Predicting Discharge Destination of Critically Ill Patients Using Machine Learning

The Effect of Clinical Decision Support Systems on Patients, Nurses, and Work Environment in ICUs: A Systematic Review

This study aimed to examine the impact of clinical decision support systems on patient outcomes, working environment outcomes, and decision-making processes in nursing. The authors conducted a systematic literature review to obtain evidence on studies about clinical decision support systems and the practices of ICU nurses. For this purpose, the authors searched 10 electronic databases, including PubMed, CINAHL, Web of Science, Scopus, Cochrane Library, Ovid MEDLINE, Science Direct, Tr-Dizin, Harman, and DergiPark. Search terms included "clinical decision support systems," "decision making," "intensive care," "nurse/nursing," "patient outcome," and "working environment" to identify relevant studies published during the period from the year 2007 to October 2022. Our search yielded 619 articles, of which 39 met the inclusion criteria. A higher percentage of studies compared with others were descriptive (20%), conducted through a qualitative (18%), and carried out in the United States (41%). According to the results of the narrative analysis, the authors identified three main themes: "patient care outcomes," "work environment outcomes," and the "decision-making process in nursing." Clinical decision support systems, which target practices of ICU nurses and patient care outcomes, have positive effects on outcomes and show promise in improving the quality of care; however, available studies are limited.

Improved patient mortality predictions in emergency departments with deep learning data-synthesis and ensemble models

The triage process in emergency departments (EDs) relies on the subjective assessment of medical practitioners, making it unreliable in certain aspects. There is a need for a more accurate and objective algorithm to determine the urgency of patients. This paper explores the application of advanced data-synthesis algorithms, machine learning (ML) algorithms, and ensemble models to predict patient mortality. Patients predicted to be at risk of mortality are in a highly critical condition, signifying an urgent need for immediate medical intervention. This paper aims to determine the most effective method for predicting mortality by enhancing the F1 score while maintaining high area under the receiver operating characteristic curve (AUC) score. This study used a dataset of 7325 patients who visited the Yonsei Severance Hospital's ED, located in Seoul, South Korea. The patients were divided into two groups: patients who deceased in the ED and patients who didn't. Various data-synthesis techniques, such as SMOTE, ADASYN, CTGAN, TVAE, CopulaGAN, and Gaussian Copula, were deployed to generate synthetic patient data. Twenty two ML models were then utilized, including tree-based algorithms like Decision tree, AdaBoost, LightGBM, CatBoost, XGBoost, NGBoost, TabNet, which are deep neural network algorithms, and statistical algorithms such as Support Vector Machine, Logistic Regression, Random Forest, k-nearest neighbors, and Gaussian Naive Bayes, as well as Ensemble Models which use the results from the ML models. Based on 21 patient information features used in the pandemic influenza triage algorithm (PITA), the models explained previously were applied to aim for the prediction of patient mortality. In evaluating ML algorithms using an imbalanced medical dataset, conventional metrics like accuracy scores or AUC can be misleading. This paper emphasizes the importance of using the F1 score as the primary performance measure, focusing on recall and specificity in detecting patient mortality. The highest-ranked model for predicting mortality utilized the Gaussian Copula data-synthesis technique and the CatBoost classifier, achieving an AUC of 0.9731 and an F1 score of 0.7059. These findings highlight the effectiveness of machine learning algorithms and data-synthesis techniques in improving the prediction performance of mortality in EDs.

Severity-of-Illness Scores and Discharge Disposition in Patients Admitted to Long-Term Acute Care Hospitals

BACKGROUND

After an intensive care unit (ICU) admission, nearly 20% of survivors of chronic critical illness require admission to a long-term acute care hospital (LTACH) for continued subspecialty care. The effect of the burden of medical comorbidities on discharge disposition after LTACH admission remains unclear.

METHODS

A retrospective cohort study was performed involving patients with chronic critical illness who were discharged from the medical ICU and admitted to an LTACH between 2016 and 2018. The patients' Acute Physiology and Chronic Health Evaluation II (APACHE II), Sequential Organ Failure Assessment (SOFA), Nutrition Risk in the Critically Ill (NUTRIC), and Charlson Comorbidity Index (CCI) scores at the time of LTACH admission were calculated from electronic medical records. The mean scores on each instrument were compared by discharge disposition.

RESULTS

A total of 156 patients were admitted to the LTACH from the medical ICU between 2016 and 2018. They had a mean (SD) age of 61.5 (13.3) years, a mean (SD) body mass index of 28.1 (8.3), a median (IQR) ICU stay of 16.3 (1-108) days, and a median (IQR) LTACH stay of 38.2 (1-227) days. Patients who were discharged home had lower mean (SD) APACHE II (14.6 [5.0] vs 18.2 [5.4], P = .01), SOFA (3.3 [2.1] vs 4.6 [2.1], P = .03), NUTRIC (3.3 [1.4] vs 4.6 [1.4], P = .001), and CCI (4.3 [2.5] vs 6.1 [2.8], P = .02) scores on admission to the LTACH than those who were not discharged home.

