Clinical and inflammatory features based machine learning model for fatal risk prediction of hospitalized COVID-19 patients: results from a retrospective cohort study

Guan, Xin; Zhang, Bo; Fu, Ming; Li, Mengying; Xu, Yuan; Zhu, Yaowu; Peng, Jing; Guo, Huan; Lu, Yanjun

doi:10.1080/07853890.2020.1868564

Cited by 99 publications

(87 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Concerning disease biomarkers, a recent study has demonstrated that laboratory parameters such as CRP, LDH or Ferritin could be considered as predictors of worsening and/or death in COVID-19 patients. 35 The study demonstrated that these parameters were higher in patients who did not survive to COVID-19 disease. In fact, a statistically significant increase was observed in the group of 72 non-survivors among the 1270 patients evaluated with increase in CRP (+76.2 mg/L, +76.0%), LDH (+175 U/L, +39.5%) and Ferritin (+878.5 ng/mL, +61.5%).…”

Section: Discussionmentioning

confidence: 88%

Potential Clinical Benefits of Quercetin in the Early Stage of COVID-19: Results of a Second, Pilot, Randomized, Controlled and Open-Label Clinical Trial

Pierro¹,

Iqtadar²,

Khan³

et al. 2021

IJGM

124

View full text Add to dashboard Cite

Background The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is the cause of the ongoing global pandemic known as COVID-19. Based on the potential antiviral role of quercetin, and on its described anti-blood clotting, anti-inflammatory and antioxidant properties, we hypothesize that subjects with mild COVID-19 treated with Quercetin Phytosome ® (QP), a novel bioavailable form of quercetin, may have a shorter time to virus clearance, a milder symptomatology, and higher probabilities of a benign earlier resolution of the disease. Methods In our 2-week, randomized, open-label, and controlled clinical study, we have enrolled 42 COVID-19 outpatients. Twenty-one have been treated with the standard of care (SC), and 21 with QP as add-on supplementation to the SC. Our main aims were to check virus clearance and symptoms. Results The interim results reveal that after 1 week of treatment, 16 patients of the QP group were tested negative for SARS-CoV-2 and 12 patients had all their symptoms diminished; in the SC group, 2 patients were tested SARS-CoV-2 negative and 4 patients had their symptoms partially improved. By 2 weeks, the remaining 5 patients of the QP group tested negative for SARS-CoV-2, whereas in the SC group out of 19 remaining patients, 17 tested negatives by week 2, one tested negative by week 3 and one patient, still positive, expired by day 20. Concerning blood parameters, the add on therapy with QP, reduced LDH (−35.5%), Ferritin (−40%), CRP (−54.8%) and D-dimer (−11.9%). Conclusion QP statistically shortens the timing of molecular test conversion from positive to negative, reducing at the same time symptoms severity and negative predictors of COVID-19.

show abstract

Section: Discussionmentioning

confidence: 88%

Potential Clinical Benefits of Quercetin in the Early Stage of COVID-19: Results of a Second, Pilot, Randomized, Controlled and Open-Label Clinical Trial

Pierro¹,

Iqtadar²,

Khan³

et al. 2021

IJGM

124

View full text Add to dashboard Cite

show abstract

“…In the field of clinical diagnosis of COVID-19, these predictive analyses grounded on biomarkers can help optimise the screening of patients with severe disease, minimising mortality and hospitalisation, and reducing care delays. Previous machine learning studies highlight that some demographic variables, patients’ comorbidities, and laboratory findings can be predictive factors for COVID-19 mortality [ [9] , [10] , [11] , [12] ]. However, most of these studies included a small sample size, which may impact the model’s robustness and reliability of findings (e.g., low sensitivity) [ [14] , [15] , [16] ] and prevent its use in practice.…”

Section: Introductionmentioning

confidence: 99%

Diagnosis and prediction of COVID-19 severity: can biochemical tests and machine learning be used as prognostic indicators?

