Early Prediction of Seven-Day Mortality in Intensive Care Unit Using a Machine Learning Model: Results from the SPIN-UTI Project

Barchitta, Martina; Maugeri, Andrea; Favara, Giuliana; Riela, Paolo Marco; Gallo, Giovanni; Mura, I; Agodi, Antonella

doi:10.3390/jcm10050992

“…Overall, the evidence supported here and elsewhere [19] suggests that HCQ treatment may be more effective in specific subtypes of Covid-19 patients and indicates machine learning as a useful approach to identify the most "promising" patients in terms of success rate of this treatment. In the future, further studies on independent datasets are warranted, possibly using supervised ML techniques as in other clinical settings (e.g., [39,40]), to validate this hypothesis and test the feasibility of predicting responsiveness to HCQ before intervention. Ideally, a trial administering low dosages of HCQ (≤400 mg/day) and randomizing subjects based on their Covid-19 subtype profile rather than on single characteristics may be warranted to clarify the effects of HCQ on mortality risk in SARS-CoV-2 infection, especially within those patients with a "low risk" profile.…”

Section: Discussionmentioning

confidence: 99%

Disentangling the Association of Hydroxychloroquine Treatment with Mortality in Covid-19 Hospitalized Patients through Hierarchical Clustering

Castelnuovo

¹

,

Gialluisi

²

,

Antinori

³

et al. 2021

Journal of Healthcare Engineering

Self Cite

View full text Add to dashboard Cite

The efficacy of hydroxychloroquine (HCQ) in treating SARS-CoV-2 infection is harshly debated, with observational and experimental studies reporting contrasting results. To clarify the role of HCQ in Covid-19 patients, we carried out a retrospective observational study of 4,396 unselected patients hospitalized for Covid-19 in Italy (February–May 2020). Patients’ characteristics were collected at entry, including age, sex, obesity, smoking status, blood parameters, history of diabetes, cancer, cardiovascular and chronic pulmonary diseases, and medications in use. These were used to identify subtypes of patients with similar characteristics through hierarchical clustering based on Gower distance. Using multivariable Cox regressions, these clusters were then tested for association with mortality and modification of effect by treatment with HCQ. We identified two clusters, one of 3,913 younger patients with lower circulating inflammation levels and better renal function, and one of 483 generally older and more comorbid subjects, more prevalently men and smokers. The latter group was at increased death risk adjusted by HCQ (HR[CI95%] = 3.80[3.08-4.67]), while HCQ showed an independent inverse association (0.51[0.43-0.61]), as well as a significant influence of cluster∗HCQ interaction ( p < 0.001 ). This was driven by a differential association of HCQ with mortality between the high (0.89[0.65-1.22]) and the low risk cluster (0.46[0.39-0.54]). These effects survived adjustments for additional medications in use and were concordant with associations with disease severity and outcome. These findings suggest a particularly beneficial effect of HCQ within low risk Covid-19 patients and may contribute to clarifying the current controversy on HCQ efficacy in Covid-19 treatment.

show abstract

“…Finally, we did not explore the prognostic value by combining leucine and other independent predictors of adverse outcomes in ICU. There is a strong need for improving outcome prediction by contemporary techniques such as machine learning models [27].…”

Section: 1mentioning

confidence: 99%

U-Shape Relationship between Plasma Leucine Level and Mortality in the Intensive Care Unit

Wang

¹

,

Wang

²

,

Chen

³

et al. 2022

View full text Add to dashboard Cite

Patients in the intensive care unit (ICU) are at high risk of mortality which is not well predicted. Previous studies noted that leucine has prognostic value in a variety of diseases. This study investigated whether leucine concentration was a useful biomarker of metabolic and nutritional status and 6-month mortality in ICU. We recruited 454 subjects admitted to ICU (348 and 106 in the initiation and validation cohorts, respectively) with an acute physiology and chronic health evaluation (APACHE II) score ≥ 15 . We measured plasma leucine concentrations, traditional biomarkers, and calculated APACHE II and sequential organ failure assessment (SOFA) scores. Leucine levels were weakly correlated with albumin, prealbumin, and transferrin levels ( r = 0.30 , 0.12, and 0.15, p = 0.001 , 0.029, and 0.007, respectively). During follow-up, 116 (33.3%) patients died. Compared to patients with leucine levels between 109 and 174 μM, patients with leucine > 174 μ M or <109 μM had a lower cumulative survival rate. Death was also associated with age, higher APACHE II and SOFA scores, C-reactive protein, and longer stays in the ICU, but with lower albumin, prealbumin, and transferrin. Patients with leucine levels > 174 μ M had higher alanine aminotransferase levels, but no significant differences in other variables; patients with leucine levels < 109 μ M had higher APACHE II and SOFA scores, higher incidence of using inotropic agents, longer ICU and hospital stays, but lower albumin and transferrin levels. Multivariable analysis demonstrated that leucine > 174 μ M was an independent predictor of mortality, especially early mortality. However, among patients who stayed in ICU longer than two weeks, leucine < 109 μ M was an independent predictor of mortality. In addition, leucine < 109 μ M was associated with worse ventilator weaning profiles. These findings were similar in the validation cohort. Our study demonstrated a U-shape relationship between leucine levels and mortality rate in ICU.

