Artificial intelligence-derived risk score for mortality in secondary mitral regurgitation treated by transcatheter edge-to-edge repair: the EuroSMR risk score.

BACKGROUND AND AIMS: Risk stratification for mitral valve transcatheter edge-to-edge repair (M-TEER) is paramount in the decision-making process to appropriately select patients with severe secondary mitral regurgitation (SMR). This study sought to develop and validate an artificial intelligence-derived risk score (EuroSMR score) to predict 1-year outcomes (survival or survival + clinical improvement) in patients with SMR undergoing M-TEER. METHODS: An artificial intelligence-derived risk score was developed from the EuroSMR cohort (4172 and 428 patients treated with M-TEER in the derivation and validation cohorts, respectively). The EuroSMR score was validated and compared with established risk models. RESULTS: The EuroSMR risk score, which is based on 18 clinical, echocardiographic, laboratory, and medication parameters, allowed for an improved discrimination of surviving and non-surviving patients (hazard ratio 4.3, 95% confidence interval 3.7-5.0; P < .001), and outperformed established risk scores in the validation cohort. Prediction for 1-year mortality (area under the curve: 0.789, 95% confidence interval 0.737-0.842) ranged from <5% to >70%, including the identification of an extreme-risk population (2.6% of the entire cohort), which had a very high probability for not surviving beyond 1 year (hazard ratio 6.5, 95% confidence interval 3.0-14; P < .001). The top 5% of patients with the highest EuroSMR risk scores showed event rates of 72.7% for mortality and 83.2% for mortality or lack of clinical improvement at 1-year follow-up. CONCLUSIONS: The EuroSMR risk score may allow for improved prognostication in heart failure patients with severe SMR, who are considered for a M-TEER procedure. The score is expected to facilitate the shared decision-making process with heart team members and patients.

Development of a comorbidity index for patients undergoing abdominal aortic aneurysm repair.

BACKGROUND: Endovascular aneurysm repair (EVAR) and open surgical repair (OSR) are two modalities to treat patients with abdominal aortic aneurysm (AAA). Alternative to individual comorbidity adjustment, a summary comorbidity index is a weighted composite score of all comorbidities that can be used as standard metric to control for comorbidity burden in clinical studies. This study aimed to develop summary comorbidity indices for patients who underwent AAA repair. METHODS: Patients who went under EVAR or OSR were identified in National Inpatient Sample (NIS) between the last quarter of 2015 to 2020. In each group, patients were randomly sampled into experimental (2/3) and validation (1/3) groups. The weights of Elixhauser comorbidities were determined from a multivariable logistic regression and single comorbidity indices were developed for EVAR and OAR groups, respectively. RESULTS: There were 34,668 patients underwent EVAR (2.19% mortality) and 4792 underwent OSR (10.98% mortality). Both comorbidity indices had moderate discriminative power (EVAR c-statistic, 0.641; 95% confidence interval [CI], 0.616-0.665; OSR c-statistic, 0.600; 95% CI, 0.563-0.630) and good calibration (EVAR Brier score, 0.021; OSR Brier score, 0.096). The indices had significantly better discriminative power (DeLong P <.001) than the Elixhauser Comorbidity Index (ECI) (EVAR c-statistic, 0.572; 95% CI, 0.546-0.597; OSR c-statistic, 0.502; 95% CI, 0.472-0.533). For internal validation, both indices had similar performance compared with individual comorbidity adjustment (EVAR DeLong P = .650; OSR DeLong P = .431). These indices demonstrated good external validation, exhibiting comparable performance to their respective validation groups (EVAR DeLong P = .891; OSR DeLong P = .757). CONCLUSIONS: ECI, the comorbidity index formulated for the general population, exhibited suboptimal performance in patients who underwent AAA repair. In response, we developed summary comorbidity indices for both EVAR and OSR for AAA repair, which were internally and externally validated. The EVAR and OSR comorbidity indices outperformed the ECI in discriminating in-hospital mortality rates. They can standardize comorbidity measurement for clinical studies in AAA repair, especially for studies with small samples such as single-institute data sources to facilitate replication and comparison of results across studies.

Validation of the Geriatrics at Risk Score (GeRi-Score) on 120-day follow-up, the influence of preoperative geriatric visits, and the time to surgery on the outcome of hip fracture patients: an analysis from the Registry for Geriatric Trauma (ATR-DGU).

