A Machine-Learning Approach for Dynamic Prediction of Sepsis-Induced Coagulopathy in Critically Ill Patients With Sepsis

Association between the anion-gap and 28-day mortality in critically ill adult patients with sepsis: A retrospective cohort study

Metabolic acidosis is usually associated with the severity of the condition of patients with sepsis or septic shock. Serum anion gap (AG) is one of the indicators of response metabolism. This study was performed to investigate whether the initial serum AG is associated with the 28-day mortality in critically ill adult patients with sepsis. This retrospective cohort study, a total of 15,047 patients with confirmed Sepsis disease from 2008 to 2019 from the Medical Information Mart for Intensive Care IV (MIMIC-IV) v1.0 database. The MIMIC-IV database is a comprehensive, de-identified clinical dataset originating from the Beth Israel Deaconess Medical Center in Boston, it includes extensive data on intensive care unit (ICU) patients, such as vital signs, lab results, and medication orders, spanning multiple years, accessible to researchers through an application process. AG can be obtained by direct extraction in the MIMIC-IV database (itemid = 50,868 from the laboratory events table of mimic_hosp), inclusion of AG values for the first test on first day of ICU admission. The patients were grouped into quartiles according to the AG interquartile range. The primary outcome was the 28-day mortality. Multiple logistic regression analysis was used to calculate the odds ratio (OR), while accounting for potential confounders, and the robustness of the results were evaluated in subgroup analyses. Among the 15,047 patients included in this study, the average age was 65.9 ± 16.0 years, 42.5% were female, 66.1% were Caucasian, and the 28-day mortality rate was 17.9% (2686/15,047). Multiple logistic regression analysis revealed the 28-day mortality in every increase of AG (per SD mEq/L), there is an associated 1.2 times (OR 1.2, 95% CI 1.12-1.29, P < .001) increase. Increased 28-day mortality (OR 1.53, 95% confidence interval 1.29-1.81, P < .001) in the group with the AG (15-18 mEq/L), and (OR 1.69, 95% confidence interval 1.4-2.04, P < .001) in the group with the highest AG (≥18 mEq/L), AG (<12 mEq/L) as a reference group, in the fully adjusted model. In adult patients with sepsis, the early AG at the time of ICU admission is an independent risk factor for prognosis.

Predicting the onset of overweight in Chinese high school students: a machine-learning approach in a one-year prospective cohort study

OBJECTIVE

This study aimed to develop and evaluate machine-learning models for predicting the onset of overweight in adolescents aged 14‒17, utilizing easily collectible personal information.

METHODS

This study was a one-year prospective cohort study. Baseline data were collected through anthropometric measurements and questionnaires, and the incidence of overweight was calculated one year later via anthropometric measurements. Predictive factors were selected through univariate analysis. Six machine-learning models were developed for predicting the onset of overweight. The SHapley Additive exPlanations (SHAP) was used for global and local interpretation of the models.

RESULTS

Out of 1,241 adolescents, 204 (16.4%) were identified as overweight after one year. Nineteen features were associated with the overweight incidence in univariable analysis. Participants were randomly divided into a training group and a testing group in a 7:3 ratio. The Light Gradient Boosting Machine (LGBM) algorithm achieved outperformed other models, achieving the following metrics: Accuracy (0.956), Recall (0.812), Specificity (0.983), F1-score (0.855), AUC (0.961). Importance ranking revealed that the top 11 minimal feature set can maintain the stability of model performance.

CONCLUSIONS

The onset of overweight in adolescents was accurately predicted using easily collectible personal information. The LGBM-based model exhibited superior performance. Oversampling technique notably improved model performance. The model interpretation technique provided innovative strategies for managing adolescent overweight/obesity.

Early prediction of sepsis-induced respiratory tract infection using a biomarker-based machine-learning algorithm

Early and differential diagnosis of sepsis is essential to avoid unnecessary antibiotic use and further reduce patient morbidity and mortality. Here, we aimed to identify predictors of sepsis and advance a machine-learning strategy to predict sepsis-induced respiratory tract infection (RTI). Patients with sepsis and RTI were selected via retrospective analysis, and essential population characteristics and laboratory parameters were recorded. To improve the performance of the primary model and avoid over-fitting, a recursive feature elimination with cross-validation (RFECV) strategy was used to screen the optimal subset of biomarkers and construct nine machine-learning models based on this subset; the average accuracy, precision, recall, and F1-score were used for evaluation of the models. We identified 430 patients with sepsis and 686 patients with RTI. A total of 39 features were collected, with 23 features identified for initial model construction. Using the RFECV algorithm, we found that the XGBoost classifier, which only needed to include seven biomarkers, demonstrated the best performance among all prediction models, with an average accuracy of 89.24 ± 2.28, while the Ridge classifier, which included 11 biomarkers, had an average accuracy of only 83.87 ± 4.69. The remaining models had prediction accuracies greater than 88%. We developed nine models for predicting sepsis using a strategy that combined RFECV with machine learning. Among these models, the XGBoost classifier, which included seven biomarkers, showed the best performance and highest accuracy for predicting sepsis and may be a promising tool for the timely identification of sepsis.

