Risk Factors Analysis of COVID-19 Patients with ARDS and Prediction Based on Machine Learning

Xu, Wei; Sun, Nannan; Gao, Hainv; Chen, Zhiyuan; Yang, Yong; Ju, Bin; Tang, Lingling

doi:10.21203/rs.3.rs-77820/v1

Cited by 16 publications

(29 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In future research, we hope to include more patients, while expanding to other cause-specific ARDS, such as COVID-19, as the current ARDS predictions for patients with COVID-19 are either risk factor analysis [35], real-time risk scoring systems [36], and/or utilising relatively complex predictors, such as the neutrophil-to-lymphocyte ratio [37]. We hope to accomplish this aim by integrating this early prediction model in our own ICU risk management system to evaluate its effectiveness.…”

Section: Discussionmentioning

confidence: 99%

Early prediction of cause-specific acute respiratory distress syndrome via interpretable machine-learning.

Jing

Liu²,

Xie

et al. 2021

Preprint

View full text Add to dashboard Cite

Background: Several studies have investigated the correlation between physiological parameters and the risk of acute respiratory distress syndrome (ARDS); however, cause-specific ARDS and its early prediction have not been well-studied. We aimed to develop and validate a machine-learning model for the early prediction of inhalation-induced ARDS. Methods: Clinical expertise was applied with data-driven analysis. Using data from electronic intensive care units (retrospective derivation cohort) and the three most accessible vital signs (i.e. heart rate, temperature, and respiratory rate) together with feature engineering, we applied a random forest approach during the time window of 90 hours that ended 6 hours prior to the onset of moderate-to-severe respiratory failure (the ratio of partial pressure of arterial oxygen to fraction of inspired oxygen ≤200 mmHg). Results: The trained random forest classifier was validated using an independent validation cohort, with an area under the curve of 0.9026 (95% confidence interval 0.8075-1) and optimal threshold probability, the accuracy, sensitivity, and specificity values were 0.8947, 1, and 0.875, respectively. A Stable and Interpretable RUle Set (SIRUS) was used to extract rules from the RF to provide guidelines for clinicians. We identified several predictive factors, including resp_96h_6h_min < 9, resp_96h_6h_mean ≥ 16.1, HR_96h_6h_mean ≥ 102, and temp_96h_6h_max > 100, that could be used for predicting inhalation-induced ARDS (moderate-to-severe condition) 6 hours prior to onset in critical care units. (‘xxx_96h_6h_min/mean/max’: the minimum/mean/maximum values of the xxx vital sign collected during a 90 h time window beginning 96 h prior to the onset of ARDS and ending 6 h prior to the onset from every recorded blood gas test) Conclusions: This newly established random forest‑based interpretable model shows good predictive ability for inhalation-induced ARDS and may assist clinicians in decision-making, as well as facilitate the enrolment of patients in prevention programmes to improve their outcomes.

show abstract

Section: Discussionmentioning

confidence: 99%

Early prediction of cause-specific acute respiratory distress syndrome via interpretable machine-learning.

Jing

Liu²,

Xie

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…This section presents two different evaluations of our model's ability to account for important clinical factors. First, we begin with profiling the model's predicted risk score distribution in terms of well-established univariate risk factors in the clinical literature 28,29 , namely the chronic conditions extracted from clinical notes and demographics information such as age, gender, and sex. Next, we examine the ability of the model to account for important multi-comorbidities.…”

Section: Model Interpretability Via Single and Multicomorbidity Analysismentioning

confidence: 99%

“…This provides a better understanding of the severity of a patient's condition, and (ii) predict the likelihood of a patient requiring mechanical ventilation. Collectively, these prediction tasks capture the inherent challenges of inpatient resource planning such as those to predict which patients are most likely to experience poor outcomes over a span of next 3-7 days 9,[27][28][29][30] .…”

