after admission.

RESULTS

The model performance to predict the need for oxygen by AI in collaboration with a clinician was better than that by AI alone. Sodium chloride difference >33.5 emerged as a novel indicator to predict the need for oxygen in patients with COVID-19. To prevent severe COVID-19 in older patients, dehydration compensation may be considered in pre-hospitalization care.

CONCLUSION

In clinical practice, our approach enables the building of a better predictive model with prompt clinician feedback even in new scenarios. These can be applied not only to current and future pandemic situations but also to other diseases within the healthcare system.

Mortality Analysis of Patients with COVID-19 in Mexico Based on Risk Factors Applying Machine Learning Techniques

The new pandemic caused by the COVID-19 virus has generated an overload in the quality of medical care in clinical centers around the world. Causes that originate this fact include lack of medical personnel, infrastructure, medicines, among others. The rapid and exponential increase in the number of patients infected by COVID-19 has required an efficient and speedy prediction of possible infections and their consequences with the purpose of reducing the health care quality overload. Therefore, intelligent models are developed and employed to support medical personnel, allowing them to give a more effective diagnosis about the health status of patients infected by COVID-19. This paper aims to propose an alternative algorithmic analysis for predicting the health status of patients infected with COVID-19 in Mexico. Different prediction models such as KNN, logistic regression, random forests, ANN and majority vote were evaluated and compared. The models use risk factors as variables to predict the mortality of patients from COVID-19. The most successful scheme is the proposed ANN-based model, which obtained an accuracy of 90% and an F1 score of 89.64%. Data analysis reveals that pneumonia, advanced age and intubation requirement are the risk factors with the greatest influence on death caused by virus in Mexico.

Biotechnological Perspectives to Combat the COVID-19 Pandemic: Precise Diagnostics and Inevitable Vaccine Paradigms

The outbreak of the novel severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is the cause for the ongoing global public health emergency. It is more commonly known as coronavirus disease 2019 (COVID-19); the pandemic threat continues to spread aroundthe world with the fluctuating emergence of its new variants. The severity of COVID-19 ranges from asymptomatic to serious acute respiratory distress syndrome (ARDS), which has led to a high human mortality rate and disruption of socioeconomic well-being. For the restoration of pre-pandemic normalcy, the international scientific community has been conducting research on a war footing to limit extremely pathogenic COVID-19 through diagnosis, treatment, and immunization. Since the first report of COVID-19 viral infection, an array of laboratory-based and point-of-care (POC) approaches have emerged for diagnosing and understanding its status of outbreak. The RT-PCR-based viral nucleic acid test (NAT) is one of the rapidly developed and most used COVID-19 detection approaches. Notably, the current forbidding status of COVID-19 requires the development of safe, targeted vaccines/vaccine injections (shots) that can reduce its associated morbidity and mortality. Massive and accelerated vaccination campaigns would be the most effective and ultimate hope to end the COVID-19 pandemic. Since the SARS-CoV-2 virus outbreak, emerging biotechnologies and their multidisciplinary approaches have accelerated the understanding of molecular details as well as the development of a wide range of diagnostics and potential vaccine candidates, which are indispensable to combating the highly contagious COVID-19. Several vaccine candidates have completed phase III clinical studies and are reported to be effective in immunizing against COVID-19 after their rollout via emergency use authorization (EUA). However, optimizing the type of vaccine candidates and its route of delivery that works best to control viral spread is crucial to face the threatening variants expected to emerge over time. In conclusion, the insights of this review would facilitate the development of more likely diagnostics and ideal vaccines for the global control of COVID-19.

A Machine Learning Approach to Predict the Rehabilitation Outcome in Convalescent COVID-19 Patients

Background: After the acute disease, convalescent coronavirus disease 2019 (COVID-19) patients may experience several persistent manifestations that require multidisciplinary pulmonary rehabilitation (PR). By using a machine learning (ML) approach, we aimed to evaluate the clinical characteristics predicting the effectiveness of PR, expressed by an improved performance at the 6-min walking test (6MWT). Methods: Convalescent COVID-19 patients referring to a Pulmonary Rehabilitation Unit were consecutively screened. The 6MWT performance was partitioned into three classes, corresponding to different degrees of improvement (low, medium, and high) following PR. A multiclass supervised classification learning was performed with random forest (RF), adaptive boosting (ADA-B), and gradient boosting (GB), as well as tree-based and k-nearest neighbors (KNN) as instance-based algorithms. Results: To train and validate our model, we included 189 convalescent COVID-19 patients (74.1% males, mean age 59.7 years). RF obtained the best results in terms of accuracy (83.7%), sensitivity (84.0%), and area under the ROC curve (94.5%), while ADA-B reached the highest specificity (92.7%). Conclusions: Our model enables a good performance in predicting the rehabilitation outcome in convalescent COVID-19 patients.