Effects of tacrolimus on proteinuria in Chinese and Indian patients with idiopathic membranous nephropathy: the results of machine learning study

PURPOSE

The present study aims to explore the effects of tacrolimus on proteinuria in patients with idiopathic membranous nephropathy (IMN) and recommend an appropriate dosage schedule via machine learning method.

METHODS

The Emax model was constructed to analyze the effects of tacrolimus on proteinuria in patients with IMN. Data were mined from published literature and machine learning was built up with Emax model, among which the efficacy indicator was proteinuria change rates from baseline. 463 IMN patients were included for modeling, and tacrolimus therapeutic window concentrations were 4-10 ng/ml.

RESULTS

In machine learning model, the Emax from tacrolimus effecting proteinuria in IMN patients was -72.7%, the ET50 was 0.43 months, and the time to achieving 25% Emax, 50% Emax, 75% Emax, and 80% (plateau) Emax of tacrolimus on proteinuria in patients with IMN were 0.15, 0.43, 1.29, and 1.72 months, respectively.

CONCLUSION

For achieving better therapeutic effects from tacrolimus on proteinuria in patients with IMN, tacrolimus concentration range need to be maintained at 4-10 ng/ml for at least 1.72 months.

The prediction of in-hospital mortality in elderly patients with sepsis-associated acute kidney injury utilizing machine learning models

Background

Sepsis-associated acute kidney injury (SA-AKI) is a severe complication associated with poorer prognosis and increased mortality, particularly in elderly patients. Currently, there is a lack of accurate mortality risk prediction models for these patients in clinic.

Objectives

This study aimed to develop and validate machine learning models for predicting in-hospital mortality risk in elderly patients with SA-AKI.

Methods

Machine learning models were developed and validated using the public, high-quality Medical Information Mart for Intensive Care (MIMIC)-IV critically ill database. The recursive feature elimination (RFE) algorithm was employed for key feature selection. Eleven predictive models were compared, with the best one selected for further validation. Shapley Additive Explanations (SHAP) values were used for visualization and interpretation, making the machine learning models clinically interpretable.

Results

There were 16,154 patients with SA-AKI in the MIMIC-IV database, and 8426 SA-AKI patients were included in this study (median age: 77.0 years; female: 45%). 7728 patients excluded based on these criteria. They were randomly divided into a training cohort (5,934, 70%) and a validation cohort (2,492, 30%). Nine key features were selected by the RFE algorithm. The CatBoost model achieved the best performance, with an AUC of 0.844 in the training cohort and 0.804 in the validation cohort. SHAP values revealed that AKI stage, PaO2, and lactate were the top three most important features contributing to the CatBoost model.

Conclusion

We developed a model capable of predicting the risk of in-hospital mortality in elderly patients with SA-AKI.

Association of serum phosphate and changes in serum phosphate with 28-day mortality in septic shock from MIMIC-IV database

This study aimed to investigate the relationship between serum phosphate levels, changes in serum phosphate levels, and 28-day mortality in patients with septic shock. In this retrospective study, data were collected from the Medical Information Mart for Intensive Care IV (MIMIC-IV) database between 2008 and 2019. Patients were divided into three groups according to the tertiles of serum phosphate levels. Kaplan-Meier curves and log-rank test analyses were used for survival analysis. Multivariate logistic regression, and restricted cubic spline (RCS) curve were used to explore the association between serum phosphate, delta serum phosphate levels and 28-day mortality. In total, 3296 patients with septic shock were included in the study, and the 28-day mortality was 30.0%. Serum phosphate levels were significantly higher in the non-survivor group than in the survivor group. The Kaplan-Meier curves showed significant differences among the three groups. Multivariate logistic regression analysis and the RCS curve showed that serum phosphate levels were independently and positively associated with the 28-day mortality of septic shock. Non-survivors had higher delta serum phosphate levels than survivors. Survival analysis showed that patients with higher delta serum phosphate levels had higher 28-day mortality. A non-linear relationship was detected between delta serum phosphate and 28-day mortality with a point of inflection at - 0.3 mg/dL. Serum phosphate levels were positively and independently associated with 28-day mortality in septic shock. Delta serum phosphate level was a high-risk factor for patients with septic shock.