A validation of the GeRi-Score on 120-day mortality, the impact of a pre-operative visit by a geriatrician, and timing of surgery on the outcome was conducted. The score has predictive value for 120-day mortality. No advantage was found for surgery within 24 h or a preoperative geriatric visit. PURPOSE: Numerous tools predict mortality among patients with hip fractures, but they include many variables, require time-consuming assessment, and are difficult to calculate. The GeRi-Score provides a quick method of pre-operative assessment. The aim of this study is to validate the score in the 120-day follow-up and determine the impact of a pre-operative visit by a geriatrician and timing of surgery on the patient outcome. METHODS: A retrospective analysis of the AltersTraumaRegister DGU® from 2017 to 2021 was conducted, including all proximal femur fractures. The patients were divided into low-, moderate-, and high-risk groups based on the GeRi-Score. Mortality was analyzed using logistic regression. To determine the influence of the time to surgery and the preoperative visit by a geriatrician, matching was performed using the exact GeRi-Score, preoperative walking ability, type of fracture, and the time to surgery. RESULTS: The study included 38,570 patients, divided into 12,673 low-risk, 18,338 moderate-risk, and 7,559 high-risk patients. The moderate-risk group had three times the mortality risk of the low-risk group (OR 3.19 (95% CI 2.68-3.79; p<0.001)), while the high-risk group had almost eight times the mortality risk than the low-risk group (OR 7.82 (95% CI 6.51-9.93; p<0.001)). No advantage was found for surgery within the first 24 h across all groups. There was a correlation of a preoperative geriatric visit and mortality showing an increase in the moderate and high-risk group on in-house mortality. CONCLUSIONS: The GeRi-Score has predictive value for 120-day mortality. No advantage was found for surgery within 24 h. The analysis did not demonstrate a benefit of the preoperative geriatric visit, but more data are needed.

Improved pediatric ICU mortality prediction for respiratory diseases: machine learning and data subdivision insights.

The growing concern of pediatric mortality demands heightened preparedness in clinical settings, especially within intensive care units (ICUs). As respiratory-related admissions account for a substantial portion of pediatric illnesses, there is a pressing need to predict ICU mortality in these cases. This study based on data from 1188 patients, addresses this imperative using machine learning techniques and investigating different class balancing methods for pediatric ICU mortality prediction. This study employs the publicly accessible "Paediatric Intensive Care database" to train, validate, and test a machine learning model for predicting pediatric patient mortality. Features were ranked using three machine learning feature selection techniques, namely Random Forest, Extra Trees, and XGBoost, resulting in the selection of 16 critical features from a total of 105 features. Ten machine learning models and ensemble techniques are used to make accurate mortality predictions. To tackle the inherent class imbalance in the dataset, we applied a unique data partitioning technique to enhance the model's alignment with the data distribution. The CatBoost machine learning model achieved an area under the curve (AUC) of 72.22%, while the stacking ensemble model yielded an AUC of 60.59% for mortality prediction. The proposed subdivision technique, on the other hand, provides a significant improvement in performance metrics, with an AUC of 85.2% and an accuracy of 89.32%. These findings emphasize the potential of machine learning in enhancing pediatric mortality prediction and inform strategies for improved ICU readiness.

External Validation of Predictors of Mortality in Polytrauma Patients.

INTRODUCTION: The Parkland Trauma Index of Mortality (PTIM) is an integrated, machine learning 72-h mortality prediction model that automatically extracts and analyzes demographic, laboratory, and physiological data in polytrauma patients. We hypothesized that this validated model would perform equally as well at another level 1 trauma center. METHODS: A retrospective cohort study was performed including ∼5000 adult level 1 trauma activation patients from January 2022 to September 2023. Demographics, physiologic and laboratory values were collected. First, a test set of models using PTIM clinical variables (CVs) was used as external validation, named PTIM+. Then, multiple novel mortality prediction models were developed considering all CVs designated as the Cincinnati Trauma Index of Mortality (CTIM). The statistical performance of the models was then compared. RESULTS: PTIM CVs were found to have similar predictive performance within the PTIM + external validation model. The highest correlating CVs used in CTIM overlapped considerably with those of the PTIM, and performance was comparable between models. Specifically, for prediction of mortality within 48 h (CTIM versus PTIM): positive prediction value was 35.6% versus 32.5%, negative prediction value was 99.6% versus 99.3%, sensitivity was 81.0% versus 82.5%, specificity was 97.3% versus 93.6%, and area under the curve was 0.98 versus 0.94. CONCLUSIONS: This external cohort study suggests that the variables initially identified via PTIM retain their predictive ability and are accessible in a different level 1 trauma center. This work shows that a trauma center may be able to operationalize an effective predictive model without undertaking a repeated time and resource intensive process of full variable selection.

