Novel pneumonia score based on a machine learning model for predicting mortality in pneumonia patients on admission to the intensive care unit

Machine Learning Accurately Predicts Need for Critical Care Support in Patients Admitted to Hospital for Community-Acquired Pneumonia

OBJECTIVES

Hospitalized community-acquired pneumonia (CAP) patients are admitted for ventilation, vasopressors, and renal replacement therapy (RRT). This study aimed to develop a machine learning (ML) model that predicts the need for such interventions and compare its accuracy to that of logistic regression (LR).

DESIGN

This retrospective observational study trained separate models using random-forest classifier (RFC), support vector machines (SVMs), Extreme Gradient Boosting (XGBoost), and multilayer perceptron (MLP) to predict three endpoints: eventual use of invasive ventilation, vasopressors, and RRT during hospitalization. RFC-based models were overall most accurate in a derivation COVID-19 CAP cohort and were validated in one COVID-19 CAP and two non-COVID-19 CAP cohorts.

SETTING

This study is part of the Community-Acquired Pneumonia: Toward InnoVAtive Treatment (CAPTIVATE) Research program.

PATIENTS

Two thousand four hundred twenty COVID-19 and 1909 non-COVID-19 CAP patients over 18 years old hospitalized and not needing invasive ventilation, vasopressors, and RRT on the day of admission were included.

INTERVENTIONS

None.

MEASUREMENTS AND MAIN RESULTS

Performance was evaluated with area under the receiver operating characteristic curve (AUROC) and accuracy. RFCs performed better than XGBoost, SVM, and MLP models. For comparison, we evaluated LR models in the same cohorts. AUROC was very high ranging from 0.74 to 0.95 in predicting ventilation, vasopressors, and RRT use in our derivation and validation cohorts. ML used and variables such as Fio2, Glasgow Coma Scale, and mean arterial pressure to predict ventilator, vasopressor use, creatinine, and potassium to predict RRT use. LR was less accurate than ML, with AUROC ranging 0.66 to 0.8.

CONCLUSIONS

A ML algorithm more accurately predicts need of invasive ventilation, vasopressors, or RRT in hospitalized non-COVID-19 CAP and COVID-19 patients than regression models and could augment clinician judgment for triage and care of hospitalized CAP patients.

Assessing severe pneumonia risk in children via clinical prognostic model based on laboratory markers

Pneumonia represents a significant cause of mortality in children globally, emphasizing the importance of identifying high-risk patients to improve clinical outcomes. There is a lack of reliable laboratory markers and robust risk stratification models for clinical decision support in pediatric pneumonia. This study extracted data from the Paediatric Intensive Care database for 749 children under 3 years with severe pneumonia. The relationship between laboratory parameters and prognostic outcomes was evaluated using Cox proportional hazards regression analyses. Oxygen saturation, hemoglobin, lipase, urea, and uric acid were identified as laboratory parameters significantly associated with severe pneumonia outcomes. Leveraging these laboratory markers, a prognosis model was constructed employing the XGBoost classifier. The model was validated in a hold-out test cohort and an external validation cohort, with its performance assessed by the area under the receiver operating characteristic curve (AUC). The validation cohort was derived from 129 children with severe pneumonia admitted to the PICU of the Children's Hospital, Zhejiang University School of Medicine in 2019. The model demonstrated efficacy in predicting the death and survival of patients (AUC = 0.943), as well as in distinguishing between children at high- and low-risk of death in advance (HR = 2.930, 95 % CI: 2.551-3.366, P < 0.001). The robust performance of this model was further validated in the test cohort (AUC = 0.871), and the validation cohort (AUC = 0.872). In conclusion, this novel model enables the prediction of individualized mortality risk in children diagnosed with severe pneumonia, offering personalized risk assessments to inform and enhance clinical decision-making processes.

Machine learning-based model for predicting all-cause mortality in severe pneumonia

BACKGROUND

Severe pneumonia has a poor prognosis and high mortality. Current severity scores such as Acute Physiology and Chronic Health Evaluation (APACHE-II) and Sequential Organ Failure Assessment (SOFA), have limited ability to help clinicians in classification and management decisions. The goal of this study was to analyse the clinical characteristics of severe pneumonia and develop a machine learning-based mortality-prediction model for patients with severe pneumonia.