Explainable machine learning model for predicting furosemide responsiveness in patients with oliguric acute kidney injury

BACKGROUND

Although current guidelines didn't support the routine use of furosemide in oliguric acute kidney injury (AKI) management, some patients may benefit from furosemide administration at an early stage. We aimed to develop an explainable machine learning (ML) model to differentiate between furosemide-responsive (FR) and furosemide-unresponsive (FU) oliguric AKI.

METHODS

From Medical Information Mart for Intensive Care-IV (MIMIC-IV) and eICU Collaborative Research Database (eICU-CRD), oliguric AKI patients with urine output (UO) < 0.5 ml/kg/h for the first 6 h after ICU admission and furosemide infusion ≥ 40 mg in the following 6 h were retrospectively selected. The MIMIC-IV cohort was used in training a XGBoost model to predict UO > 0.65 ml/kg/h during 6-24 h succeeding the initial 6 h for assessing oliguria, and it was validated in the eICU-CRD cohort. We compared the predictive performance of the XGBoost model with the traditional logistic regression and other ML models.

RESULTS

6897 patients were included in the MIMIC-IV training cohort, with 2235 patients in the eICU-CRD validation cohort. The XGBoost model showed an AUC of 0.97 (95% CI: 0.96-0.98) for differentiating FR and FU oliguric AKI. It outperformed the logistic regression and other ML models in correctly predicting furosemide diuretic response, achieved 92.43% sensitivity (95% CI: 90.88-93.73%) and 95.12% specificity (95% CI: 93.51-96.3%).

CONCLUSION

A boosted ensemble algorithm can be used to accurately differentiate between patients who would and would not respond to furosemide in oliguric AKI. By making the model explainable, clinicians would be able to better understand the reasoning behind the prediction outcome and make individualized treatment.

A Retrospective Cohort Study: Predicting 90-Day Mortality for ICU Trauma Patients with a Machine Learning Algorithm Using XGBoost Using MIMIC-III Database

Objective

The aim of this study was to develop and validate a machine learning-based predictive model that predicts 90-day mortality in ICU trauma patients.

Methods

Data of patients with severe trauma were extracted from the Medical Information Mart for Intensive Care III (MIMIC-III) database. The performances of mortality prediction models generated using nine machine learning extreme gradient boosting (XGBoost), logistic regression, random forest, AdaBoost, multilayer perceptron (MLP) neural networks, support vector machine (SVM), light gradient boosting machine (GBM), k nearest neighbors (KNN) and gaussian naive bayes (GNB). The performance of the model was evaluated in terms of discrimination, calibration and clinical application.

Results

We found that the accuracy, sensitivity, specificity, PPV, NPV and F1 score of our proposed XGBoost model were 82.8%, 79.7%, 77.6%, 51.2%, 91.5% and 0.624, respectively. Among the nine models, the XGBoost model performed best. Compared with traditional logistic regression, the calibration curves of the XGBoost model and decision curve analysis (DCA) performed well.

Conclusion

Our study shows that the XGBoost model outperforms other machine learning models in predicting 90-day mortality in trauma patients. It can be used to assist clinicians in the early identification of mortality risk factors and early intervention to reduce mortality.

Comparison of the prognostic value of four different critical illness scores in patients with sepsis-induced coagulopathy

In patients with sepsis-induced coagulopathy (SIC), the Chinese DIC scoring system (CDSS) of the Chinese Society of Thrombosis and Hemostasis score, the Japanese Association for Acute Medicine (JAAM) score, the International Society of Thrombosis and Hemostasis (ISTH), and the Can Rapid risk stratification of Unstable angina patients Suppress Adverse outcomes with Early implementation of the ACC/AHA Guidelines (CRUSADE) score were compared for their predictive significance (SIC). From August 2021 through August 2022, 92 SIC patients hospitalized in our hospital's Department of Critical Care Medicine served as study participants. Groups of patients were created with a bad prognosis (n = 35) and a favorable prognosis (n = 57) 14 days following admission. Electronic medical records were used to compile patient information such as demographics (gender, age, and body mass index), medical history (hypertension, diabetes, chronic obstructive pulmonary disease, and chronic kidney disease), treatment (mechanical ventilation, APACHE II score at admission), and outcomes (results). All patients' JAAM, CDSS, ISTH, and CRUSADE scores were recorded. The APACHE II scores of the group with a poor prognosis were noticeably (p < 0.05) higher upon admission than those of the group with a favorable prognosis. The poor prognosis group had higher JAAM, ISTH, CDSS, and CRUSADE scores than the good prognosis group (all p < 0.05). Partial coagulation indicators in fibrinogen, D-dimer, activated partial thromboplastin time, and prothrombin time were positively linked with JAAM, ISTH, CDSS, and CRUSADE (all p < 0.05). At admission, the JAAM, ISTH, CDSS, CRUSADE, and APACHE II scores were independently linked with SIC patients' prognosis (all p < 0.05) in a multivariate logistic regression analysis. According to receiver operating characteristic analysis, the area under the curve for predicting the prognosis of SIC patients using the JAAM, ISTH, CDSS, and CRUSADE4 scores was 0.896, 0.870, 0.852, and 0.737, respectively, with 95% CI being 0.840-0.952, 0.805-0.936, 0.783-0.922 and 0.629-0.845, respectively (all p < 0.05). The prognosis of SIC patients may be predicted in part by their JAAM, ISTH, CDSS, and CRUSADE4 scores, with the CDSS score being the most accurate. This research provides important recommendations for improving the care of patients with SIC.