The impact of norepinephrine dose reporting heterogeneity on mortality prediction in septic shock patients.

BACKGROUND: Norepinephrine (NE) is a cornerstone drug in the management of septic shock, with its dose being used clinically as a marker of disease severity and as mortality predictor. However, variations in NE dose reporting either as salt formulations or base molecule may lead to misinterpretation of mortality risks and hinder the process of care. METHODS: We conducted a retrospective analysis of the MIMIC-IV database to assess the impact of NE dose reporting heterogeneity on mortality prediction in a cohort of septic shock patients. NE doses were converted from the base molecule to equivalent salt doses, and their ability to predict 28-day mortality at common severity dose cut-offs was compared. RESULTS: 4086 eligible patients with septic shock were identified, with a median age of 68 [57-78] years, an admission SOFA score of 7 [6-10], and lactate at diagnosis of 3.2 [2.4-5.1] mmol/L. Median peak NE dose at day 1 was 0.24 [0.12-0.42] µg/kg/min, with a 28-day mortality of 39.3%. The NE dose showed significant heterogeneity in mortality prediction depending on which formulation was reported, with doses reported as bitartrate and tartrate presenting 65 (95% CI 79-43)% and 67 (95% CI 80-47)% lower ORs than base molecule, respectively. This divergence in prediction widened at increasing NE doses. When using a 1 µg/kg/min threshold, predicted mortality was 54 (95% CI 52-56)% and 83 (95% CI 80-87)% for tartrate formulation and base molecule, respectively. CONCLUSIONS: Heterogeneous reporting of NE doses significantly affects mortality prediction in septic shock. Standardizing NE dose reporting as base molecule could enhance risk stratification and improve processes of care. These findings underscore the importance of consistent NE dose reporting practices in critical care settings.

Assessing fairness in machine learning models: A study of racial bias using matched counterparts in mortality prediction for patients with chronic diseases.

OBJECTIVE: Existing approaches to fairness evaluation often overlook systematic differences in the social determinants of health, like demographics and socioeconomics, among comparison groups, potentially leading to inaccurate or even contradictory conclusions. This study aims to evaluate racial disparities in predicting mortality among patients with chronic diseases using a fairness detection method that considers systematic differences. METHODS: We created five datasets from Mass General Brigham's electronic health records (EHR), each focusing on a different chronic condition: congestive heart failure (CHF), chronic kidney disease (CKD), chronic obstructive pulmonary disease (COPD), chronic liver disease (CLD), and dementia. For each dataset, we developed separate machine learning models to predict 1-year mortality and examined racial disparities by comparing prediction performances between Black and White individuals. We compared racial fairness evaluation between the overall Black and White individuals versus their counterparts who were Black and matched White individuals identified by propensity score matching, where the systematic differences were mitigated. RESULTS: We identified significant differences between Black and White individuals in age, gender, marital status, education level, smoking status, health insurance type, body mass index, and Charlson comorbidity index (p-value < 0.001). When examining matched Black and White subpopulations identified through propensity score matching, significant differences between particular covariates existed. We observed weaker significance levels in the CHF cohort for insurance type (p = 0.043), in the CKD cohort for insurance type (p = 0.005) and education level (p = 0.016), and in the dementia cohort for body mass index (p = 0.041); with no significant differences for other covariates. When examining mortality prediction models across the five study cohorts, we conducted a comparison of fairness evaluations before and after mitigating systematic differences. We revealed significant differences in the CHF cohort with p-values of 0.021 and 0.001 in terms of F1 measure and Sensitivity for the AdaBoost model, and p-values of 0.014 and 0.003 in terms of F1 measure and Sensitivity for the MLP model, respectively. DISCUSSION AND CONCLUSION: This study contributes to research on fairness assessment by focusing on the examination of systematic disparities and underscores the potential for revealing racial bias in machine learning models used in clinical settings.

