Developing a stroke alert trigger for clinical decision support at emergency triage using machine learning

Analyzing the impact of heavy metal exposure on osteoarthritis and rheumatoid arthritis: an approach based on interpretable machine learning

Introduction

This investigation leverages advanced machine learning (ML) techniques to dissect the complex relationship between heavy metal exposure and its impacts on osteoarthritis (OA) and rheumatoid arthritis (RA). Utilizing a comprehensive dataset from the National Health and Nutrition Examination Survey (NHANES) spanning from 2003 to 2020, this study aims to elucidate the roles specific heavy metals play in the incidence and differentiation of OA and RA.

Methods

Employing a phased ML strategy that encompasses a range of methodologies, including LASSO regression and SHapley Additive exPlanations (SHAP), our analytical framework integrates demographic, laboratory, and questionnaire data. Thirteen distinct ML models were applied across seven methodologies to enhance the predictability and interpretability of clinical outcomes. Each phase of model development was meticulously designed to progressively refine the algorithm's performance.

Results

The results reveal significant associations between certain heavy metals and an increased risk of arthritis. The phased ML approach enabled the precise identification of key predictors and their contributions to disease outcomes.

Discussion

These findings offer new insights into potential pathways for early detection, prevention, and management strategies for arthritis associated with environmental exposures. By improving the interpretability of ML models, this research provides a potent tool for clinicians and researchers, facilitating a deeper understanding of the environmental determinants of arthritis.

Predicting stroke events with a proactive fusion system: a comprehensive study on imbalance class handling in computational biomechanics

Stroke, as a critical global health concern and the second leading cause of death, occurs when blood flow to the brain is interrupted. Although machine learning has advanced in medical safety, there is limited research on stroke prediction using information fusion systems. This study presents a fusion framework that combines multiple base classifiers and a Meta classifier to improve stroke prediction performance. The research utilizes Grid Search optimized models, such as Random Forest, Support Vector Machine, K Nearest Neighbors, AdaBoost, Gradient Boosting, Light Gradient Boosting, Categorical Boosting, and eXtreme Gradient Boosting for in-depth stroke analysis. Since stroke events are rare and unpredictable, classification outcomes can be deceptive due to imbalanced data. This article examines stroke probability by comparing three data balancing methods: over-sampling, under-sampling, and tomek-link synthetic minority over-sampling (SMOTE-TL) to enhance prediction accuracy. The findings reveal that AdaBoost as a meta-classifier attains the highest performance in the fusion framework, with a peak of 88.09% Recall and 83.66% F1 score. This innovative approach provides crucial insights into stroke prediction and can be a valuable resource for strengthening intervention efforts in advanced healthcare systems.

Systematic Review and Meta-Analysis of Prehospital Machine Learning Scores as Screening Tools for Early Detection of Large Vessel Occlusion in Patients With Suspected Stroke

BACKGROUND

Enhanced detection of large vessel occlusion (LVO) through machine learning (ML) for acute ischemic stroke appears promising. This systematic review explored the capabilities of ML models compared with prehospital stroke scales for LVO prediction.

METHODS AND RESULTS

Six bibliographic databases were searched from inception until October 10, 2023. Meta-analyses pooled the model performance using area under the curve (AUC), sensitivity, specificity, and summary receiver operating characteristic curve. Of 1544 studies screened, 8 retrospective studies were eligible, including 32 prehospital stroke scales and 21 ML models. Of the 9 prehospital scales meta-analyzed, the Rapid Arterial Occlusion Evaluation had the highest pooled AUC (0.82 [95% CI, 0.79-0.84]). Support Vector Machine achieved the highest AUC of 9 ML models included (pooled AUC, 0.89 [95% CI, 0.88-0.89]). Six prehospital stroke scales and 10 ML models were eligible for summary receiver operating characteristic analysis. Pooled sensitivity and specificity for any prehospital stroke scale were 0.72 (95% CI, 0.68-0.75) and 0.77 (95% CI, 0.72-0.81), respectively; summary receiver operating characteristic curve AUC was 0.80 (95% CI, 0.76-0.83). Pooled sensitivity for any ML model for LVO was 0.73 (95% CI, 0.64-0.79), specificity was 0.85 (95% CI, 0.80-0.89), and summary receiver operating characteristic curve AUC was 0.87 (95% CI, 0.83-0.89).

