Biomarkers of severe COVID-19 pneumonia on admission using data-mining powered by common laboratory blood tests-datasets

Biomarkers for SARS-CoV-2 infection. A narrative review

COVID-19 is an infectious disease caused by SARS-CoV-2 with devastating effects on health-care systems. The magnitude of the problem has moved physicians and investigators to identify strategies to detect patients at a high risk of severe disease. The aim of this study was to identify the most relevant biomarkers in the published literature and their correlation with clinical outcomes. To this end, we performed a revision of studies that investigated laboratory abnormalities in patients with COVID-19, comparing non-severe and severe patients. Blood biomarkers were classified into five main categories: hematological, coagulation related to the liver or kidney, and inflammatory. From our analysis, the most relevant biomarkers associated with severe infection for each category were increased levels of leukocytes, neutrophils, and neutrophil-to-lymphocyte ratio; decreased platelet count; and high levels of aspartate transaminase, alanine transaminase, creatine kinase, troponin, creatinine, and blood urea nitrogen, C-reactive protein, ferritin, and IL-6. Moreover, lactate dehydrogenase and D-dimer levels were independent risk factors for death.

Red cell distribution width-to-platelet ratio (RPR) as a predictor of prolonged stay at hospital for COVID-19 inpatients

AIMS/BACKGROUND

We looked at novel hematological composites like the neutrophil-to-lymphocyte ratio, platelet-to-lymphocyte ratio, monocyte-to-lymphocyte ratio, red cell distribution width-to-lymphocyte ratio, red cell distribution width-to-platelet ratio, leukocyte-to-C reactive protein ratio, and lymphocyte-to-C reactive protein ratio as explanatory variables for COVID-19 patients´ hospital length of stay (LoS).

METHODS

The association of hematological indices with LoS was analyzed on 2930 COVID-19 patients using the univariate and multivariable Cox proportional hazards regression models with enter method. The Kaplan-Meier survival estimates were applied to LoS.

RESULTS

The survivors´ mean LoS was 7.8 ± 24.0 days, but the deaths´ mean LoS was 38.6 ± 41.9 days (W = 31338, p < 0.01). Every hematological scores representative of the inflammatory status was significantly correlated in the univariate analysis with a prolonged LoS (p < 0.001). In the multivariate analysis, it was discovered that just the monocyte-to-lymphocyte and lymphocyte-to-C reactive protein ratios had not achieved statistical significance. However, most systemic inflammation measures showed hazards ratios close to one. One exemption was the red cell distribution width-to-platelet ratio (RPR) index, which can increase the probability of a longer hospital stay by up to ten times (HR(IC95%) = 0.092(0.03-0.29); p < 0.001).

CONCLUSION

The most effective biomarker to identify COVID-19 patients at high risk for extended hospital stay was RPR.

HIGHLIGHTS

IntroductionDetermining hospital Length of Stay (LoS) is vital for resource management, especially for future COVID-19 outbreaks.Previous studies have primarily focused on sociodemographic and clinical attributes, along with resource availability, but have not accounted for other factors like routine laboratory tests, which can significantly impact LoS predictions.This study examines novel hematology scores as predictors of LoS, emphasizing their importance in resource-limited settings like Ecuador.MethodsThis retrospective cohort study analyzed 2,930 COVID-19 patients admitted to Hospital IESS Quito Sur in Ecuador focusing on confirmed cases with complete blood count (CBC) values to assess LoS.The study explored various hematological ratios, such as the neutrophil-to-lymphocyte ratio (NLR) and red cell distribution width-to-lymphocyte ratio (RLR), as potential predictors of LoS and in-hospital outcomes for COVID-19 patients, using a combination of univariate and multivariable Cox proportional hazards regression models.Kaplan-Meier survival estimates and log-rank tests were used to analyze survival and discharge probabilities over time, highlighting sex-dependent effects and the significant association between selected hematological indices and patient outcomes.ResultsThe mean LoS for survivors was significantly shorter compared to those who died (p < 0.001), indicating that longer hospitalization was associated with higher mortality risk.Women had a shorter average LoS (with lower mortality risk (p < 0.001), suggesting an asymmetrical hospitalization pattern based on sex.While most hematological markers had minimal clinical relevance for LoS, the red cell distribution width-to-platelet ratio (RPR) stood out, increasing the likelihood of a longer hospital stay by up to tenfold, making it a critical factor in predicting prolonged hospitalization.DiscussionMen had longer hospital stays and higher mortality rates than women, likely due to differing inflammatory responses, though hyperinflammation markers like NLR, PLR, and Leu-CPR had minimal clinical impact.RPR had the strongest link to longer hospital stays, indicating a higher risk of extended hospitalization in severe COVID-19 cases.Elevated RPR is tied to oxidative stress and coagulation issues, suggesting early identification could help reduce prolonged stays and complications.ConclusionResearch generally points to clinical complications from COVID-19 as the main factor behind prolonged hospitalizations, underlining the importance of early identification and management of these issues.