Section: Introductionmentioning

confidence: 99%

“…We also present a new methodology to interpret and evaluate model predictions via multi-comorbidity analysis. Much of the literature on EHR models focus on characterizing the prediction performance through univariate analysis of the well understood risk factors 29,33 . However, clinical presentation of a disease and it's severity can markedly vary depending on the constellation of symptoms, prior health conditions and risk factors.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Preparing For The Next Pandemic: Transfer Learning From Existing Diseases Via Hierarchical Multi-Modal BERT Models to Predict COVID-19 Outcomes

Agarwal

Choudhury

Tipirneni

et al. 2021

Preprint

View full text Add to dashboard Cite

Developing prediction models for emerging infectious diseases from relatively small numbers of cases is a critical need for improving pandemic preparedness. Using COVID-19 as an exemplar, we propose a transfer learning methodology for developing predictive models from multi-modal electronic healthcare records by leveraging information from more prevalent diseases with shared clinical characteristics. Our novel hierarchical, multi-modal model (TransMED) integrates baseline risk factors from the natural language processing of clinical notes at admission, time-series measurements of biomarkers obtained from laboratory tests, and discrete diagnostic, procedure and drug codes. We demonstrate the alignment of TransMED's predictions with well-established clinical knowledge about COVID-19 through univariate and multivariate risk factor driven sub-cohort analysis. TransMED's superior performance over state-of-the-art methods shows that leveraging patient data across modalities and transferring prior knowledge from similar disorders is critical for accurate prediction of patient outcomes, and this approach may serve as an important tool in the early response to future pandemics.

show abstract

“…AI has been extensively applied to analyze various COVID-19 data, including to aid diagnostics and in therapeutic design [2]. In RWD AI, machine learning has been used to predict the probability of ARDS based on the clinical characteristics of COVID patients [3]. A further study, on 3,194 COVID-19 cases in the Emory Healthcare network, assessed whether a COVID-19 patient's need for hospitalization can be predicted at the time of their RT-PCR test using electronic medical records data prior to the test [4].…”

Section: Introduction Background and Significancementioning

confidence: 99%

Ondansetron use is associated with lower COVID-19 mortality in a Real-World Data network-based analysis

Miller

Ellis

Sarangarajan

et al. 2021

Preprint

View full text Add to dashboard Cite

Objective: The COVID-19 pandemic generated a massive amount of clinical data, which potentially holds yet undiscovered answers related to COVID-19 morbidity, mortality, long term effects, and therapeutic solutions. The objective of this study was to generate insights on COVID-19 mortality-associated factors and identify potential new therapeutic options for COVID-19 patients by employing artificial intelligence analytics on real-world data. Materials and Methods: A Bayesian statistics-based artificial intelligence data analytics tool (bAIcis®) within Interrogative Biology® platform was used for network learning, inference causality and hypothesis generation to analyze 16,277 PCR positive patients from a database of 279,281 inpatients and outpatients tested for SARS-CoV-2 infection by antigen, antibody, or PCR methods during the first pandemic year in Central Florida. This approach generated causal networks that enabled unbiased identification of significant predictors of mortality for specific COVID-19 patient populations. These findings were validated by logistic regression, regression by least absolute shrinkage and selection operator, and bootstrapping. Results: We found that in the SARS-CoV-2 PCR positive patient cohort, early use of the antiemetic agent ondansetron was associated with increased survival in mechanically ventilated patients. Conclusions: The results demonstrate how real world COVID-19 focused data analysis using artificial intelligence can generate valid insights that could possibly support clinical decision-making and minimize the future loss of lives and resources.

show abstract

Risk Factors Analysis of COVID-19 Patients with ARDS and Prediction Based on Machine Learning

Cited by 16 publications

References 35 publications

Early prediction of cause-specific acute respiratory distress syndrome via interpretable machine-learning.

Early prediction of cause-specific acute respiratory distress syndrome via interpretable machine-learning.

Preparing For The Next Pandemic: Transfer Learning From Existing Diseases Via Hierarchical Multi-Modal BERT Models to Predict COVID-19 Outcomes

Ondansetron use is associated with lower COVID-19 mortality in a Real-World Data network-based analysis

Contact Info

Product

Resources

About