Does multidimensional daily information predict the onset of myopia? A 1-year prospective cohort study

PURPOSE

This study aimed to develop an interpretable machine learning model to predict the onset of myopia based on individual daily information.

METHOD

This study was a prospective cohort study. At baseline, non-myopia children aged 6-13 years old were recruited, and individual data were collected through interviewing students and parents. One year after baseline, the incidence of myopia was evaluated based on visual acuity test and cycloplegic refraction measurement. Five algorithms, Random Forest, Support Vector Machines, Gradient Boosting Decision Tree, CatBoost and Logistic Regression were utilized to develop different models and their performance was validated by area under curve (AUC). Shapley Additive exPlanations was applied to interpret the model output on the individual and global level.

RESULT

Of 2221 children, 260 (11.7%) developed myopia in 1 year. In univariable analysis, 26 features were associated with the myopia incidence. Catboost algorithm had the highest AUC of 0.951 in the model validation. The top 3 features for predicting myopia were parental myopia, grade and frequency of eye fatigue. A compact model using only 10 features was validated with an AUC of 0.891.

CONCLUSION

The daily information contributed reliable predictors for childhood's myopia onset. The interpretable Catboost model presented the best prediction performance. Oversampling technology greatly improved model performance. This model could be a tool in myopia preventing and intervention that can help identify children who are at risk of myopia, and provide personalized prevention strategies based on contributions of risk factors to the individual prediction result.

Machine learning for the prediction of all-cause mortality in patients with sepsis-associated acute kidney injury during hospitalization

Background

Sepsis-associated acute kidney injury (S-AKI) is considered to be associated with high morbidity and mortality, a commonly accepted model to predict mortality is urged consequently. This study used a machine learning model to identify vital variables associated with mortality in S-AKI patients in the hospital and predict the risk of death in the hospital. We hope that this model can help identify high-risk patients early and reasonably allocate medical resources in the intensive care unit (ICU).

Methods

A total of 16,154 S-AKI patients from the Medical Information Mart for Intensive Care IV database were examined as the training set (80%) and the validation set (20%). Variables (129 in total) were collected, including basic patient information, diagnosis, clinical data, and medication records. We developed and validated machine learning models using 11 different algorithms and selected the one that performed the best. Afterward, recursive feature elimination was used to select key variables. Different indicators were used to compare the prediction performance of each model. The SHapley Additive exPlanations package was applied to interpret the best machine learning model in a web tool for clinicians to use. Finally, we collected clinical data of S-AKI patients from two hospitals for external validation.

Results

In this study, 15 critical variables were finally selected, namely, urine output, maximum blood urea nitrogen, rate of injection of norepinephrine, maximum anion gap, maximum creatinine, maximum red blood cell volume distribution width, minimum international normalized ratio, maximum heart rate, maximum temperature, maximum respiratory rate, minimum fraction of inspired O2, minimum creatinine, minimum Glasgow Coma Scale, and diagnosis of diabetes and stroke. The categorical boosting algorithm model presented significantly better predictive performance [receiver operating characteristic (ROC): 0.83] than other models [accuracy (ACC): 75%, Youden index: 50%, sensitivity: 75%, specificity: 75%, F1 score: 0.56, positive predictive value (PPV): 44%, and negative predictive value (NPV): 92%]. External validation data from two hospitals in China were also well validated (ROC: 0.75).

Conclusions

After selecting 15 crucial variables, a machine learning-based model for predicting the mortality of S-AKI patients was successfully established and the CatBoost model demonstrated best predictive performance.