Insights into the baseline blood pH homeostasis at admission and the risk of in-hospital mortality in COVID-19 patients

Aim: A laboratory finding in critically ill COVID-19 patients is blood academia (pH <7.35). We investigated its cause in connection with the admission baseline blood pH homeostasis.Patients & methods: We retrospectively monitored the baseline blood pH homeostasis of 1215 COVID-19 patients who were admitted with pneumonia using data-driven knowledge. Two categories of patients were identified: non-survivors (107) and survivors (1108).Results: Non-survivors showed greater levels of lactate and lower blood pH, saturation, and partial pressure of oxygen than survivors. A bivariate Spearman's correlation matrix showed that the [HCO3-]/pCO2 and pCO2 of non-survivors exhibited an unmatched connection, but not in the survivor group. When comparing non-survivors to survivors, the dendrograms derived from the bivariate comparison matrix showed differences in gasometry parameters like blood pH, [HCO3-]/pCO2 ratio, anion gap and pO2.Conclusion: The little variations in the gasometry readings between survivors and non-survivors upon admission suggested abnormal changes in the complementary renal and respiratory systems that bring blood pH back to normal. In advanced COVID-19, modest blood acid-base imbalances could become blood acidemia if these compensatory strategies were overused. Data-driven monitoring of acid-base parameters may help predict abnormal blood pH and the advancement of metabolic acidemia before it is too late.

Discovering associations between radiological features and COVID-19 patients' deterioration

Background and Aims

Data mining methods are effective and well-known tools for developing predictive models and extracting useful information from various data of patients. The present study aimed to predict the severity of patients with COVID-19 by applying the rule mining method using characteristics of medical images.

Methods

This retrospective study has analyzed the radiological data from 104 COVID-19 hospitalized patients diagnosed with COVID-19 in a hospital in Iran. A data set containing 75 binary features was generated. Apriori method is utilized for association rule mining on this data set. Only rules with confidence equal to one were generated. The performance of rules is calculated by support, coverage, and lift indexes.

Results

Ten rules were extracted with only X-ray-related features on cases referred to ICU. The Support and Coverage index of all of these rules was 0.087, and the Lift index of them was 1.58. Thirteen rules were extracted from only CT scan-related features on cases referred to ICU. The CXR_Pleural effusion feature has appeared in all the rules. The CXR_Left upper zone feature appears in 9 rules out of 10. The Support and Coverage index of all rules was 0.15, and the Lift index of all rules was 1.63. the CT_Adjacent pleura thickening feature has appeared in all rules, and the CT_Right middle lobe appeared in 9 rules out of 13.

Conclusion

This study could reveal the application and efficacy of CXR and CT scan imaging modalities in predicting ICU admission to a major COVID-19 infection via data mining methods. The findings of this study could help data scientists, radiologists, and clinicians in the future development and implementation of these methods in similar conditions and timely and appropriately save patients from adverse disease outcomes.

Whale optimization with random contraction and Rosenbrock method for COVID-19 disease prediction