CONCLUSIONS

Both prehospital stroke scales and ML models demonstrated varying accuracies in predicting LVO. Despite ML potential for improved LVO detection in the prehospital setting, application remains limited by the absence of prospective external validation, limited sample sizes, and lack of real-world performance data in a prehospital setting.

Usability evaluation of a glove-type wearable device for efficient biometric collection during triage

To efficiently allocate medical resources at disaster sites, medical workers perform triage to prioritize medical treatments based on the severity of the wounded or sick. In such instances, evaluators often assess the severity status of the wounded or sick quickly, but their measurements are qualitative and rely on experience. Therefore, we developed a wearable device called Medic Hand in this study to extend the functionality of a medical worker's hand so as to measure multiple biometric indicators simultaneously without increasing the number of medical devices to be carried. Medic Hand was developed to quantitatively and efficiently evaluate "perfusion" during triage. Speed is essential during triage at disaster sites, where time and effort are often spared to attach medical devices to patients, so the use of Medic Hand as a biometric measurement device is more efficient for collecting biometric information. For Medic Hand to be handy during disasters, it is essential to understand and improve upon factors that facilitate its public acceptance. To this end, this paper reports on the usability evaluation of Medic Hand through a questionnaire survey of nonmedical workers.

Explaining predictors of discharge destination assessed along the patients' acute stroke journey

INTRODUCTION

Accurate prediction of outcome destination at an early stage would help manage patients presenting with stroke. This study assessed the predictive ability of three machine learning (ML) algorithms to predict outcomes at four different stages as well as compared the predictive power of stroke scores.

METHODS

Patients presenting with acute stroke to the Canberra Hospital between 2015 and 2019 were selected retrospectively. 16 potential predictors and one target variable (discharge destination) were obtained from the notes. k-Nearest Neighbour (kNN) and two ensemble-based classification algorithms (Adaptive Boosting and Bootstrap Aggregation) were employed to predict outcomes. Predictive accuracy was assessed at each of the four stages using both overall and per-class accuracy. The contribution of each variable to the prediction outcome was evaluated by the ensemble-based algorithm and using the Relief feature selection algorithm. Various combinations of stroke scores were tested using the aforementioned models.

RESULTS

Of the three ML models, Adaptive Boosting demonstrated the highest accuracy (90%) at Stage 4 in predicting death while the highest overall accuracy (81.7%) was achieved by kNN (k=2/City-block distance). Feature importance analysis has shown that the most important features are the 24-hour Scandinavian Stroke Scale (SSS) and 24-hour National Institutes of Health Stroke Scale (NIHSS) scores, dyslipidaemia, hypertension and premorbid mRS score. For the initial and 24-hour scores, there was a higher correlation (0.93) between SSS scores than for NIHSS scores (0.81). Reducing the overall four scores to InitSSS/24hrNIHSS increased accuracy to 95% in predicting death (Adaptive Boosting) and overall accuracy to 85.4% (kNN). Accuracies at Stage 2 (pre-treatment, 11 predictors) were not far behind those at Stage 4.

CONCLUSION

Our findings suggest that even in the early stages of management, a clinically useful prediction regarding discharge destination can be made. Adaptive Boosting might be the best ML model, especially when it comes to predicting death. The predictors' importance analysis also showed that dyslipidemia and hypertension contributed to the discharge outcome even more than expected. Further, surprisingly using mixed score systems might also lead to higher prediction accuracies.

Intersectional analysis of inequalities in self-reported breast cancer screening attendance using supervised machine learning and PROGRESS-Plus framework

Background

Breast cancer is a critical public health concern in Spain, and organized screening programs have been in place since the 1990s to reduce its incidence. However, despite the bi-annual invitation for breast cancer screening (BCS) for women aged 45-69, significant attendance inequalities persist among different population groups. This study employs a quantitative intersectional perspective to identify intersectional positions at risk of not undergoing breast cancer screening in Spain.