METHODS

Consecutive patients with severe pneumonia between 2013 and 2022 admitted to Beijing Chaoyang Hospital affiliated with Capital Medical University were included. In-hospital all-cause mortality was the outcome of this study. We performed a retrospective analysis of the cohort, stratifying patients into survival and non-survival groups, using mainstream machine learning algorithms (light gradient boosting machine, support vector classifier and random forest). We aimed to construct a mortality-prediction model for patients with severe pneumonia based on their accessible clinical and laboratory data. The discriminative ability was evaluated using the area under the receiver operating characteristic curve (AUC). The calibration curve was used to assess the fit goodness of the model, and decision curve analysis was performed to quantify clinical utility. By means of logistic regression, independent risk factors for death in severe pneumonia were figured out to provide an important basis for clinical decision-making.

RESULTS

A total of 875 patients were included in the development and validation cohorts, with the in-hospital mortality rate of 14.6%. The AUC of the model in the internal validation set was 0.8779 (95% CI, 0.738 to 0.974), showing a competitive discrimination ability that outperformed those of traditional clinical scoring systems, that is, APACHE-II, SOFA, CURB-65 (confusion, urea, respiratory rate, blood pressure, age ≥65 years), Pneumonia Severity Index. The calibration curve showed that the in-hospital mortality in severe pneumonia predicted by the model fit reasonably with the actual hospital mortality. In addition, the decision curve showed that the net clinical benefit was positive in both training and validation sets of hospitalised patients with severe pneumonia. Based on ensemble machine learning algorithms and logistic regression technique, the level of ferritin, lactic acid, blood urea nitrogen, creatine kinase, eosinophil and the requirement of vasopressors were identified as top independent predictors of in-hospital mortality with severe pneumonia.

CONCLUSION

A robust clinical model for predicting the risk of in-hospital mortality after severe pneumonia was successfully developed using machine learning techniques. The performance of this model demonstrates the effectiveness of these techniques in creating accurate predictive models, and the use of this model has the potential to greatly assist patients and clinical doctors in making well-informed decisions regarding patient care.

Machine Learning-Based Prediction of Postoperative Pneumonia Among Super-Aged Patients With Hip Fracture

Background

Hip fractures have become a significant health concern, particularly among super-aged patients, who were at a high risk of postoperative pneumonia due to their frailty and the presence of multiple comorbidities. This study aims to establish and validate a model to predict postoperative pneumonia among super-aged patients with hip fracture.

Methods

Data were derived from the Chinese PLA General Hospital (PLAGH) Hip Fracture Cohort Study, and we included 555 super-aged patients (≧80 years old) with hip fracture treated with surgery. Patient's demographics, comorbidities, laboratory tests, and surgery types were collected for analysis. All patients were randomly splitting into a training group and a validation group according to the ratio of 7:3. The majority of patients were used to train models, which was tuned using a series of algorithms, including decision tree (DT), random forest (RF), extreme gradient boosting machine (eXGBM), support vector machine (SVM), neural network (NN), and logistic regression (LR).

Results

The incidence of postoperative pneumonia was 7.2% (40/555). Among the six developed models, the eXGBM model demonstrated the optimal model, with the area under the curve (AUC) value of 0.929 (95% CI: 0.900-0.959), followed by the RF model (AUC: 0.916, 95% CI: 0.885-0.948). The LR model had an AUC value of 0.720 (95% CI: 0.662-0.778). In addition, the eXGBM model demonstrated the optimal prediction performance in terms of accuracy (0.858), precision (0.870), F1 score (0.855), Brier score (0.104), and log loss (0.349). It also showed favorable calibration ability and favorable clinical net benefits across various threshold risk.

Conclusion

This study develops and validates a reliable machine learning-based model to predict pneumonia specifically among super-aged patients with hip fracture following surgery. This model can serve as a useful tool to identify postoperative pneumonia and guide clinical strategies for super-aged patients with hip fracture.

Application of machine learning algorithms in an epidemiologic study of mortality

PURPOSE

Epidemiologic studies are important in assessing risk factors of mortality. Machine learning (ML) is efficient in analyzing multidimensional data to unravel dependencies between risk factors and health outcomes.