Coronavirus Disease 2019 (COVID-19), instigated by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), has hugely impacted global public health. To identify and intervene in critically ill patients early, this paper proposes an efficient, intelligent prediction model based on the machine learning approach, which combines the improved whale optimization algorithm (RRWOA) with the k-nearest neighbor (KNN) classifier. In order to improve the problem that WOA is prone to fall into local optimum, an improved version named RRWOA is proposed based on the random contraction strategy (RCS) and the Rosenbrock method. To verify the capability of the proposed algorithm, RRWOA is tested against nine classical metaheuristics, nine advanced metaheuristics, and seven well-known WOA variants based on 30 IEEE CEC2014 competition functions, respectively. The experimental results in mean, standard deviation, the Friedman test, and the Wilcoxon signed-rank test are considered, proving that RRWOA won first place on 18, 24, and 25 test functions, respectively. In addition, a binary version of the algorithm, called BRRWOA, is developed for feature selection problems. An efficient prediction model based on BRRWOA and KNN classifier is proposed and compared with seven existing binary metaheuristics based on 15 datasets of UCI repositories. The experimental results show that the proposed algorithm obtains the smallest fitness value in eleven datasets and can solve combinatorial optimization problems, indicating that it still performs well in discrete cases. More importantly, the model was compared with five other algorithms on the COVID-19 dataset. The experiment outcomes demonstrate that the model offers a scientific framework to support clinical diagnostic decision-making. Therefore, RRWOA is an effectively improved optimizer with efficient value.

Automatic detection of A-line in lung ultrasound images using deep learning and image processing

BACKGROUND

Auxiliary diagnosis and monitoring of lung diseases based on lung ultrasound (LUS) images is important clinical research. A-line is one of the most common indicators of LUS that can offer support for the assessment of lung diseases. A traditional A-line detection method mainly relies on experienced clinicians, which is inefficient and cannot meet the needs of these areas with backward medical level. Therefore, how to realize the automatic detection of A-line in LUS image is important.

PURPOSE

In order to solve the disadvantages of traditional A-line detection methods, realize automatic and accurate detection, and provide theoretical support for clinical application, we proposed a novel A-line detection method for LUS images with different probe types in this paper.

METHODS

First, the improved Faster R-CNN model with a selection strategy of localization box was designed to accurately locate the pleural line. Then, the LUS image below the pleural line was segmented for independent analysis excluding the influence of other similar structures. Next, image-processing methods based on total variation, matched filter, and gray difference were applied to achieve the automatic A-line detection. Finally, the "depth" index was designed to verify the accuracy by judging whether the automatic measurement results belong to corresponding manual results (±5%). In experiments, 3000 convex array LUS images were used for training and validating the improved pleural line localization model by five-fold cross validation. 850 convex array LUS images and 1080 linear array LUS images were used for testing the trained pleural line localization model and the proposed image-processing-based A-line detection method. The accuracy analysis, error statistics, and Harsdorff distance were employed to evaluate the experimental results.

RESULTS

After 100 epochs, the mean loss value of training and validation set of improved Faster R-CNN model reached 0.6540 and 0.7882, with the validation accuracy of 98.70%. The trained pleural line localization model was applied in the testing set of convex and linear probes and reached the accuracy of 97.88% and 97.11%, respectively, which were 3.83% and 8.70% higher than the original Faster R-CNN model. The accuracy, sensitivity, and specificity of A-line detection reached 95.41%, 0.9244%, 0.9875%, and 94.63%, 0.9230%, and 0.9766% for convex and linear probes, respectively. Compared to the experienced clinicians' results, the mean value and p value of depth error were 1.5342 ± 1.2097 and 0.9021, respectively, and the Harsdorff distance was 5.7305 ± 1.8311. In addition, the accumulated accuracy of the two-stage experiment (pleural line localization and A-line detection) was calculated as the final accuracy of the whole A-line detection system. They were 93.39% and 91.90% for convex and linear probes, respectively, which were higher than these previous methods.

CONCLUSIONS

The proposed method combining image processing and deep learning can automatically and accurately detect A-line in LUS images with different probe types, which has important application value for clinical diagnosis.

Predictive models for COVID-19 detection using routine blood tests and machine learning

The problem of accurate, fast, and inexpensive COVID-19 tests has been urgent till now. Standard COVID-19 tests need high-cost reagents and specialized laboratories with high safety requirements, are time-consuming. Data of routine blood tests as a base of SARS-CoV-2 invasion detection allows using the most practical medicine facilities. But blood tests give general information about a patient's state, which is not directly associated with COVID-19. COVID-19-specific features should be selected from the list of standard blood characteristics, and decision-making software based on appropriate clinical data should be created. This review describes the abilities to develop predictive models for COVID-19 detection using routine blood tests and machine learning.