Predicting Mortality in Sepsis-Associated Acute Respiratory Distress Syndrome: A Machine Learning Approach Using the MIMIC-III Database.

BACKGROUND: To develop and validate a mortality prediction model for patients with sepsis-associated Acute Respiratory Distress Syndrome (ARDS). METHODS: This retrospective cohort study included 2466 patients diagnosed with sepsis and ARDS within 24 h of ICU admission. Demographic, clinical, and laboratory parameters were extracted from Medical Information Mart for Intensive Care III (MIMIC-III) database. Feature selection was performed using the Boruta algorithm, followed by the construction of seven ML models: logistic regression, Naive Bayes, k-nearest neighbor, support vector machine, decision tree, Random Forest, and extreme gradient boosting. Model performance was evaluated using the area under the receiver operating characteristic curve, accuracy, sensitivity, specificity, positive predictive value, and negative predictive value. RESULTS: The study identified 24 variables significantly associated with mortality. The optimal ML model, a Random Forest model, demonstrated an AUC of 0.8015 in the test set, with high accuracy and specificity. The model highlighted the importance of blood urea nitrogen, age, urine output, Simplified Acute Physiology Score II, and albumin levels in predicting mortality. CONCLUSIONS: The model's superior predictive performance underscores the potential for integrating advanced analytics into clinical decision-making processes, potentially improving patient outcomes and resource allocation in critical care settings.

Predicting mortality risk in hospitalized COVID-19 patients: an early model utilizing clinical symptoms.

BACKGROUND: Despite global efforts to control the COVID-19 pandemic, the emergence of new viral strains continues to pose a significant threat. Accurate patient stratification, optimized resource allocation, and appropriate treatment are crucial in managing COVID-19 cases. To address this, a simple and accurate prognostic tool capable of rapidly identifying individuals at high risk of mortality is urgently needed. Early prognosis facilitates predicting treatment outcomes and enables effective patient management. The aim of this study was to develop an early predictive model for assessing mortality risk in hospitalized COVID-19 patients, utilizing baseline clinical factors. METHODS: We conducted a descriptive cross-sectional study involving a cohort of 375 COVID-19 patients admitted and treated at the COVID-19 Patient Treatment Center in Military Hospital 175 from October 2021 to December 2022. RESULTS: Among the 375 patients, 246 and 129 patients were categorized into the survival and mortality groups, respectively. Our findings revealed six clinical factors that demonstrated independent predictive value for mortality in COVID-19 patients. These factors included age greater than 50 years, presence of multiple underlying diseases, dyspnea, acute confusion, saturation of peripheral oxygen below 94%, and oxygen demand exceeding 5 L per minute. We integrated these factors to develop the Military Hospital 175 scale (MH175), a prognostic scale demonstrating significant discriminatory ability with an area under the curve (AUC) of 0.87. The optimal cutoff value for predicting mortality risk using the MH175 score was determined to be ≥ 3 points, resulting in a sensitivity of 96.1%, specificity of 63.4%, positive predictive value of 58%, and negative predictive value of 96.9%. CONCLUSIONS: The MH175 scale demonstrated a robust predictive capacity for assessing mortality risk in patients with COVID-19. Implementation of the MH175 scale in clinical settings can aid in patient stratification and facilitate the application of appropriate treatment strategies, ultimately reducing the risk of death. Therefore, the utilization of the MH175 scale holds significant potential to improve clinical outcomes in COVID-19 patients. TRIAL REGISTRATION: An independent ethics committee approved the study (Research Ethics Committee of Military Hospital 175 (No. 3598GCN-HDDD; date: October 8, 2021), which was performed in accordance with the Declaration of Helsinki, Guidelines for Good Clinical Practice.

Development and Validation of an Interpretable Conformal Predictor to Predict Sepsis Mortality Risk: Retrospective Cohort Study.