METHODS

Using a representative sample from the National Health and Nutrition Examination Survey data collected from 2009 to 2016 linked to the National Death Index public-use mortality data through December 31, 2019, we applied logistic, random forests, k-Nearest Neighbors, multivariate adaptive regression splines, support vector machines, extreme gradient boosting, and super learner ML algorithms to study risk factors of all-cause mortality. We evaluated the algorithms using area under the receiver operating curve (AUC-ROC), sensitivity, negative predictive value (NPV) among other metrics and interpreted the results using SHapley Additive exPlanation.

RESULTS

The AUC-ROC ranged from 0.80 ─ 0.87. The super learner had the highest AUC-ROC of 0.87 (95 % CI, 0.86 ─ 0.88), sensitivity of 0.86 (95 % CI, 0.84 ─ 0.88) and NPV of 0.98 (95 % CI, 0.98 ─ 0.99). Key risk factors of mortality included advanced age, larger waist circumference, male and systolic blood pressure. Being married, high annual household income, and high education level were linked with low risk of mortality.

CONCLUSIONS

Machine learning can be used to identify risk factors of mortality, which is critical for individualized targeted interventions in epidemiologic studies.

Prediction of mortality risk in patients with severe community-acquired pneumonia in the intensive care unit using machine learning

The aim of this study was to develop and validate a machine learning-based mortality risk prediction model for patients with severe community-acquired pneumonia (SCAP) in the intensive care unit (ICU). We collected data from two centers as the development and external validation cohorts. Variables were screened using the Recursive Feature Elimination method. Five machine learning algorithms were used to build predictive models. Models were evaluated through nested cross-validation to select the best one. The model was interpreted using Shapley Additive Explanations. We selected the optimal model to generate the web calculator. A total of 23 predictive features were selected. The Light Gradient Boosting Machine (LightGBM) model had an area under the receiver operating characteristic curve (AUC) of 0.842 (95% CI: 0.757-0.927), with an external 5-fold cross-validation average AUC of 0.842 ± 0.038, which was superior to the other models. External validation results also demonstrated good performance by the LightGBM model with an AUC of 0.856 (95% CI: 0.792-0.921). Based on this, we generated a web calculator by combining five high importance predictive factors. The LightGBM model was confirmed to be efficient and stable in predicting the mortality risk of patients with SCAP admitted to the ICU. The web calculator based on the LightGBM model can provide clinicians with a prognostic evaluation tool.

Employing a low-code machine learning approach to predict in-hospital mortality and length of stay in patients with community-acquired pneumonia

Community-acquired pneumonia (CAP) is associated with high mortality rates and often results in prolonged hospital stays. The potential of machine learning to enhance prediction accuracy in this context is significant, yet clinicians often lack the programming skills required for effective data mining. This study aimed to assess the effectiveness of a low-code approach for assisting clinicians with data mining for mortality and length of stay (LOS) prediction in patients with CAP. A retrospective study was conducted using a low-code platform and the PyCaret library in Google Colab on data from patients with community-acquired pneumonia (CAP) admitted between January 2013 and December 2021 to two medical facilities. Model performance was assessed using the area under the receiver operating characteristic curve (AUC) for mortality prediction and the R2 score for LOS prediction, with benchmarks set at AUC > 0.9 and R2 > 0.5. The Shapley Additive Explanations (SHAP) method was used for interpreting individual predictions. A total of 669 CAP patients were enrolled in the analysis.Fifteen models were evaluated for mortality prediction, and nineteen models were evaluated for LOS prediction utilizing the PyCaret library. The Light Gradient Boosting Machine model yielded the highest AUC (0.963) for mortality prediction. In predicting LOS, the Extratrees Regressor model achieved the highest R2 score of 0.585. Factors such as the severity of pneumonia and the Charlson Comorbidity Index (CCI) were significant factors influencing mortality. For the LOS, the CCI score, activities of daily living, and social support were significant predictors. The low-code approach enables medical professionals with limited technical expertise to effectively employ data science in their clinical decision-making process. This approach proved to be a valuable tool in the analysis of CAP patient data.