Management Strategies in Septic Coagulopathy: A Review of the Current Literature

One of the 'organs' that can be affected by sepsis is the coagulation system. Coagulopathy in sepsis may take the form of sepsis-induced coagulopathy (SIC) or sepsis-associated disseminated intravascular coagulation (DIC). It is important to identify SIC early, as at this stage of coagulopathy anticoagulants may be of the greatest benefit. The most recent diagnostic scoring systems for septic coagulopathy come from the International Society on Thrombosis and Hemostasis and the Japanese Association for Acute Medicine. Recommendations regarding the management of septic coagulopathy differ between organizations. Moreover, septic coagulopathy is an area of intense research in recent years. Therefore we searched three databases to review the most recent management strategies in septic coagulopathy. The mainstream management strategies in septic coagulopathy include the causal treatment of sepsis, unfractionated heparin, low-molecular-weight heparin, antithrombin, and recombinant human thrombomodulin. The last two have been associated with the highest survival benefit. Nevertheless, the indiscriminate use of these anticoagulants should be avoided due to the lack of mortality benefit and increased risk of bleeding. The early diagnosis of SIC and monitoring of coagulation status during sepsis is crucial for the timely management and selection of the most suitable treatment at a time. New directions in septic coagulopathy include new diagnostic biomarkers, dynamic diagnostic models, genetic markers for SIC management, and new therapeutic agents. These new research avenues may potentially result in timelier SIC diagnosis and improved management of all stages of septic coagulopathy by making it more effective, safe, and personalized.

Predicting pneumonia during hospitalization in flail chest patients using machine learning approaches

Objective

Pneumonia is a common pulmonary complication of flail chest, causing high morbidity and mortality rates in affected patients. The existing methods for identifying pneumonia have low accuracy, and their use may delay antimicrobial therapy. However, machine learning can be combined with electronic medical record systems to identify information and assist in quick clinical decision-making. Our study aimed to develop a novel machine-learning model to predict pneumonia risk in flail chest patients.

Methods

From January 2011 to December 2021, the electronic medical records of 169 adult patients with flail chest at a tertiary teaching hospital in an urban level I Trauma Centre in Chongqing were retrospectively analysed. Then, the patients were randomly divided into training and test sets at a ratio of 7:3. Using the Fisher score, the best subset of variables was chosen. The performance of the seven models was evaluated by computing the area under the receiver operating characteristic curve (AUC). The output of the XGBoost model was shown using the Shapley Additive exPlanation (SHAP) method.

Results

Of 802 multiple rib fracture patients, 169 flail chest patients were eventually included, and 86 (50.80%) were diagnosed with pneumonia. The XGBoost model performed the best among all seven machine-learning models. The AUC of the XGBoost model was 0.895 (sensitivity: 84.3%; specificity: 80.0%).Pneumonia in flail chest patients was associated with several features: systolic blood pressure, pH value, blood transfusion, and ISS.

Conclusion

Our study demonstrated that the XGBoost model with 32 variables had high reliability in assessing risk indicators of pneumonia in flail chest patients. The SHAP method can identify vital pneumonia risk factors, making the XGBoost model's output clinically meaningful.

Potential Biomarkers for Early Diagnosis, Evaluation, and Prognosis of Sepsis-Induced Coagulopathy

Sepsis-induced coagulopathy (SIC) is a life-threatening complication characterized by the systemic activation of coagulation in sepsis. The diagnostic criteria of SIC consist of three items, including Sequential Organ Failure Assessment (SOFA) score, platelet count, and prothrombin time (PT)-international normalized ratio (INR). SIC has a high prevalence and it can lead to a higher mortality rate and longer length of hospital and ICU stay. Thus, the early detection of SIC is extremely important. It is unfortunate that there is still no precise biomarker for early diagnosis and assessment of the prognosis of SIC. We reviewed the current literature and discovered that some potential biomarkers, such as soluble thrombomodulin (sTM), thrombin-antithrombin complex (TAT), tissue plasminogen activator-inhibitor complex (t-PAIC), α2-plasmin inhibitor-plasmin complex (PIC), C-type lectin-like receptor 2 (CLEC-2), neutrophil extracellular traps (NETs), prothrombin fragment 1.2 (F1.2), Angiopoietin-2 (Ang-2), plasminogen activator inhibitor-1 (PAI-1), and tissue inhibitor of metalloproteinase-1 (TIMP-1) may be useful for early diagnosis, evaluation, and prognosis of SIC. Early initiation of treatment without missing any therapeutic opportunities may improve SIC patients' prognosis. Further large-scale clinical studies are still needed to confirm the role of these biomarkers in the diagnosis and prognosis assessment of SIC.