Methods

Women were selected from the 2020 European Health Interview Survey in Spain, which surveyed the adult population (> 15 years old) living in private households (N = 22,072; 59% response rate). Inequality indicators based on the PROGRESS-Plus framework were used to disentangle existing social intersections. To identify intersectional groups, decision tree models, including classification and regression trees (CARTs), chi-squared automatic interaction detector (CHAID), conditional inference rees (CITs), and C5.0, along with an ensemble algorithm, extreme gradient boosting (XGBoost), were applied.

Results

XGBoost (AUC 78.8%) identified regional differences (Autonomous Community) as the most important factor for classifying BCS attendance, followed by education, age, and marital status. The C5.0 model (balanced accuracy 81.1%) highlighted that the relative importance of individual characteristics, such as education, marital status, or age, for attendance differs based on women's place of residence and their degree of interaction. The highest risk of not attending BCS was observed among illiterate older women in lower social classes who were born in Spain, were residing in Asturias, Cantabria, Basque Country, Castile and León, Extremadura, Galicia, Madrid, Murcia, La Rioja, or Valencian Community, and were married, divorced, or widowed. Subsequently, the risk of not attending BCS extends to three other groups of women: women living in Ceuta and Melilla; single or legally separated women living in the rest of Spain; and women not born in Spain who were married, divorced, or widowed and not residing in Ceuta or Melilla.

Conclusion

The combined use of decision trees and ensemble algorithms can be a valuable tool in identifying intersectional positions at a higher risk of not utilizing public resources and, thus, can aid substantially in developing targeted interventions to increase BCS attendance.

Artificial intelligence in emergency medicine. A systematic literature review

Motivation and objective: Emergency medicine is becoming a popular application area for artificial intelligence methods but remains less investigated than other healthcare branches. The need for time-sensitive decision-making on the basis of high data volumes makes the use of quantitative technologies inevitable. However, the specifics of healthcare regulations impose strict requirements for such applications. Published contributions cover separate parts of emergency medicine and use disparate data and algorithms. This study aims to systematize the relevant contributions, investigate the main obstacles to artificial intelligence applications in emergency medicine, and propose directions for further studies.

METHODS

The contributions selection process was conducted with systematic electronic databases querying and filtering with respect to established exclusion criteria. Among the 380 papers gathered from IEEE Xplore, ACM Digital Library, Springer Library, ScienceDirect, and Nature databases 116 were considered to be a part of the survey. The main features of the selected papers are the focus on emergency medicine and the use of machine learning or deep learning algorithms.

FINDINGS AND DISCUSSION

The selected papers were classified into two branches: diagnostics-specific and triage-specific. The former ones are focused on either diagnosis prediction or decision support. The latter covers such applications as mortality, outcome, admission prediction, condition severity estimation, and urgent care prediction. The observed contributions are highly specialized within a single disease or medical operation and often use privately collected retrospective data, making them incomparable. These and other issues can be addressed by creating an end-to-end solution based on human-machine interaction.

CONCLUSION

Artificial intelligence applications are finding their place in emergency medicine, while most of the corresponding studies remain isolated and lack higher generalization and more sophisticated methodology, which can be a matter of forthcoming improvements.

Triage body temperature and its influence on patients with acute myocardial infarction

BACKGROUND

Fever can occur after acute myocardial infarction (MI). The influence of body temperature (BT) after hospital arrival on patients with acute MI has rarely been investigated.

METHODS

Patients who were diagnosed with acute MI in the emergency department (ED) of a tertiary teaching hospital between 1 January 2020 and 31 December 2020 were enrolled. Based on the tympanic temperature obtained at the ED triage, patients were categorized into normothermic (35.5°C-37.5°C), hypothermic (< 35.5°C), or hyperthermic (> 37.5°C) groups. The primary outcome was in-hospital cardiac arrest (IHCA), while the secondary outcomes were adverse events. Statistical significance was set at p < 0.05.

