A Comparison of Decision Tree Algorithms in the Assessment of Biomedical Data

Use of machine learning to identify protective factors for death from COVID-19 in the ICU: a retrospective study

Background

Patients in serious condition due to COVID-19 often require special care in intensive care units (ICUs). This disease has affected over 758 million people and resulted in 6.8 million deaths worldwide. Additionally, the progression of the disease may vary from individual to individual, that is, it is essential to identify the clinical parameters that indicate a good prognosis for the patient. Machine learning (ML) algorithms have been used for analyzing complex medical data and identifying prognostic indicators. However, there is still an urgent need for a model to elucidate the predictors related to patient outcomes. Therefore, this research aimed to verify, through ML, the variables involved in the discharge of patients admitted to the ICU due to COVID-19.

Methods

In this study, 126 variables were collected with information on demography, hospital length stay and outcome, chronic diseases and tumors, comorbidities and risk factors, complications and adverse events, health care, and vital indicators of patients admitted to an ICU in southern Brazil. These variables were filtered and then selected by a ML algorithm known as decision trees to identify the optimal set of variables for predicting patient discharge using logistic regression. Finally, a confusion matrix was performed to evaluate the model's performance for the selected variables.

Results

Of the 532 patients evaluated, 180 were discharged: female (16.92%), with a central venous catheter (23.68%), with a bladder catheter (26.13%), and with an average of 8.46- and 23.65-days using bladder catheter and submitted to mechanical ventilation, respectively. In addition, the chances of discharge increase by 14% for each additional day in the hospital, by 136% for female patients, 716% when there is no bladder catheter, and 737% when no central venous catheter is used. However, the chances of discharge decrease by 3% for each additional year of age and by 9% for each other day of mechanical ventilation. The performance of the training data presented a balanced accuracy of 0.81, sensitivity of 0.74, specificity of 0.88, and the kappa value was 0.64. The test performance had a balanced accuracy of 0.85, sensitivity 0.75, specificity 0.95, and kappa value of 0.73. The McNemar test found that there were no significant differences in the error rates in the training and test data, suggesting good classification. This work showed that female, the absence of a central venous catheter and bladder catheter, shorter mechanical ventilation, and bladder catheter duration were associated with a greater chance of hospital discharge. These results may help develop measures that lead to a good prognosis for the patient.

MMP14high macrophages orchestrate progressive pulmonary fibrosis in SR-Ag-induced hypersensitivity pneumonitis

Fibrotic hypersensitivity pneumonitis (FHP) is a fatal interstitial pulmonary disease with limited treatment options. Lung macrophages are a heterogeneous cell population that exhibit distinct subsets with divergent functions, playing pivotal roles in the progression of pulmonary fibrosis. However, the specific macrophage subpopulations and underlying mechanisms involved in the disease remain largely unexplored. In this study, a decision tree model showed that matrix metalloproteinase-14 (MMP14) had higher scores for important features in the up-regulated genes in macrophages from mice exposed to the Saccharopolyspora rectivirgula antigen (SR-Ag). Using single-cell RNA sequencing (scRNA-seq) analysis of hypersensitivity pneumonitis (HP) mice profiles, we identified MMP14high macrophage subcluster with a predominant M2 phenotype that exhibited higher activity in promoting fibroblast-to myofibroblast transition (FMT). We demonstrated that suppressing toll-like receptor 2 (TLR2) and nuclear factor kappa-B (NF-κB) could attenuate MMP14 expression and exosome secretion in macrophages stimulation with SR-Ag. The exosomes derived from MMP14-overexpressing macrophages were found to be more effective in regulating the transition of fibroblasts through exosomal MMP14. Importantly, it was observed that the transfer of MMP14-overexpressing macrophages into mice promoted lung inflammation and fibrosis induced by SR-Ag. NSC-405020 binding to the hemopexin domain (PEX) of MMP-14 ameliorated lung inflammation and fibrosis induced by SR-Ag in mice. Thus, MMP14-overexpressing macrophages may be an important mechanism contributing to the exacerbation of allergic reactions. Our results indicated that MMP14 in macrophages has the potential to be a therapeutic target for HP.

Retracted: A Comparison of Decision Tree Algorithms in the Assessment of Biomedical Data

[This retracts the article DOI: 10.1155/2022/9449497.].

IMPLEMENTATION OF THE DECISION TREE METHOD IN EXPERT ANALYSIS OF THE MEDICAL ERRORS IN OBSTETRIC PRACTICE

OBJECTIVE

Aim: To identify expert patterns in cases of improper medical care in obstetric practice based on the analysis of the materials of judicial and investigative cases initiated against obstetrician-gynaecologists in cases of improper performance of their professional duties, using the decision tree method.

PATIENTS AND METHODS

Materials and methods: A retrospective review of all alleged medical malpractice cases (a total 350) between 2007 and 2016 handled at Ternopil Regional Bu¬reau of Forensic Medical Examination, Chernivtsi Regional Bureau of Forensic Medical Examination and Zhytomir Regional Bureau of Forensic Medical Examination (Ukraine) was performed.

RESULTS

Results: Expert commissions confirmed various shortcomings and omissions in provision of medical care in 232 (72.0%) of the investigated cases. Obstetricians were involved in claims in 82 (23.4%) cases. Application of intelligent data processing technology "Data Mining" with the use of the decision tree method revealed that inadequacies with regard to the medical records (attribute usage 100%) were the most informative attribute in the expert assessment of inappropriate medical care in obstetrics. Defects in the provision of obstetric care with a probability (P = 0.71) occur simultaneously both at pre-hospital and hospital levels and with a high probability (P = 0.83) result in severe consequences.

CONCLUSION

Conclusions: The use of modern technologies for data analysis and processing contributes to the formulation of mathematically substantiated statements that significantly enhance the reliability of expert opinions in cases of forensic medical examination attached to dereliction of duties by the medical practitioners.