Prognostic Accuracy of the Different Scoring Systems for Assessing Coagulopathy in Sepsis: A Retrospective Study

There is no gold standard for the diagnosis of coagulation dysfunction in sepsis, and the use of the current scoring systems is still controversial. The purpose of this study was to assess the performance of sepsis-induced coagulopathy (SIC), the Japanese Association for Acute Medicine Disseminated Intravascular Coagulation (JAAM DIC), and the International Society on Thrombosis and Haemostasis overt DIC (ISTH overt-DIC). The relationship between each scoring system and 28-day all-cause mortality was examined. Among 452 patients (mean age, 65 [48,76] years), 306 [66.7%] were men, the median SOFA score was 6 [4,9], and the median APACHE II score was 15 [11,22]. A total of 132 patients (29.2%) died within 28 days. Both the diagnosis of SIC (AUROC, 0.779 [95% CI, 0.728-0.830], P < 0.001) and ISTH overt-DIC (AUROC, 0.782 [95% CI, 0.732-0.833], P < 0.001) performed equally well in the discrimination of 28-day all-cause mortality (between-group difference: SIC versus ISTH overt-DIC, -0.003 [95% CI, -0.025-0.018], P = 0.766). However, the SIC demonstrated greater calibration for 28-day all-cause mortality than ISTH overt-DIC (the coincidence of the calibration curve of the former is higher than that of the latter). The diagnosis of JAAM DIC was not independently associated with 28-day all-cause mortality in sepsis (RR, 1.115, [95% CI 0.660-1.182], P = 0.684). The SIC scoring system demonstrated superior prognostic prediction ability in comparison with the others and is the most appropriate standard for diagnosing coagulopathy in sepsis.

Association between hypomagnesemia and coagulopathy in sepsis: a retrospective observational study

BACKGROUND

Hypomagnesemia reportedly has significant associations with poor clinical outcomes such as increased mortality and septic shock in patients with sepsis. Although the mechanism underlying these outcomes mostly remains unclear, some experimental data suggest that magnesium deficiency could potentiate coagulation activation in sepsis. However, in sepsis, the association between serum magnesium levels and coagulopathy, including disseminated intravascular coagulation (DIC), remains unknown. Thus, we aimed to investigate the relationship between serum magnesium levels and coagulation status and the association between hypomagnesemia and DIC in patients with sepsis.

METHODS

This retrospective observational study was conducted at the intensive care unit (ICU) of a university hospital from June 2011 to December 2017. Patients older than 19 years who met the Sepsis-3 definition were included. We categorized patients into three groups according to their serum magnesium levels: hypomagnesemia (< 1.6 mg/dL), normal serum magnesium level (1.6-2.4 mg/dL), and hypermagnesemia (> 2.4 mg/dL). We investigated the association between serum magnesium levels and overt DIC at the time of ICU admission according to the criteria of the International Society on Thrombosis and Haemostasis.

RESULTS

Among 753 patients included in this study, 181 had DIC, 105 had hypomagnesemia, 552 had normal serum magnesium levels, and 96 had hypermagnesemia. Patients with hypomagnesemia had a more activated coagulation status indicated by lower platelet counts, lower fibrinogen levels, higher prothrombin time-international normalized ratios, higher thrombin-antithrombin complex, and more frequent DIC than those with normal serum magnesium levels and hypermagnesemia (DIC: 41.9% vs. 20.6% vs. 24.0%, P < 0.001). The coagulation status in patients with hypomagnesemia was more augmented toward suppressed fibrinolysis than that in patients with normal serum magnesium levels and hypermagnesemia. Multivariate logistic regression revealed that hypomagnesemia was independently associated with DIC (odds ratio, 1.69; 95% confidence interval, 1.00-2.84; P = 0.048) after adjusting for several confounding variables.

CONCLUSIONS

Patients with hypomagnesemia had a significantly activated coagulation status and suppressed fibrinolysis. Hypomagnesemia was independently associated with DIC in patients with sepsis. Therefore, the treatment of hypomagnesemia may be a potential therapeutic strategy for the treatment of coagulopathy in sepsis.

Machine learning for early prediction of sepsis-associated acute brain injury

Background

Sepsis-associated encephalopathy (SAE) is defined as diffuse brain dysfunction associated with sepsis and leads to a high mortality rate. We aimed to develop and validate an optimal machine-learning model based on clinical features for early predicting sepsis-associated acute brain injury.

Methods

We analyzed adult patients with sepsis from the Medical Information Mart for Intensive Care (MIMIC III) clinical database. Candidate models were trained using random forest, support vector machine (SVM), decision tree classifier, gradients boosting machine (GBM), multiple layer perception (MLP), extreme gradient boosting (XGBoost), light gradients boosting machine (LGBM) and a conventional logistic regression model. These methods were applied to develop and validate the optimal model based on its accuracy and area under curve (AUC).

Results

In total, 12,460 patients with sepsis met inclusion criteria, and 6,284 (50.4%) patients suffered from sepsis-associated acute brain injury. Compared other models, the LGBM model achieved the best performance. The AUC for both train set and test set indicated excellent validity (Trainset AUC 0.91, Testset AUC 0.87). Feature importance analysis showed that glucose, age, mean arterial pressure, heart rate, hemoglobin, and length of ICU stay were the top 6 important clinical factors to predict occurrence of sepsis-associated acute brain injury.