RESULTS

There were 440 enrollees; significant differences were found among the normothermic (n = 369, 83.9%), hypothermic (n = 27, 6.1%), and hyperthermic (n = 44, 10.0%) groups in the triage respiratory rate (median [IQR]) (20.0 [4.0] cycles/min versus 20.0 [4.0] versus 20.0 [7.5], p = 0.009), triage heart rate (88.0 [29.0] beats/min versus 82.0 [28.0] versus 102.5 [30.5], p < 0.001), presence of ST-elevation MI (42.0% versus 66.7% versus 31.8%, p = 0.014), need for cardiac catheterization (87.3% versus 85.2% versus 72.7%, p = 0.034), initial troponin T level (165.9 [565.2] ng/L versus 49.1 [202.0] versus 318.8 [2002.0], p = 0.002), peak troponin T level (343.8 [1405.9] ng/L versus 218.7 [2318.2] versus 832.0 [2640.8], p = 0.003), length of ICU stay (2.0 [3.0] days versus 3.0 [8.0] versus 3.0 [9.5], p = 0.006), length of hospital stay (4.0 [4.5] days versus 6.0 [15.0] versus 10.5 [10.8], p < 0.001), and infection during hospitalization (19.8% versus 29.6% versus 63.6%, p < 0.001) but not in IHCA (7.6% versus 14.8% versus 11.4%, p = 0.323) or any adverse events (50.9% versus 48.1% versus 63.6%, p = 0.258). Multivariable analysis showed no significant association of triage BT with IHCA or any major complication.

CONCLUSION

Triage BT did not show a significant association with IHCA or adverse events in patients with acute MI. However, triage BT could be associated with different clinical presentations and should warrant further investigation.

Examining the predictability and prognostication of multimorbidity among older Delayed-Discharge Patients: A Machine learning analytics

BACKGROUND

Patient complexity among older delayed-discharge patients complicates discharge planning, resulting in a higher rate of adverse outcomes, such as readmission and mortality. Early prediction of multimorbidity, as a common indicator of patient complexity, can support proactive discharge planning by prioritizing complex patients and reducing healthcare inefficiencies.

OBJECTIVE

We set out to accomplish the following two objectives: 1) to examine the predictability of three common multimorbidity indices, including Charlson-Deyo Comorbidity Index (CDCI), the Elixhauser Comorbidity Index (ECI), and the Functional Comorbidity Index (FCI) using machine learning (ML), and 2) to assess the prognostic power of these indices in predicting 30-day readmission and mortality.

MATERIALS AND METHODS

We used data including 163,983 observations of patients aged 65 and older who experienced discharge delay in Ontario, Canada, during 2004 - 2017. First, we utilized various classification ML algorithms, including classification and regression trees, random forests, bagging trees, extreme gradient boosting, and logistic regression, to predict the multimorbidity status based on CDCI, ECI, and FCI. Second, we used adjusted multinomial logistic regression to assess the association between multimorbidity indices and the patient-important outcomes, including 30-day mortality and readmission.

RESULTS

For all ML algorithms and regardless of the predictive performance criteria, better predictions were established for the CDCI compared with the ECI and FCI. Remarkably, the most predictable multimorbidity index (i.e., CDCI with Area Under the Receiver Operating Characteristic Curve = 0.80, 95% CI = 0.79 - 0.81) also offered the highest prognostications regarding adverse events (RRR_mortality = 3.44, 95% CI = 3.21 - 3.68 and RRR_readmission = 1.36, 95% CI = 1.31 - 1.40).

CONCLUSIONS

Our findings highlight the feasibility and utility of predicting multimorbidity status using ML algorithms, resulting in the early detection of patients at risk of mortality and readmission. This can support proactive triage and decision-making about staffing and resource allocation, with the goal of optimizing patient outcomes and facilitating an upstream and informed discharge process through prioritizing complex patients for discharge and providing patient-centered care.