Cobre

Stremel

Noleto

et al. 2021

Computers in Biology and Medicine

View full text Add to dashboard Cite

Objective This study aimed to implement and evaluate machine learning based-models to predict COVID-19’ diagnosis and disease severity. Methods COVID-19 test samples (positive or negative results) from patients who attended a single hospital were evaluated. Patients diagnosed with COVID-19 were categorised according to the severity of the disease. Data were submitted to exploratory analysis (principal component analysis, PCA) to detect outlier samples, recognise patterns, and identify important variables. Based on patients’ laboratory tests results, machine learning models were implemented to predict disease positivity and severity. Artificial neural networks (ANN), decision trees (DT), partial least squares discriminant analysis (PLS-DA), and K nearest neighbour algorithm (KNN) models were used. The four models were validated based on the accuracy (area under the ROC curve). Results The first subset of data had 5,643 patient samples (5,086 negatives and 557 positives for COVID-19). The second subset included 557 COVID-19 positive patients. The ANN, DT, PLS-DA, and KNN models allowed the classification of negative and positive samples with >84% accuracy. It was also possible to classify patients with severe and non-severe disease with an accuracy >86%. The following were associated with the prediction of COVID-19 diagnosis and severity: hyperferritinaemia, hypocalcaemia, pulmonary hypoxia, hypoxemia, metabolic and respiratory acidosis, low urinary pH, and high levels of lactate dehydrogenase. Conclusion Our analysis shows that all the models could assist in the diagnosis and prediction of COVID-19 severity.

show abstract

“…In particular, Ramiro and colleagues showed that patients with ferritin levels higher that the median value of 1,419 µg/L had the better benefit from tocilizumab plus corticosteroids ( 46 ). Despite such robust evidence, ferritin has been rarely incorporated in COVID-19 simple multiparameter prognostic scores ( 48 ), with notable exceptions represented by complex algorithms obtained through machine learning approaches (MLA) ( 49 , 50 ). For example, Kar and colleagues applied MLA to 1,393 hospitalized patients, obtaining a multivariable mortality risk score that was prospectively validated in further 977 patients ( 50 ).…”

Section: Introductionmentioning

confidence: 99%

Iron metabolism in infections: Focus on COVID-19

Girelli

Marchi

Busti

et al. 2021

Seminars in Hematology

View full text Add to dashboard Cite

Iron is a micronutrient essential for a wide range of metabolic processes in virtually all living organisms. During infections, a battle for iron takes place between the human host and the invading pathogens. The liver peptide hepcidin, which is phylogenetically and structurally linked to defensins (antimicrobial peptides of the innate immunity), plays a pivotal role by subtracting iron to pathogens through its sequestration into host cells, mainly macrophages. While this phenomenon is well studied in certain bacterial infections, much less is known regarding viral infections. Iron metabolism also has implications on the functionality of cells of the immune system. Once primed by the contact with antigen presenting cells, lymphocytes need iron to sustain the metabolic burst required for mounting an effective cellular and humoral response. The COVID-19 pandemic has boosted an amount of clinical and translational research over the possible influences of nutrients on SARS-CoV-2 infection, in terms of either susceptibility or clinical course. Here we review the intersections between iron metabolism and COVID-19, belonging to the wider domain of the so-called “nutritional immunity”. A better understanding of such connections has potential broad implications, either from a mechanistic standpoint, or for the development of more effective strategies for managing COVID-19 and possible future pandemics.

show abstract

Clinical and inflammatory features based machine learning model for fatal risk prediction of hospitalized COVID-19 patients: results from a retrospective cohort study

Abstract: 2021) Clinical and inflammatory features based machine learning model for fatal risk prediction of hospitalized COVID-19 patients: results from a retrospective cohort study,

Cited by 99 publications

References 38 publications

Potential Clinical Benefits of Quercetin in the Early Stage of COVID-19: Results of a Second, Pilot, Randomized, Controlled and Open-Label Clinical Trial

Potential Clinical Benefits of Quercetin in the Early Stage of COVID-19: Results of a Second, Pilot, Randomized, Controlled and Open-Label Clinical Trial

Diagnosis and prediction of COVID-19 severity: can biochemical tests and machine learning be used as prognostic indicators?

Iron metabolism in infections: Focus on COVID-19

Contact Info

Product

Resources

About