Risk prediction of delirium in hospitalized patients using machine learning: An implementation and prospective evaluation study

Clinical evaluation of a machine learning-based dysphagia risk prediction tool

PURPOSE

The rise of digitization promotes the development of screening and decision support tools. We sought to validate the results from a machine learning based dysphagia risk prediction tool with clinical evaluation.

METHODS

149 inpatients in the ENT department were evaluated in real time by the risk prediction tool, as well as clinically over a 3-week period. Patients were classified by both as patients at risk/no risk.

RESULTS

The AUROC, reflecting the discrimination capability of the algorithm, was 0.97. The accuracy achieved 92.6% given an excellent specificity as well as sensitivity of 98% and 82.4% resp. Higher age, as well as male sex and the diagnosis of oropharyngeal malignancies were found more often in patients at risk of dysphagia.

CONCLUSION

The proposed dysphagia risk prediction tool proved to have an outstanding performance in discriminating risk from no risk patients in a prospective clinical setting. It is likely to be particularly useful in settings where there is a lower incidence of patients with dysphagia and less awareness among staff.

Characterizing medical patients with delirium: A cohort study comparing ICD-10 codes and a validated chart review method

BACKGROUND

Delirium is a major cause of preventable mortality and morbidity in hospitalized adults, but accurately determining rates of delirium remains a challenge.

OBJECTIVE

To characterize and compare medical inpatients identified as having delirium using two common methods, administrative data and retrospective chart review.

METHODS

We conducted a retrospective study of 3881 randomly selected internal medicine hospital admissions from six acute care hospitals in Toronto and Mississauga, Ontario, Canada. Delirium status was determined using ICD-10-CA codes from hospital administrative data and through a previously validated chart review method. Baseline sociodemographic and clinical characteristics, processes of care and outcomes were compared across those without delirium in hospital and those with delirium as determined by administrative data and chart review.

RESULTS

Delirium was identified in 6.3% of admissions by ICD-10-CA codes compared to 25.7% by chart review. Using chart review as the reference standard, ICD-10-CA codes for delirium had sensitivity 24.1% (95%CI: 21.5-26.8%), specificity 99.8% (95%CI: 99.5-99.9%), positive predictive value 97.6% (95%CI: 94.6-98.9%), and negative predictive value 79.2% (95%CI: 78.6-79.7%). Age over 80, male gender, and Charlson comorbidity index greater than 2 were associated with misclassification of delirium. Inpatient mortality and median costs of care were greater in patients determined to have delirium by ICD-10-CA codes (5.8% greater mortality, 95% CI: 2.0-9.5 and $6824 greater cost, 95%CI: 4713-9264) and by chart review (11.9% greater mortality, 95%CI: 9.5-14.2% and $4967 greater cost, 95%CI: 4415-5701), compared to patients without delirium.

CONCLUSIONS

Administrative data are specific but highly insensitive, missing most cases of delirium in hospital. Mortality and costs of care were greater for both the delirium cases that were detected and missed by administrative data. Better methods of routinely measuring delirium in hospital are needed.

The validity of electronic health data for measuring smoking status: a systematic review and meta-analysis

BACKGROUND

Smoking is a risk factor for many chronic diseases. Multiple smoking status ascertainment algorithms have been developed for population-based electronic health databases such as administrative databases and electronic medical records (EMRs). Evidence syntheses of algorithm validation studies have often focused on chronic diseases rather than risk factors. We conducted a systematic review and meta-analysis of smoking status ascertainment algorithms to describe the characteristics and validity of these algorithms.

METHODS

The Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines were followed. We searched articles published from 1990 to 2022 in EMBASE, MEDLINE, Scopus, and Web of Science with key terms such as validity, administrative data, electronic health records, smoking, and tobacco use. The extracted information, including article characteristics, algorithm characteristics, and validity measures, was descriptively analyzed. Sources of heterogeneity in validity measures were estimated using a meta-regression model. Risk of bias (ROB) in the reviewed articles was assessed using the Quality Assessment of Diagnostic Accuracy Studies-2 tool.

