UNASSIGNED: The application of machine learning in health care often necessitates the use of hierarchical codes such as the International Classification of Diseases (ICD) and Anatomical Therapeutic Chemical (ATC) systems. These codes classify diseases and medications, respectively, thereby forming extensive data dimensions. Unsupervised feature selection tackles the \"curse of dimensionality\" and helps to improve the accuracy and performance of supervised learning models by reducing the number of irrelevant or redundant features and avoiding overfitting. Techniques for unsupervised feature selection, such as filter, wrapper, and embedded methods, are implemented to select the most important features with the most intrinsic information. However, they face challenges due to the sheer volume of ICD and ATC codes and the hierarchical structures of these systems.
UNASSIGNED: The objective of this study was to compare several unsupervised feature selection methods for ICD and ATC code databases of patients with coronary artery disease in different aspects of performance and complexity and select the best set of features representing these patients.
UNASSIGNED: We compared several unsupervised feature selection methods for 2 ICD and 1 ATC code databases of 51,506 patients with coronary artery disease in Alberta, Canada. Specifically, we used the Laplacian score, unsupervised feature selection for multicluster data, autoencoder-inspired unsupervised feature selection, principal feature analysis, and concrete autoencoders with and without ICD or ATC tree weight adjustment to select the 100 best features from over 9000 ICD and 2000 ATC codes. We assessed the selected features based on their ability to reconstruct the initial feature space and predict 90-day mortality following discharge. We also compared the complexity of the selected features by mean code level in the ICD or ATC tree and the interpretability of the features in the mortality prediction task using Shapley analysis.
UNASSIGNED: In feature space reconstruction and mortality prediction, the concrete autoencoder-based methods outperformed other techniques. Particularly, a weight-adjusted concrete autoencoder variant demonstrated improved reconstruction accuracy and significant predictive performance enhancement, confirmed by DeLong and McNemar tests (P<.05). Concrete autoencoders preferred more general codes, and they consistently reconstructed all features accurately. Additionally, features selected by weight-adjusted concrete autoencoders yielded higher Shapley values in mortality prediction than most alternatives.
UNASSIGNED: This study scrutinized 5 feature selection methods in ICD and ATC code data sets in an unsupervised context. Our findings underscore the superiority of the concrete autoencoder method in selecting salient features that represent the entire data set, offering a potential asset for subsequent machine learning research. We also present a novel weight adjustment approach for the concrete autoencoders specifically tailored for ICD and ATC code data sets to enhance the generalizability and interpretability of the selected features.

BACKGROUND: Older patients presenting to the emergency department (ED) after falling are increasingly prevalent. Falls are associated with functional decline and death. Biomarkers predicting short-term mortality might facilitate decisions regarding resource allocation and disposition. D-dimer levels are used to rule out thromboembolic disease, while copeptin and adrenomedullin (MR-proADM) may be used as measures of the patient`s stress level. These nonspecific biomarkers were selected as potential predictors for mortality.
METHODS: Prospective, international, multicenter, cross-sectional observation was performed in two tertiary and two regional hospitals in Germany and Switzerland. Patients aged 65 years or older presenting to the ED after a fall were enrolled. Demographic data, Activities of Daily Living (ADL), and D-dimers were collected upon presentation. Copeptin and MR-proADM levels were determined from frozen samples. Primary outcome was 30-day mortality; and secondary outcomes were mortality at 90, 180, and 365 days.
RESULTS: Five hundred and seventy-two patients were included. Median age was 83 [IQR 78, 89] years, 236 (67.7%) were female. Mortality overall was 3.1% (30 d), 5.4% (90 d), 7.5% (180 d), and 13.8% (365 d), respectively. Non-survivors were older, had a lower ADL index and higher levels of all three biomarkers. Elevated levels of MR-proADM and D-dimer were associated with higher risk of mortality. MR-proADM and D-dimer showed high sensitivity and low negative likelihood ratio regarding short-term mortality, whereas copeptin did not.
CONCLUSIONS: D-dimer and MR-proADM levels might be useful as prognostic markers in older patients presenting to the ED after a fall, by identifying patients at low risk of short-term mortality.
BACKGROUND: ClinicalTrials.gov Identifier: NCT02244983.

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.
CONCLUSIONS: 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.