BACKGROUND: Early and reliable identification of patients with sepsis who are at high risk of mortality is important to improve clinical outcomes. However, 3 major barriers to artificial intelligence (AI) models, including the lack of interpretability, the difficulty in generalizability, and the risk of automation bias, hinder the widespread adoption of AI models for use in clinical practice. OBJECTIVE: This study aimed to develop and validate (internally and externally) a conformal predictor of sepsis mortality risk in patients who are critically ill, leveraging AI-assisted prediction modeling. The proposed approach enables explaining the model output and assessing its confidence level. METHODS: We retrospectively extracted data on adult patients with sepsis from a database collected in a teaching hospital at Beth Israel Deaconess Medical Center for model training and internal validation. A large multicenter critical care database from the Philips eICU Research Institute was used for external validation. A total of 103 clinical features were extracted from the first day after admission. We developed an AI model using gradient-boosting machines to predict the mortality risk of sepsis and used Mondrian conformal prediction to estimate the prediction uncertainty. The Shapley additive explanation method was used to explain the model. RESULTS: A total of 16,746 (80%) patients from Beth Israel Deaconess Medical Center were used to train the model. When tested on the internal validation population of 4187 (20%) patients, the model achieved an area under the receiver operating characteristic curve of 0.858 (95% CI 0.845-0.871), which was reduced to 0.800 (95% CI 0.789-0.811) when externally validated on 10,362 patients from the Philips eICU database. At a specified confidence level of 90% for the internal validation cohort the percentage of error predictions (n=438) out of all predictions (n=4187) was 10.5%, with 1229 (29.4%) predictions requiring clinician review. In contrast, the AI model without conformal prediction made 1449 (34.6%) errors. When externally validated, more predictions (n=4004, 38.6%) were flagged for clinician review due to interdatabase heterogeneity. Nevertheless, the model still produced significantly lower error rates compared to the point predictions by AI (n=1221, 11.8% vs n=4540, 43.8%). The most important predictors identified in this predictive model were Acute Physiology Score III, age, urine output, vasopressors, and pulmonary infection. Clinically relevant risk factors contributing to a single patient were also examined to show how the risk arose. CONCLUSIONS: By combining model explanation and conformal prediction, AI-based systems can be better translated into medical practice for clinical decision-making.

Improving the Prognostic Evaluation Precision of Hospital Outcomes for Heart Failure Using Admission Notes and Clinical Tabular Data: Multimodal Deep Learning Model.

BACKGROUND: Clinical notes contain contextualized information beyond structured data related to patients' past and current health status. OBJECTIVE: This study aimed to design a multimodal deep learning approach to improve the evaluation precision of hospital outcomes for heart failure (HF) using admission clinical notes and easily collected tabular data. METHODS: Data for the development and validation of the multimodal model were retrospectively derived from 3 open-access US databases, including the Medical Information Mart for Intensive Care III v1.4 (MIMIC-III) and MIMIC-IV v1.0, collected from a teaching hospital from 2001 to 2019, and the eICU Collaborative Research Database v1.2, collected from 208 hospitals from 2014 to 2015. The study cohorts consisted of all patients with critical HF. The clinical notes, including chief complaint, history of present illness, physical examination, medical history, and admission medication, as well as clinical variables recorded in electronic health records, were analyzed. We developed a deep learning mortality prediction model for in-hospital patients, which underwent complete internal, prospective, and external evaluation. The Integrated Gradients and SHapley Additive exPlanations (SHAP) methods were used to analyze the importance of risk factors. RESULTS: The study included 9989 (16.4%) patients in the development set, 2497 (14.1%) patients in the internal validation set, 1896 (18.3%) in the prospective validation set, and 7432 (15%) patients in the external validation set. The area under the receiver operating characteristic curve of the models was 0.838 (95% CI 0.827-0.851), 0.849 (95% CI 0.841-0.856), and 0.767 (95% CI 0.762-0.772), for the internal, prospective, and external validation sets, respectively. The area under the receiver operating characteristic curve of the multimodal model outperformed that of the unimodal models in all test sets, and tabular data contributed to higher discrimination. The medical history and physical examination were more useful than other factors in early assessments. CONCLUSIONS: The multimodal deep learning model for combining admission notes and clinical tabular data showed promising efficacy as a potentially novel method in evaluating the risk of mortality in patients with HF, providing more accurate and timely decision support.

Interpretable prediction of 30-day mortality in patients with acute pancreatitis based on machine learning and SHAP.