Predicting Blood Parasite Load and Influence of Expression of iNOS on the Effect Size of Clinical Laboratory Parameters in Acute Trypanosoma cruzi Infection With Different Inoculum Concentrations in C57BL/6 Mice

In Chagas disease, the initial responses of phagocyte-mediated innate immunity are strongly associated with the control of Trypanosoma cruzi and are mediated by various signaling pathways, including the inducible nitric oxide synthetase (iNOS) pathway. The clinical and laboratory manifestations of Chagas disease depend on the parasite–host relationship, i.e., the responsive capacity of the host immune system and the immunogenicity of the parasite. Here, we evaluated effect sizes in clinical and laboratory parameters mediated by acute infection with different concentrations of T. cruzi inoculum in mice immunosuppressed via iNOS pathway inactivation. Infection was induced in C57BL/6 wild-type and iNOS^-/- mice with the “Y” strain of T. cruzi at three inoculum concentrations (3 × 10², 3 × 10³, and 3 × 10⁴). Parasitemia and mortality in both mouse strains were monitored. Immunohistochemistry was performed to quantify amastigotes in cardiac tissues and cardiac musculature cells. Biochemical parameters, such as blood urea nitrogen, sodium, albumin, and globulin concentrations, among others, were measured, and cytokine concentrations were also measured. Effect sizes were determined by the eta squared formula. Compared with that in wild-type animals, mice with an absence of iNOS expression demonstrated a greater parasite load, with earlier infection and a delayed parasitemia peak. Inoculum concentration was positively related to death in the immunosuppressed subgroup. Nineteen parameters (hematological, biochemical, cytokine-related, and histopathological) in the immunocompetent subgroup and four in the immunosuppressed subgroup were associated with parasitemia. Parasitemia, biochemical parameters, and hematological parameters were found to be predictors in the knockout group. The impact of effect sizes on the markers evaluated based on T. cruzi inoculum concentration was notably high in the immunocompetent group (Cohen’s d = 88.50%; p <.001). These findings contribute to the understanding of physiopathogenic mechanisms underlying T. cruzi infection and also indicate the influence of the concentration of T. cruzi during infection and the immunosuppression through the iNOS pathway in clinical laboratory heterogeneity reported in acute Chagas disease.

Challenges of Multiplex Assays for COVID-19 Research: A Machine Learning Perspective

Multiplex assays that provide simultaneous measurement of multiple analytes in biological samples have now developed into widely used technologies in the study of diseases, drug discovery, and other medical areas. These approaches span multiple assay systems and can provide readouts of specific assay components with similar accuracy as the respective single assay measurements. Multiplexing allows the consumption of lower sample volumes, lower costs, and higher throughput compared with carrying out single assays. A number of recent studies have demonstrated the impact of multiplex assays in the study of the SARS-CoV-2 virus, the infectious agent responsible for the current COVID-19 pandemic. In this respect, machine learning techniques have proven to be highly valuable in capturing complex disease phenotypes and converting these insights into models which can be applied in real-world settings. This chapter gives an overview of opportunities and challenges of multiplexed biomarker analysis, with a focus on the use of machine learning aimed at identification of biological signatures for increasing our understanding of COVID-19 disease, and for improved diagnostics and prediction of disease outcomes.

Artificial intelligence approach towards assessment of condition of COVID-19 patients - Identification of predictive biomarkers associated with severity of clinical condition and disease progression

BACKGROUND AND OBJECTIVES

Although ML has been studied for different epidemiological and clinical issues as well as for survival prediction of COVID-19, there is a noticeable shortage of literature dealing with ML usage in prediction of disease severity changes through the course of the disease. In that way, predicting disease progression from mild towards moderate, severe and critical condition, would help not only to respond in a timely manner to prevent lethal results, but also to minimize the number of patients in hospitals where this is not necessary.

METHODS

We present a methodology for the classification of patients into 4 distinct categories of the clinical condition of COVID-19 disease. Classification of patients is based on the values of blood biomarkers that were assessed by Gradient boosting regressor and which were selected as biomarkers that have the greatest influence in the classification of patients with COVID-19.

RESULTS

The results show that among several tested algorithms, XGBoost classifier achieved best results with an average accuracy of 94% and an average F1-score of 94.3%. We have also extracted 10 best features from blood analysis that are strongly associated with patient condition and based on those features we can predict the severity of the clinical condition.

CONCLUSIONS

The main advantage of our system is that it is a decision tree-based algorithm which is easier to interpret, instead of the use of black box models, which are not appealing in medical practice.