BACKGROUND: Machine learning techniques are starting to be used in various health care data sets to identify frail persons who may benefit from interventions. However, evidence about the performance of machine learning techniques compared to conventional regression is mixed. It is also unclear what methodological and database factors are associated with performance.
OBJECTIVE: This study aimed to compare the mortality prediction accuracy of various machine learning classifiers for identifying frail older adults in different scenarios.
METHODS: We used deidentified data collected from older adults (65 years of age and older) assessed with interRAI-Home Care instrument in New Zealand between January 1, 2012, and December 31, 2016. A total of 138 interRAI assessment items were used to predict 6-month and 12-month mortality, using 3 machine learning classifiers (random forest [RF], extreme gradient boosting [XGBoost], and multilayer perceptron [MLP]) and regularized logistic regression. We conducted a simulation study comparing the performance of machine learning models with logistic regression and interRAI Home Care Frailty Scale and examined the effects of sample sizes, the number of features, and train-test split ratios.
RESULTS: A total of 95,042 older adults (median age 82.66 years, IQR 77.92-88.76; n=37,462, 39.42% male) receiving home care were analyzed. The average area under the curve (AUC) and sensitivities of 6-month mortality prediction showed that machine learning classifiers did not outperform regularized logistic regressions. In terms of AUC, regularized logistic regression had better performance than XGBoost, MLP, and RF when the number of features was ≤80 and the sample size ≤16,000; MLP outperformed regularized logistic regression in terms of sensitivities when the number of features was ≥40 and the sample size ≥4000. Conversely, RF and XGBoost demonstrated higher specificities than regularized logistic regression in all scenarios.
CONCLUSIONS: The study revealed that machine learning models exhibited significant variation in prediction performance when evaluated using different metrics. Regularized logistic regression was an effective model for identifying frail older adults receiving home care, as indicated by the AUC, particularly when the number of features and sample sizes were not excessively large. Conversely, MLP displayed superior sensitivity, while RF exhibited superior specificity when the number of features and sample sizes were large.

UNASSIGNED: Globally, trauma cases have significant morbidity and mortality. Hence, various scoring systems have been designed to improve the prognosis in trauma cases. Trauma and Injury Severity Score (TRISS) is one of the widely used models to predict mortality; however, it has certain limitation. We have aimed to evaluate the survival prediction of new model TRISS-oxygen saturation (SpO2) and to compare with original TRISS score in trauma study participants.
UNASSIGNED: This was a prospective cohort study conducted on 380 trauma study participants admitted to the surgery department from January 20, 2021, to November 28, 2021. The proposed model includes TRISS-SpO2 which replaces pulse SpO2 instead of revised trauma score in the original TRISS score. Probability of survival (Ps) was calculated for both models using coefficients derived from Walker-Duncan regression analysis analyzed from the Major Trauma Outcome Study. Receiver operating characteristic curve analysis was used to predict model performance and the accuracy was calculated.
UNASSIGNED: The mortality rate in the present study was 30 (7.9%). The predictive accuracy of original TRISS score which calculated Ps based on respiratory rate was 97.11%, and for the proposed model of TRISS score which calculated Ps based on SpO2 was found 97.11%, and thus there is no significant difference in the performance.
UNASSIGNED: The new proposed model TRISS-SpO2 showed a good accuracy which is similar to original TRISS score. However, the new tool TRISS-SpO2 might be easier to use for robust performance in the clinical setting.

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.

BACKGROUND: Prognostic indices can enhance personalized predictions of health burdens. However, a simple, practical, and reproducible tool is lacking for clinical use. This study aimed to develop a machine learning-based prognostic index for predicting all-cause mortality in community-dwelling older individuals.
METHODS: We utilized the Healthy Aging Longitudinal Study in Taiwan (HALST) cohort, encompassing data from 5 663 participants. Over the 5-year follow-up, 447 deaths were confirmed. A machine learning-based routine blood examination prognostic index (MARBE-PI) was developed using common laboratory tests based on machine learning techniques. Participants were grouped into multiple risk categories by stratum-specific likelihood ratio analysis based on their MARBE-PI scores. The MARBE-PI was subsequently externally validated with an independent population-based cohort from Japan.
RESULTS: Beyond age, sex, education level, and BMI, 6 laboratory tests (low-density lipoprotein, albumin, aspartate aminotransferase, lymphocyte count, high-sensitivity C-reactive protein, and creatinine) emerged as pivotal predictors via stepwise logistic regression (LR) for 5-year mortality. The area under curves of MARBE-PI constructed by LR were 0.799 (95% confidence interval [95% CI]: 0.778-0.819) and 0.756 (95% CI: 0.694-0.814) for the internal and external validation data sets, and were 0.801 (95% CI: 0.790-0.811) and 0.809 (95% CI: 0.774-0.845) for the extended 10-year mortality in both data sets, respectively. Risk categories stratified by MARBE-PI showed a consistent dose-response association with mortality. The MARBE-PI also performed comparably with indices constructed with clinical health deficits and/or laboratory results.
CONCLUSIONS: The MARBE-PI is considered the most applicable measure for risk stratification in busy clinical settings. It holds potential to pinpoint older individuals at elevated mortality risk, thereby aiding clinical decision-making.