BACKGROUND: Severe acute pancreatitis (SAP) can be fatal if left unrecognized and untreated. The purpose was to develop a machine learning (ML) model for predicting the 30-day all-cause mortality risk in SAP patients and to explain the most important predictors. METHODS: This research utilized six ML methods, including logistic regression (LR), k-nearest neighbors(KNN), support vector machines (SVM), naive Bayes (NB), random forests(RF), and extreme gradient boosting(XGBoost), to construct six predictive models for SAP. An extensive evaluation was conducted to determine the most effective model and then the Shapley Additive exPlanations (SHAP) method was applied to visualize key variables. Utilizing the optimized model, stratified predictions were made for patients with SAP. Further, the study employed multivariable Cox regression analysis and Kaplan-Meier survival curves, along with subgroup analysis, to explore the relationship between the machine learning-based score and 30-day mortality. RESULTS: Through LASSO regression and recursive feature elimination (RFE), 25 optimal feature variables are selected. The XGBoost model performed best, with an area under the curve (AUC) of 0.881, a sensitivity of 0.5714, a specificity of 0.9651 and an F1 score of 0.64. The first six most important feature variables were the use of vasopressor, high Charlson comorbidity index, low blood oxygen saturation, history of malignant tumor, hyperglycemia and high APSIII score. Based on the optimal threshold of 0.62, patients were divided into high and low-risk groups, and the 30-day survival rate in the high-risk group decreased significantly. COX regression analysis further confirmed the positive correlation between high-risk scores and 30-day mortality. In the subgroup analysis, the model showed good risk stratification ability in patients with different gender, renal replacement therapy and with or without a history of malignant tumor, but it was not effective in predicting peripheral vascular disease. CONCLUSIONS: the XGBoost model effectively predicts the severity of SAP, serving as a valuable tool for clinicians to identify SAP early.

Machine learning-based prognostic model for 30-day mortality prediction in Sepsis-3.

BACKGROUND: Sepsis poses a critical threat to hospitalized patients, particularly those in the Intensive Care Unit (ICU). Rapid identification of Sepsis is crucial for improving survival rates. Machine learning techniques offer advantages over traditional methods for predicting outcomes. This study aimed to develop a prognostic model using a Stacking-based Meta-Classifier to predict 30-day mortality risks in Sepsis-3 patients from the MIMIC-III database. METHODS: A cohort of 4,240 Sepsis-3 patients was analyzed, with 783 experiencing 30-day mortality and 3,457 surviving. Fifteen biomarkers were selected using feature ranking methods, including Extreme Gradient Boosting (XGBoost), Random Forest, and Extra Tree, and the Logistic Regression (LR) model was used to assess their individual predictability with a fivefold cross-validation approach for the validation of the prediction. The dataset was balanced using the SMOTE-TOMEK LINK technique, and a stacking-based meta-classifier was used for 30-day mortality prediction. The SHapley Additive explanations analysis was performed to explain the model's prediction. RESULTS: Using the LR classifier, the model achieved an area under the curve or AUC score of 0.99. A nomogram provided clinical insights into the biomarkers' significance. The stacked meta-learner, LR classifier exhibited the best performance with 95.52% accuracy, 95.79% precision, 95.52% recall, 93.65% specificity, and a 95.60% F1-score. CONCLUSIONS: In conjunction with the nomogram, the proposed stacking classifier model effectively predicted 30-day mortality in Sepsis patients. This approach holds promise for early intervention and improved outcomes in treating Sepsis cases.

Prediction of 30-day mortality for ICU patients with Sepsis-3.

BACKGROUND: There is a growing demand for advanced methods to improve the understanding and prediction of illnesses. This study focuses on Sepsis, a critical response to infection, aiming to enhance early detection and mortality prediction for Sepsis-3 patients to improve hospital resource allocation. METHODS: In this study, we developed a Machine Learning (ML) framework to predict the 30-day mortality rate of ICU patients with Sepsis-3 using the MIMIC-III database. Advanced big data extraction tools like Snowflake were used to identify eligible patients. Decision tree models and Entropy Analyses helped refine feature selection, resulting in 30 relevant features curated with clinical experts. We employed the Light Gradient Boosting Machine (LightGBM) model for its efficiency and predictive power. RESULTS: The study comprised a cohort of 9118 Sepsis-3 patients. Our preprocessing techniques significantly improved both the AUC and accuracy metrics. The LightGBM model achieved an impressive AUC of 0.983 (95% CI: [0.980-0.990]), an accuracy of 0.966, and an F1-score of 0.910. Notably, LightGBM showed a substantial 6% improvement over our best baseline model and a 14% enhancement over the best existing literature. These advancements are attributed to (I) the inclusion of the novel and pivotal feature Hospital Length of Stay (HOSP_LOS), absent in previous studies, and (II) LightGBM's gradient boosting architecture, enabling robust predictions with high-dimensional data while maintaining computational efficiency, as demonstrated by its learning curve. CONCLUSIONS: Our preprocessing methodology reduced the number of relevant features and identified a crucial feature overlooked in previous studies. The proposed model demonstrated high predictive power and generalization capability, highlighting the potential of ML in ICU settings. This model can streamline ICU resource allocation and provide tailored interventions for Sepsis-3 patients.