Robust extraction of pneumonia-associated clinical states from electronic health records

Mining of electronic health records (EHR) promises to automate the identification of comprehensive disease phenotypes. However, the realization of this promise is hindered by the unavailability of generalizable ground-truth information, data incompleteness and heterogeneity, and the lack of generalization to multiple cohorts. We present here a data-driven approach to identify clinical states that we implement for 585 critical care patients with suspected pneumonia recruited by the SCRIPT study, which we compare to and integrate with 9,918 pneumonia patients from the MIMIC-IV dataset. We extract and curate from their structured EHRs a primary set of clinical features (53 and 59 features for SCRIPT and MIMIC-IV, respectively), including disease severity scores, vital signs, and so on, at various degrees of completeness. We aggregate irregular time series into daily frequency, resulting in 12,495 and 94,684 patient-day pairs for SCRIPT and MIMIC, respectively. We define a "common-sense" ground truth that we then use in a semisupervised pipeline to optimize choices for data preprocessing, and reduce the feature space to four principal components. We describe and validate an ensemble-based clustering method that enables us to robustly identify five clinical states, and use a Gaussian mixture model to quantify uncertainty in cluster assignment. Demonstrating the clinical relevance of the identified states, we find that three states are strongly associated with disease outcomes (dying vs. recovering), while the other two reflect disease etiology. The outcome associated clinical states provide significantly increased discrimination of mortality rates over standard approaches.

Artificial intelligence and prescription of antibiotic therapy: present and future

INTRODUCTION

In the past few years, the use of artificial intelligence in healthcare has grown exponentially. Prescription of antibiotics is not exempt from its rapid diffusion, and various machine learning (ML) techniques, from logistic regression to deep neural networks and large language models, have been explored in the literature to support decisions regarding antibiotic prescription.

AREAS COVERED

In this narrative review, we discuss promises and challenges of the application of ML-based clinical decision support systems (ML-CDSSs) for antibiotic prescription. A search was conducted in PubMed up to April 2024.

EXPERT OPINION

Prescribing antibiotics is a complex process involving various dynamic phases. In each of these phases, the support of ML-CDSSs has shown the potential, and also the actual ability in some studies, to favorably impacting relevant clinical outcomes. Nonetheless, before widely exploiting this massive potential, there are still crucial challenges ahead that are being intensively investigated, pertaining to the transparency of training data, the definition of the sufficient degree of prediction explanations when predictions are obtained through black box models, and the legal and ethical framework for decision responsibility whenever an antibiotic prescription is supported by ML-CDSSs.

Interpretable mortality prediction model for ICU patients with pneumonia: using shapley additive explanation method

BACKGROUND

Pneumonia, a leading cause of morbidity and mortality worldwide, often necessitates Intensive Care Unit (ICU) admission. Accurate prediction of pneumonia mortality is crucial for tailored prevention and treatment plans. However, existing mortality prediction models face limited adoption in clinical practice due to their lack of interpretability.

OBJECTIVE

This study aimed to develop an interpretable model for predicting pneumonia mortality in ICUs. Leveraging the Shapley Additive Explanation (SHAP) method, we sought to elucidate the Extreme Gradient Boosting (XGBoost) model and identify prognostic factors for pneumonia.

METHODS

Conducted as a retrospective cohort study, we utilized electronic health records from the eICU-CRD (2014-2015) for all adult pneumonia patients. The first 24 h of each ICU admission records were considered, with 70% of the dataset allocated for model training and 30% for validation. The XGBoost model was employed, and performance was assessed using the area under the receiver operating characteristic curve (AUC). The SHAP method provided insights into the XGBoost model.

RESULTS

Among 10,962 pneumonia patients, in-hospital mortality was 16.33%. The XGBoost model demonstrated superior predictive performance (AUC: 0.778 ± 0.016)) compared to traditional scoring systems and other machine learning method, which achieved an improvement of 10% points. SHAP analysis identified Aspartate Aminotransferase (AST) as the most crucial predictor.

CONCLUSIONS

Interpretable predictive models enhance mortality risk assessment for pneumonia patients in the ICU, fostering transparency. AST emerged as the foremost predictor, followed by patient age, albumin, BMI et al. These insights, rooted in strong correlations with mortality, facilitate improved clinical decision-making and resource allocation.