CONCLUSION

Severity-of-illness scores on admission to an LTACH can be used to predict patients' likelihood of being discharged home.

Unlocking the Power of EHRs: Harnessing Unstructured Data for Machine Learning-based Outcome Predictions

The integration of Electronic Health Records (EHRs) with Machine Learning (ML) models has become imperative in examining patient outcomes due to the vast amounts of clinical data they provide. However, critical information regarding social and behavioral factors that affect health, such as social isolation, stress, and mental health complexities, is often recorded in unstructured clinical notes, hindering its accessibility. This has resulted in an over-reliance on clinical data in current EHR-based research, potentially leading to disparities in health outcomes. This study aims to evaluate the impact of incorporating patient-specific context from unstructured EHR data on the accuracy and stability of ML algorithms for predicting mortality, using the MIMIC III database. Results from the study confirmed the significance of incorporating patient-specific information into prediction models, leading to a notable improvement in the discriminatory power and robustness of the ML algorithms. Furthermore, the findings underline the importance of considering non-clinical factors related to a patient's daily life, in addition to clinical factors, when making predictions about patient outcomes. The advent of advanced generative models, such as GPT-4, presents new opportunities for effectively extracting social and behavioral factors from unstructured clinical notes, further enhancing the accuracy and stability of ML algorithms in predicting patient outcomes. The results of our study have significant ramifications for improving ML in clinical decision support and patient outcome predictions, specifically highlighting the potential role of generative models like GPT-4 in advancing ML-based outcome predictions.

Cloud computing and machine learning for analysis of health care data based on neuro fuzzy logistic regression

Healthcare data is the most sensitive information for processing through machine learning and cloud computing in the various healthcare organizations. Electronic Health Record (EHR) manipulation are now on the rise, and we need to focus on using the data generated by the healthcare applications. Many sensitive data are associated with various health care domains, particularly neurology and cardiology. Previous approaches, such as manual data records, had significant disadvantages, and hence disease prediction based on the above records was found ineffective resulting with improper diagnosis on the patients. These data records require special attention, and current frameworks focused on these areas must implement sophisticated technologies to predict specific patterns. To address the above concerns, the proposed work incorporates the integration of Neuro Fuzzy Logistic Regression (NFLR) machine learning algorithm and cloud computing storage management to solve these problems. The usage of cloud storage reduces data duplication while handling the storage of EHRs where the proposed ML algorithm accurately predict the disease. In the proposed research, the features are extracted using a specific algorithm –Self-organizing Clustering (SOC) which forms a clustered data with highest weight. To select the maximum number of features, and to predict the disease risk factors, the S2NO algorithm and NFLR algorithms are used in this work. Further, the database storage estimation with fuzzy rules, logistic analysis, and other benefits such as experimental learning of different ML tools, data privacy constraints related to healthcare are considered in this paper.

Early prediction of patient discharge disposition in acute neurological care using machine learning

BACKGROUND

Acute neurological complications are some of the leading causes of death and disability in the U.S. The medical professionals that treat patients in this setting are tasked with deciding where (e.g., home or facility), how, and when to discharge these patients. It is important to be able to predict potential patient discharge outcomes as early as possible during the patient's hospital stay and to know what factors influence the development of discharge planning. This study carried out two parallel experiments: A multi-class outcome (patient discharge targets of 'home', 'nursing facility', 'rehab', 'death') and binary class outcome ('home' vs. 'non-home'). The goal of this study is to develop early predictive models for each experiment exploring which patient characteristics and clinical variables significantly influence discharge planning of patients based on the data that are available only within 24 h of their hospital admission.  METHOD: Our methodology centers around building and training five different machine learning models followed by testing and tuning those models to find the best-suited predictor for each experiment with a dataset of 5,245 adult patients with neurological conditions taken from the eICU-CRD database.

RESULTS

The results of this study show XGBoost to be the most effective model for predicting between four common discharge outcomes of 'home', 'nursing facility', 'rehab', and 'death', with 71% average c-statistic. The XGBoost model was also the best-performer in the binary outcome experiment with a c-statistic of 76%. This article also explores the accuracy, reliability, and interpretability of the best performing models in each experiment by identifying and analyzing the features that are most impactful to the predictions.

CONCLUSIONS

The acceptable accuracy and interpretability of the predictive models based on early admission data suggests that the models can be used in a suggestive context to help guide healthcare providers in efforts of planning effective and equitable discharge recommendations.

An Interpretable Machine Learning Approach to Predict Fall Risk Among Community-Dwelling Older Adults: a Three-Year Longitudinal Study

BACKGROUND

Adverse health effects resulting from falls are a major public health concern. Although studies have identified risk factors for falls, none have examined long-term prediction of fall risk. Furthermore, recent evidence suggests that there are additional risk factors, such as psychosocial factors.