Conclusion

Almost half of patients admitted to ICU with sepsis had sepsis-associated acute brain injury. The LGBM model better identify patients with sepsis-associated acute brain injury than did other machine-learning models. Glucose, age, and mean arterial pressure were the three most important clinical factors to predict occurrence of sepsis-associated acute brain injury.

Predicting acute kidney injury risk in acute myocardial infarction patients: An artificial intelligence model using medical information mart for intensive care databases

Background

Predictive models based on machine learning have been widely used in clinical practice. Patients with acute myocardial infarction (AMI) are prone to the risk of acute kidney injury (AKI), which results in a poor prognosis for the patient. The aim of this study was to develop a machine learning predictive model for the identification of AKI in AMI patients.

Methods

Patients with AMI who had been registered in the Medical Information Mart for Intensive Care (MIMIC) III and IV database were enrolled. The primary outcome was the occurrence of AKI during hospitalization. We developed Random Forests (RF) model, Naive Bayes (NB) model, Support Vector Machine (SVM) model, eXtreme Gradient Boosting (xGBoost) model, Decision Trees (DT) model, and Logistic Regression (LR) models with AMI patients in MIMIC-IV database. The importance ranking of all variables was obtained by the SHapley Additive exPlanations (SHAP) method. AMI patients in MIMIC-III databases were used for model evaluation. The area under the receiver operating characteristic curve (AUC) was used to compare the performance of each model.

Results

A total of 3,882 subjects with AMI were enrolled through screening of the MIMIC database, of which 1,098 patients (28.2%) developed AKI. We randomly assigned 70% of the patients in the MIMIC-IV data to the training cohort, which is used to develop models in the training cohort. The remaining 30% is allocated to the testing cohort. Meanwhile, MIMIC-III patient data performs the external validation function of the model. 3,882 patients and 37 predictors were included in the analysis for model construction. The top 5 predictors were serum creatinine, activated partial prothrombin time, blood glucose concentration, platelets, and atrial fibrillation, (SHAP values are 0.670, 0.444, 0.398, 0.389, and 0.381, respectively). In the testing cohort, using top 20 important features, the models of RF, NB, SVM, xGBoost, DT model, and LR obtained AUC of 0.733, 0.739, 0.687, 0.689, 0.663, and 0.677, respectively. Placing RF models of number of different variables on the external validation cohort yielded their AUC of 0.711, 0.754, 0.778, 0.781, and 0.777, respectively.

Conclusion

Machine learning algorithms, particularly the random forest algorithm, have improved the accuracy of risk stratification for AKI in AMI patients and are applied to accurately identify the risk of AKI in AMI patients.

Artificial Intelligence and Machine Learning in Patient Blood Management: A Scoping Review

Machine learning (ML) and artificial intelligence (AI) are widely used in many different fields of modern medicine. This narrative review gives, in the first part, a brief overview of the methods of ML and AI used in patient blood management (PBM) and, in the second part, aims at describing which fields have been analyzed using these methods so far. A total of 442 articles were identified by a literature search, and 47 of them were judged as qualified articles that applied ML and AI techniques in PBM. We assembled the eligible articles to provide insights into the areas of application, quality measures of these studies, and treatment outcomes that can pave the way for further adoption of this promising technology and its possible use in routine clinical decision making. The topics that have been investigated most often were the prediction of transfusion (30%), bleeding (28%), and laboratory studies (15%). Although in the last 3 years a constantly increasing number of questions of ML in PBM have been investigated, there is a vast scientific potential for further application of ML and AI in other fields of PBM.

A Machine Learning Approach for the Prediction of Traumatic Brain Injury Induced Coagulopathy

Background: Traumatic brain injury-induced coagulopathy (TBI-IC), is a disease with poor prognosis and increased mortality rate.

Objectives: Our study aimed to identify predictors as well as develop machine learning (ML) models to predict the risk of coagulopathy in this population.

Methods: ML models were developed and validated based on two public databases named Medical Information Mart for Intensive Care (MIMIC)-IV and the eICU Collaborative Research Database (eICU-CRD). Candidate predictors, including demographics, family history, comorbidities, vital signs, laboratory findings, injury type, therapy strategy and scoring system were included. Models were compared on area under the curve (AUC), accuracy, sensitivity, specificity, positive and negative predictive values, and decision curve analysis (DCA) curve.