Machine learning-based derivation and validation of three immune phenotypes for risk stratification and prognosis in community-acquired pneumonia: a retrospective cohort study

Background

The clinical presentation of Community-acquired pneumonia (CAP) in hospitalized patients exhibits heterogeneity. Inflammation and immune responses play significant roles in CAP development. However, research on immunophenotypes in CAP patients is limited, with few machine learning (ML) models analyzing immune indicators.

Methods

A retrospective cohort study was conducted at Xinhua Hospital, affiliated with Shanghai Jiaotong University. Patients meeting predefined criteria were included and unsupervised clustering was used to identify phenotypes. Patients with distinct phenotypes were also compared in different outcomes. By machine learning methods, we comprehensively assess the disease severity of CAP patients.

Results

A total of 1156 CAP patients were included in this research. In the training cohort (n=809), we identified three immune phenotypes among patients: Phenotype A (42.0%), Phenotype B (40.2%), and Phenotype C (17.8%), with Phenotype C corresponding to more severe disease. Similar results can be observed in the validation cohort. The optimal prognostic model, SuperPC, achieved the highest average C-index of 0.859. For predicting CAP severity, the random forest model was highly accurate, with C-index of 0.998 and 0.794 in training and validation cohorts, respectively.

Conclusion

CAP patients can be categorized into three distinct immune phenotypes, each with prognostic relevance. Machine learning exhibits potential in predicting mortality and disease severity in CAP patients by leveraging clinical immunological data. Further external validation studies are crucial to confirm applicability.

Harnessing of Artificial Intelligence for the Diagnosis and Prevention of Hospital-Acquired Infections: A Systematic Review

Healthcare-associated infections (HAIs) are the most common adverse events in healthcare and constitute a major global public health concern. Surveillance represents the foundation for the effective prevention and control of HAIs, yet conventional surveillance is costly and labor intensive. Artificial intelligence (AI) and machine learning (ML) have the potential to support the development of HAI surveillance algorithms for the understanding of HAI risk factors, the improvement of patient risk stratification as well as the prediction and timely detection and prevention of infections. AI-supported systems have so far been explored for clinical laboratory testing and imaging diagnosis, antimicrobial resistance profiling, antibiotic discovery and prediction-based clinical decision support tools in terms of HAIs. This review aims to provide a comprehensive summary of the current literature on AI applications in the field of HAIs and discuss the future potentials of this emerging technology in infection practice. Following the PRISMA guidelines, this study examined the articles in databases including PubMed and Scopus until November 2023, which were screened based on the inclusion and exclusion criteria, resulting in 162 included articles. By elucidating the advancements in the field, we aim to highlight the potential applications of AI in the field, report related issues and shortcomings and discuss the future directions.

Development and validation of a predictive model for 30-day mortality in patients with severe community-acquired pneumonia in intensive care units

Background

Based on the high prevalence and fatality rates associated with severe community-acquired pneumonia (SCAP), this study endeavored to construct an innovative nomogram for early identification of individuals at high risk of all-cause death within a 30-day period among SCAP patients receiving intensive care units (ICU) treatment.

Methods

In this single-center, retrospective study, 718 SCAP patients were screened from the Medical Information Mart for Intensive Care IV (MIMIC-IV) database for the development of a predictive model. A total of 97 patients eligible for inclusion were included from Chongqing General Hospital, China between January 2020 and July 2023 for external validation. Clinical data and short-term prognosis were collected. Risk factors were determined using the least absolute shrinkage and selection operator (LASSO) and multiple logistic regression analysis. The model's performance was evaluated through area under the curve (AUC), calibration curve, and decision curve analysis (DCA).

Results

Eight risk predictors, including age, presence of malignant cancer, heart rate, mean arterial pressure, albumin, blood urea nitrogen, prothrombin time, and lactate levels were adopted in a nomogram. The nomogram exhibited high predictive accuracy, with an AUC of 0.803 (95% CI: 0.756-0.845) in the training set, 0.756 (95% CI: 0.693-0.816) in the internal validation set, 0.778 (95% CI: 0.594-0.893) in the external validation set concerning 30-day mortality. Meanwhile, the nomogram demonstrated effective calibration through well-fitted calibration curves. DCA confirmed the clinical application value of the nomogram.

Conclusion

This simple and reliable nomogram can help physicians assess the short-term prognosis of patients with SCAP quickly and effectively, and could potentially be adopted widely in clinical settings after more external validations.