show abstract

“…Notably, the main novelties of our approach concerned the analysis of HABSIs from different points of view (predictive and statistical analyses) in a cohort of 1203 patients at the level III NICU of the “Federico II” University Hospital in Naples from 2016 to 2020 (60 months). Despite different efforts in the literature for predicting the risk of HABSIs and other adverse outcomes in the ICU, these studies primarily focused on the use of Support Vector Machine (SVM) [ 16 , 17 ] or divided patients into clusters [ 18 ] without considering more recent Artificial Intelligence approaches (e.g., XGB or Catboost) or the large set of features in our analyses. Notably, our approach achieved better classification performances in accuracy and Area Under Curve (AUC) [ 16 , 17 ].…”

Section: Introductionmentioning

confidence: 99%

“…Despite different efforts in the literature for predicting the risk of HABSIs and other adverse outcomes in the ICU, these studies primarily focused on the use of Support Vector Machine (SVM) [ 16 , 17 ] or divided patients into clusters [ 18 ] without considering more recent Artificial Intelligence approaches (e.g., XGB or Catboost) or the large set of features in our analyses. Notably, our approach achieved better classification performances in accuracy and Area Under Curve (AUC) [ 16 , 17 ].…”

Section: Introductionmentioning

confidence: 99%

Predictive Analysis of Healthcare-Associated Blood Stream Infections in the Neonatal Intensive Care Unit Using Artificial Intelligence: A Single Center Study

Montella¹,

Ferraro²,

Sperlí

³

et al. 2022

IJERPH

View full text Add to dashboard Cite

Background: Neonatal infections represent one of the six main types of healthcare-associated infections and have resulted in increasing mortality rates in recent years due to preterm births or problems arising from childbirth. Although advances in obstetrics and technologies have minimized the number of deaths related to birth, different challenges have emerged in identifying the main factors affecting mortality and morbidity. Dataset characterization: We investigated healthcare-associated infections in a cohort of 1203 patients at the level III Neonatal Intensive Care Unit (ICU) of the “Federico II” University Hospital in Naples from 2016 to 2020 (60 months). Methods: The present paper used statistical analyses and logistic regression to identify an association between healthcare-associated blood stream infection (HABSIs) and the available risk factors in neonates and prevent their spread. We designed a supervised approach to predict whether a patient suffered from HABSI using seven different artificial intelligence models. Results: We analyzed a cohort of 1203 patients and found that birthweight and central line catheterization days were the most important predictors of suffering from HABSI. Conclusions: Our statistical analyses showed that birthweight and central line catheterization days were significant predictors of suffering from HABSI. Patients suffering from HABSI had lower gestational age and birthweight, which led to longer hospitalization and umbilical and central line catheterization days than non-HABSI neonates. The predictive analysis achieved the highest Area Under Curve (AUC), accuracy and F1-macro score in the prediction of HABSIs using Logistic Regression (LR) and Multi-layer Perceptron (MLP) models, which better resolved the imbalanced dataset (65 infected and 1038 healthy).

show abstract

Early Prediction of Seven-Day Mortality in Intensive Care Unit Using a Machine Learning Model: Results from the SPIN-UTI Project

Cited by 22 publications

References 69 publications

Disentangling the Association of Hydroxychloroquine Treatment with Mortality in Covid-19 Hospitalized Patients through Hierarchical Clustering

Disentangling the Association of Hydroxychloroquine Treatment with Mortality in Covid-19 Hospitalized Patients through Hierarchical Clustering

U-Shape Relationship between Plasma Leucine Level and Mortality in the Intensive Care Unit

Predictive Analysis of Healthcare-Associated Blood Stream Infections in the Neonatal Intensive Care Unit Using Artificial Intelligence: A Single Center Study

Contact Info

Product

Resources

About