Results: Of 999 patients in MIMIC-IV included in the final cohort, a total of 493 (49.35%) patients developed coagulopathy following TBI. Recursive feature elimination (RFE) selected 15 variables, including international normalized ratio (INR), prothrombin time (PT), sepsis related organ failure assessment (SOFA), activated partial thromboplastin time (APTT), platelet (PLT), hematocrit (HCT), red blood cell (RBC), hemoglobin (HGB), blood urea nitrogen (BUN), red blood cell volume distribution width (RDW), creatinine (CRE), congestive heart failure, myocardial infarction, sodium, and blood transfusion. The external validation in eICU-CRD demonstrated that adapting boosting (Ada) model had the highest AUC of 0.924 (95% CI: 0.902–0.943). Furthermore, in the DCA curve, the Ada model and the extreme Gradient Boosting (XGB) model had relatively higher net benefits (ie, the correct classification of coagulopathy considering a trade-off between false- negatives and false-positives)—over other models across a range of threshold probability values.

Conclusions: The ML models, as indicated by our study, can be used to predict the incidence of TBI-IC in the intensive care unit (ICU).

A Novel Intercellular Communication-Associated Gene Signature for Prognostic Prediction and Clinical Value in Patients With Lung Adenocarcinoma

Background

Lung cancer remains the leading cause of cancer death globally, with lung adenocarcinoma (LUAD) being its most prevalent subtype. This study aimed to identify the key intercellular communication-associated genes (ICAGs) in LUAD.

Methods

Eight publicly available datasets were downloaded from the Gene Expression Omnibus (GEO) and The Cancer Genome Atlas (TCGA) databases. The prognosis-related ICAGs were identified and a risk score was developed by using survival analysis. Machine learning models were trained to predict LUAD recurrence based on the selected ICAGs and clinical information. Comprehensive analyses on ICAGs and tumor microenvironment were performed. A single-cell RNA-sequencing dataset was assessed to further elucidate aberrant changes in intercellular communication.

Results

Eight ICAGs with prognostic potential were identified in the present study, and a risk score was derived accordingly. The best machine-learning model to predict relapse was developed based on clinical information and the expression levels of these eight ICAGs. This model achieved a remarkable area under receiver operator characteristic curves of 0.841. Patients were divided into high- and low-risk groups according to their risk scores. DNA replication and cell cycle were significantly enriched by the differentially expressed genes between the high- and the low-risk groups. Infiltrating immune cells, immune functions were significantly related to ICAGs expressions and risk scores. Additionally, the changes of intercellular communication were modeled by analyzing the single-cell sequencing dataset.

Conclusion

The present study identified eight key ICAGs in LUAD, which could contribute to patient stratification and act as novel therapeutic targets.

Development and Validation of a Machine-Learning Model for Prediction of Extubation Failure in Intensive Care Units

Background: Extubation failure (EF) can lead to an increased chance of ventilator-associated pneumonia, longer hospital stays, and a higher mortality rate. This study aimed to develop and validate an accurate machine-learning model to predict EF in intensive care units (ICUs). Methods: Patients who underwent extubation in the Medical Information Mart for Intensive Care (MIMIC)-IV database were included. EF was defined as the need for ventilatory support (non-invasive ventilation or reintubation) or death within 48 h following extubation. A machine-learning model called Categorical Boosting (CatBoost) was developed based on 89 clinical and laboratory variables. SHapley Additive exPlanations (SHAP) values were calculated to evaluate feature importance and the recursive feature elimination (RFE) algorithm was used to select key features. Hyperparameter optimization was conducted using an automated machine-learning toolkit (Neural Network Intelligence). The final model was trained based on key features and compared with 10 other models. The model was then prospectively validated in patients enrolled in the Cardiac Surgical ICU of Zhongshan Hospital, Fudan University. In addition, a web-based tool was developed to help clinicians use our model. Results: Of 16,189 patients included in the MIMIC-IV cohort, 2,756 (17.0%) had EF. Nineteen key features were selected using the RFE algorithm, including age, body mass index, stroke, heart rate, respiratory rate, mean arterial pressure, peripheral oxygen saturation, temperature, pH, central venous pressure, tidal volume, positive end-expiratory pressure, mean airway pressure, pressure support ventilation (PSV) level, mechanical ventilation (MV) durations, spontaneous breathing trial success times, urine output, crystalloid amount, and antibiotic types. After hyperparameter optimization, our model had the greatest area under the receiver operating characteristic (AUROC: 0.835) in internal validation. Significant differences in mortality, reintubation rates, and NIV rates were shown between patients with a high predicted risk and those with a low predicted risk. In the prospective validation, the superiority of our model was also observed (AUROC: 0.803). According to the SHAP values, MV duration and PSV level were the most important features for prediction. Conclusions: In conclusion, this study developed and prospectively validated a CatBoost model, which better predicted EF in ICUs than other models.