Early Prediction of Mortality for Septic Patients Visiting Emergency Room Based on Explainable Machine Learning: A Real-World Multicenter Study.

BACKGROUND: Worldwide, sepsis is the leading cause of death in hospitals. If mortality rates in patients with sepsis can be predicted early, medical resources can be allocated efficiently. We constructed machine learning (ML) models to predict the mortality of patients with sepsis in a hospital emergency department. METHODS: This study prospectively collected nationwide data from an ongoing multicenter cohort of patients with sepsis identified in the emergency department. Patients were enrolled from 19 hospitals between September 2019 and December 2020. For acquired data from 3,657 survivors and 1,455 deaths, six ML models (logistic regression, support vector machine, random forest, extreme gradient boosting [XGBoost], light gradient boosting machine, and categorical boosting [CatBoost]) were constructed using fivefold cross-validation to predict mortality. Through these models, 44 clinical variables measured on the day of admission were compared with six sequential organ failure assessment (SOFA) components (PaO2/FIO2 [PF], platelets (PLT), bilirubin, cardiovascular, Glasgow Coma Scale score, and creatinine). The confidence interval (CI) was obtained by performing 10,000 repeated measurements via random sampling of the test dataset. All results were explained and interpreted using Shapley's additive explanations (SHAP). RESULTS: Of the 5,112 participants, CatBoost exhibited the highest area under the curve (AUC) of 0.800 (95% CI, 0.756-0.840) using clinical variables. Using the SOFA components for the same patient, XGBoost exhibited the highest AUC of 0.678 (95% CI, 0.626-0.730). As interpreted by SHAP, albumin, lactate, blood urea nitrogen, and international normalization ratio were determined to significantly affect the results. Additionally, PF and PLTs in the SOFA component significantly influenced the prediction results. CONCLUSION: Newly established ML-based models achieved good prediction of mortality in patients with sepsis. Using several clinical variables acquired at the baseline can provide more accurate results for early predictions than using SOFA components. Additionally, the impact of each variable was identified.

Mortality Prediction by Kinetic Parameters of Lactate and S-Adenosylhomocysteine in a Cohort of Critically Ill Patients.

Tissue hypoxia is associated with the development of organ dysfunction and death in critically ill patients commonly captured using blood lactate. The kinetic parameters of serial lactate evaluations are superior at predicting mortality compared with single values. S-adenosylhomocysteine (SAH), which is also associated with hypoxia, was recently established as a useful predictor of septic organ dysfunction and death. We evaluated the performance of kinetic SAH parameters for mortality prediction compared with lactate parameters in a cohort of critically ill patients. For lactate and SAH, maxima and means as well as the normalized area scores were calculated for two periods: the first 24 h and the total study period of up to five days following ICU admission. Their performance in predicting in-hospital mortality were compared in 99 patients. All evaluated parameters of lactate and SAH were significantly higher in non-survivors compared with survivors. In univariate analysis, the predictive power for mortality of SAH was higher compared with lactate in all forms of application. Multivariable models containing SAH parameters demonstrated higher predictive values for mortality than models based on lactate parameters. The optimal models for mortality prediction incorporated both lactate and SAH parameters. Compared with lactate, SAH displayed stronger predictive power for mortality in static and dynamic application in critically ill patients.

Machine Learning-Based Mortality Prediction in Chronic Kidney Disease among Heart Failure Patients: Insights and Outcomes from the Jordanian Heart Failure Registry.