OBJECTIVE

In this 3-year longitudinal study, we evaluated a predictive model for risk of fall among community-dwelling older adults using machine learning methods.

DESIGN

A 3-year follow-up prospective longitudinal study (from 2010 to 2013).

SETTING

Twenty-four municipalities in nine of the 47 prefectures (provinces) of Japan.

PARTICIPANTS

Community-dwelling individuals aged ≥65 years who were functionally independent at baseline (n = 61,883).

METHODS

The baseline survey was conducted from August 2010 to January 2012, and the follow-up survey was conducted from October to December 2013. Both surveys were conducted involving self-reported questionnaires. The measured outcome at the follow-up survey was self-reported multiple falls during the previous year. The 142 variables included in the baseline survey were regarded as candidate predictors. The random-forest-based Boruta algorithm was used to select predictors, and the eXtreme Gradient Boosting algorithm with 10 repetitions of nested k-fold cross-validation was used for modeling and model evaluation. Furthermore, we used shapley additive explanations to gain insight into the behavior of the prediction model.

KEY RESULTS

Fourteen out of 142 candidate features were selected as predictors. Among these predictors, experience of falling as of the baseline survey was the most important feature, followed by self-rated health and age. Moreover, sense of coherence was newly identified as a risk factor for falls.

CONCLUSIONS

This study suggests that machine learning tools can be adapted to explore new associative factors, make accurate predictions, and provide actionable insights for fall prevention strategies.

Using Bayesian Optimization and Wavelet Decomposition in GPU for Arterial Blood Pressure Estimation

Continuous monitoring of arterial blood pressure (ABP) of patients in hospital is currently carried out in an invasive way, which could represent a risk for them. In this paper, a noninvasive methodology to optimize ABP estimators using electrocardiogram and photoplethysmography signals is proposed. For this, the XGBoost machine learning model, optimized with Bayesian techniques, is executed in a Graphics Processing Unit, which drastically reduces execution time. The methodology is evaluated using the MIMIC-III Waveform Database. Systolic and diastolic pressures are estimated with mean absolute error values of 15.85 and 11.59 mmHg, respectively, similar to those of the state of the art. The main advantage of the proposed methodology with respect to others of the current state of the art is that it allows the optimization of the estimator model to be performed automatically and more efficiently at the computational level for the data available. Clinical Relevance- This approach has the advantage of using noninvasive methods to continuously monitor patient's arterial blood pressure, reducing the risk for patients.

A Machine Learning Based Discharge Prediction of Cardiovascular Diseases Patients in Intensive Care Units

This paper targets a major challenge of how to effectively allocate medical resources in intensive care units (ICUs). We trained multiple regression models using the Medical Information Mart for Intensive Care III (MIMIC III) database recorded in the period between 2001 and 2012. The training and validation dataset included pneumonia, sepsis, congestive heart failure, hypotension, chest pain, coronary artery disease, fever, respiratory failure, acute coronary syndrome, shortness of breath, seizure and transient ischemic attack, and aortic stenosis patients’ recorded data. Then we tested the models on the unseen data of patients diagnosed with coronary artery disease, congestive heart failure or acute coronary syndrome. We included the admission characteristics, clinical prescriptions, physiological measurements, and discharge characteristics of those patients. We assessed the models’ performance using mean residuals and running times as metrics. We ran multiple experiments to study the data partition’s impact on the learning phase. The total running time of our best-evaluated model is 123,450.9 mS. The best model gives an average accuracy of 98%, highlighting the location of discharge, initial diagnosis, location of admission, drug therapy, length of stay and internal transfers as the most influencing patterns to decide a patient’s readiness for discharge.

A heterogeneous ensemble learning method for neuroblastoma survival prediction

Neuroblastoma is a pediatric cancer with high morbidity and mortality. Accurate survival prediction of patients with neuroblastoma plays an important role in the formulation of treatment plans. In this study, we proposed a heterogeneous ensemble learning method to predict the survival of neuroblastoma patients and extract decision rules from the proposed method to assist doctors in making decisions. After data preprocessing, five heterogeneous base learners were developed, which consisted of decision tree, random forest, support vector machine based on genetic algorithm, extreme gradient boosting and light gradient boosting machine. Subsequently, a heterogeneous feature selection method was devised to obtain the optimal feature subset of each base learner, and the optimal feature subset of each base learner guided the construction of the base learners as a priori knowledge. Furthermore, an area under curve-based ensemble mechanism was proposed to integrate the five heterogeneous base learners. Finally, the proposed method was compared with mainstream machine learning methods from different indicators, and valuable information was extracted by using the partial dependency plot analysis method and rule-extracted method from the proposed method. Experimental results show that the proposed method achieves an accuracy of 91.64%, recall of 91.14%, and AUC of 91.35% and is significantly better than the mainstream machine learning methods. In addition, interpretable rules with accuracy higher than 0.900 and predicted responses are extracted from the proposed method. Our study can effectively improve the performance of the clinical decision support system to improve the survival of neuroblastoma patients.