Colorectal cancer (CRC) ranks as the second leading cause of cancer-related deaths. This study aimed to predict survival outcomes of CRC patients using machine learning (ML) methods.
A retrospective analysis included 1853 CRC patients admitted to three prominent tertiary hospitals in Iran from October 2006 to July 2019. Six ML methods, namely logistic regression (LR), Naïve Bayes (NB), Support Vector Machine (SVM), Neural Network (NN), Decision Tree (DT), and Light Gradient Boosting Machine (LGBM), were developed with 10-fold cross-validation. Feature selection employed the Random Forest method based on mean decrease GINI criteria. Model performance was assessed using Area Under the Curve (AUC).
Time from diagnosis, age, tumor size, metastatic status, lymph node involvement, and treatment type emerged as crucial predictors of survival based on mean decrease GINI. The NB (AUC = 0.70, 95% Confidence Interval [CI] 0.65-0.75) and LGBM (AUC = 0.70, 95% CI 0.65-0.75) models achieved the highest predictive AUC values for CRC patient survival.
This study highlights the significance of variables including time from diagnosis, age, tumor size, metastatic status, lymph node involvement, and treatment type in predicting CRC survival. The NB model exhibited optimal efficacy in mortality prediction, maintaining a balanced sensitivity and specificity. Policy recommendations encompass early diagnosis and treatment initiation for CRC patients, improved data collection through digital health records and standardized protocols, support for predictive analytics integration in clinical decisions, and the inclusion of identified prognostic variables in treatment guidelines to enhance patient outcomes.

BACKGROUND: In intensive care units (ICUs), accurate mortality prediction is crucial for effective patient management and resource allocation. The Simplified Acute Physiology Score II (SAPS-2), though commonly used, relies heavily on comprehensive clinical data and blood samples. This study sought to develop an artificial intelligence (AI) model utilizing key hemodynamic parameters to predict ICU mortality within the first 24 h and assess its performance relative to SAPS-2.
METHODS: We conducted an analysis of select hemodynamic parameters and the structure of heart rate curves to identify potential predictors of ICU mortality. A machine-learning model was subsequently trained and validated on distinct patient cohorts. The AI algorithm\'s performance was then compared to the SAPS-2, focusing on classification accuracy, calibration, and generalizability.
RESULTS: The study included 1298 ICU admissions from March 27th, 2015, to March 27th, 2017. An additional cohort from 2022 to 2023 comprised 590 patients, resulting in a total dataset of 1888 patients. The observed mortality rate stood at 24.0%. Key determinants of mortality were the Glasgow Coma Scale score, heart rate complexity, patient age, duration of diastolic blood pressure below 50 mmHg, heart rate variability, and specific mean and systolic blood pressure thresholds. The AI model, informed by these determinants, exhibited a performance profile in predicting mortality that was comparable, if not superior, to the SAPS-2.
CONCLUSIONS: The AI model, which integrates heart rate and blood pressure curve analyses with basic clinical parameters, provides a methodological approach to predict in-hospital mortality in ICU patients. This model offers an alternative to existing tools that depend on extensive clinical data and laboratory inputs. Its potential integration into ICU monitoring systems may facilitate more streamlined mortality prediction processes.

OBJECTIVE: To designed a new model using pre-transplant data to predict post-transplant mortality for Chinese population and compared its performance to that of existing models.
METHODS: In this multicenter study, 544 recipients of liver transplants for non-tumor indications were enrolled in the training group and 276 patients in the validation group. The new Simplified Mortality Prediction Scores (SMOPS) model was compared to the MELD and four existing models using the C-statistic.
RESULTS: SMOPS model used 6 independent pre-transplantation risk factors screened from the training group (chronic liver failure/organ failure scores, fever > 37.6 ℃, ABO blood-type compatibility, arterial lactate level, leukocyte count and re-transplantation). The SMOPS accurately predicted patients\' 30-day, 90-day and 365-day mortality following liver transplantation, and its\' scores were more accurate than those of the other models. The SMOPS generated four levels of risk: low risk (<10 points), moderate risk (11-20 points), high risk (21-25 points) and futile risk (≥26 points). The survival within all risk levels was not different between MELD=40 and MELD<40. The survival within moderate-, high- or extreme-risk ALF was not different between ALF and non-ALF.
CONCLUSIONS: The SMOPS model uses pre-transplant risk factors to stratify post-transplant survival and is superior to current models for Chinese population, and has the potential to contribute to improvements in organ-allocation policies.