Background and Objectives: Heart failure (HF) is a prevalent and debilitating condition that imposes a significant burden on healthcare systems and adversely affects the quality of life of patients worldwide. Comorbidities such as chronic kidney disease (CKD), arterial hypertension, and diabetes mellitus (DM) are common among HF patients, as they share similar risk factors. This study aimed to identify the prognostic significance of multiple factors and their correlation with disease prognosis and outcomes in a Jordanian cohort. Materials and Methods: Data from the Jordanian Heart Failure Registry (JoHFR) were analyzed, encompassing medical records from acute and chronic HF patients attending public and private cardiology clinics and hospitals across Jordan. An online form was utilized for data collection, focusing on three kidney function tests, estimated glomerular filtration rate (eGFR), blood urea nitrogen (BUN), and creatinine levels, with the eGFR calculated using the Cockcroft-Gault formula. We also built six machine learning models to predict mortality in our cohort. Results: From the JoHFR, 2151 HF patients were included, with 644, 1799, and 1927 records analyzed for eGFR, BUN, and creatinine levels, respectively. Age negatively impacted all measures (p ≤ 0.001), while smokers surprisingly showed better results than non-smokers (p ≤ 0.001). Males had more normal eGFR levels compared to females (p = 0.002). Comorbidities such as hypertension, diabetes, arrhythmias, and implanted devices were inversely related to eGFR (all with p-values <0.05). Higher BUN levels were associated with chronic HF, dyslipidemia, and ASCVD (p ≤ 0.001). Higher creatinine levels were linked to hypertension, diabetes, dyslipidemia, arrhythmias, and previous HF history (all with p-values <0.05). Low eGFR levels were associated with increased mechanical ventilation needs (p = 0.049) and mortality (p ≤ 0.001), while BUN levels did not significantly affect these outcomes. Machine learning analysis employing the Random Forest Classifier revealed that length of hospital stay and creatinine >115 were the most significant predictors of mortality. The classifier achieved an accuracy of 90.02% with an AUC of 80.51%, indicating its efficacy in predictive modeling. Conclusions: This study reveals the intricate relationship among kidney function tests, comorbidities, and clinical outcomes in HF patients in Jordan, highlighting the importance of kidney function as a predictive tool. Integrating machine learning models into clinical practice may enhance the predictive accuracy of patient outcomes, thereby supporting a more personalized approach to managing HF and related kidney dysfunction. Further research is necessary to validate these findings and to develop innovative treatment strategies for the CKD population within the HF cohort.

Is it Time to Develop an Indian Sepsis-related Mortality Prediction Score?

How to cite this article: Dedeepiya VD. Is it Time to Develop an Indian Sepsis-related Mortality Prediction Score? Indian J Crit Care Med 2024;28(4):320-322.

Author Response.

How to cite this article: Rahmatinejad Z, Hoseini B, Pourmand A, Reihani H, Rahmatinejad F, Eslami S, et al. Author Response. Indian J Crit Care Med 2024;28(2):183-184.

Modified NUTRIC Score as a Predictor of All-cause Mortality in Critically Ill Patients: A Systematic Review and Meta-analysis.

Purpose: The purpose of our meta-analysis was to look at the impact of modified nutrition risk in the critically ill (mNUTRIC) on mortality in patients with critical illness. Materials and methods: Literature relevant to this meta-analysis was searched in PubMed, Web of Science, and Cochrane Library till 26 August 2023. Prospective or retrospective studies, patients >18 years of age, studies that reported on mortality and mNUTRIC (mNUTRIC cut-off score) were included. The QUIPS tool was used to evaluate the risk for bias in prognostic factors. Results: A total of 31 studies on mNUTRIC score, involving 13,271 patients were included. The summary area under the curve (sAUC) of 0.80 (95% CI: 0.76-0.83) illustrates the mNUTRIC score's strong discrimination. The pooled sensitivity was 0.79 (95% CI: 0.74-0.84) and pooled specificity was 0.68 (95% CI: 0.63-0.73). We found no discernible variation in the mNUTRIC's prediction accuracy among cut-off values of <5 and >5 in our subgroup analysis and sAUC values were 0.82 (95% CI: 0.78-0.85) and 0.78 (95% CI: 0.74-0.81), respectively. Conclusion: We observed that mNUTRIC can discriminate between critically ill individuals and predict their mortality. Prospero: CRD42023460292. How to cite this article: Prakash J, Verma S, Shrivastava P, Saran K, Kumari A, Raj K, et al. Modified NUTRIC Score as a Predictor of All-cause Mortality in Critically Ill Patients: A Systematic Review and Meta-analysis. Indian J Crit Care Med 2024;28(5):495-503.