RESULTS

The initial search yielded 2086 articles; 57 were selected for review and 116 algorithms were identified. Almost three-quarters (71.6%) of algorithms were based on EMR data. The algorithms were primarily constructed using diagnosis codes for smoking-related conditions, although prescription medication codes for smoking treatments were also adopted. About half of the algorithms were developed using machine-learning models. The pooled estimates of positive predictive value, sensitivity, and specificity were 0.843, 0.672, and 0.918 respectively. Algorithm sensitivity and specificity were highly variable and ranged from 3 to 100% and 36 to 100%, respectively. Model-based algorithms had significantly greater sensitivity (p = 0.006) than rule-based algorithms. Algorithms for EMR data had higher sensitivity than algorithms for administrative data (p = 0.001). The ROB was low in most of the articles (76.3%) that underwent the assessment.

CONCLUSIONS

Multiple algorithms using different data sources and methods have been proposed to ascertain smoking status in electronic health data. Many algorithms had low sensitivity and positive predictive value, but the data source influenced their validity. Algorithms based on machine-learning models for multiple linked data sources have improved validity.

Postoperative delirium prediction after cardiac surgery using machine learning models

OBJECTIVE

Postoperative delirium (POD) is a common postoperative complication in elderly patients, especially those undergoing cardiac surgery, which seriously affects the short- and long-term prognosis of patients. Early identification of risk factors for the development of POD can help improve the perioperative management of surgical patients. In the present study, five machine learning models were developed to predict patients at high risk of delirium after cardiac surgery and their performance was compared.

METHODS

A total of 367 patients who underwent cardiac surgery were retrospectively included in this study. Using single-factor analysis, 21 risk factors for POD were selected for inclusion in machine learning. The dataset was divided using 10-fold cross-validation for model training and testing. Five machine learning models (random forest (RF), support vector machine (SVM), radial based kernel neural network (RBFNN), K-nearest neighbour (KNN), and Kernel ridge regression (KRR)) were compared using area under the receiver operating characteristic curve (AUC-ROC), accuracy (ACC), sensitivity (SN), specificity (SPE), and Matthews coefficient (MCC).

RESULTS

Among 367 patients, 105 patients developed POD, the incidence of delirium was 28.6 %. Among the five ML models, RF had the best performance in ACC (87.99 %), SN (69.27 %), SPE (95.38 %), MCC (70.00 %) and AUC (0.9202), which was far superior to the other four models.

CONCLUSION

Delirium is common in patients after cardiac surgery. This analysis confirms the importance of the computational ML models in predicting the occurrence of delirium after cardiac surgery, especially the outstanding performance of the RF model, which has practical clinical applications for early identification of patients at risk of developing POD.

Predicting delirium and the effects of medications in hospitalized COVID-19 patients using machine learning: A retrospective study within the Korean Multidisciplinary Cohort for Delirium Prevention (KoMCoDe)

Objective

Delirium is commonly reported from the inpatients with Coronavirus disease 2019 (COVID-19) infection. As delirium is closely associated with adverse clinical outcomes, prediction and prevention of delirium is critical. We developed a machine learning (ML) model to predict delirium in hospitalized patients with COVID-19 and to identify modifiable factors to prevent delirium.

Methods

The data set (n = 878) from four medical centers was constructed. Total of 78 predictors were included such as demographic characteristics, vital signs, laboratory results and medication, and the primary outcome was delirium occurrence during hospitalization. For analysis, the extreme gradient boosting (XGBoost) algorithm was applied, and the most influential factors were selected by recursive feature elimination. Among the indicators of performance for ML model, the area under the curve of the receiver operating characteristic (AUROC) curve was selected as the evaluation metric.

Results

Regarding the performance of developed delirium prediction model, the accuracy, precision, recall, F1 score, and the AUROC were calculated (0.944, 0.581, 0.421, 0.485, 0.873, respectively). The influential factors of delirium in this model included were mechanical ventilation, medication (antipsychotics, sedatives, ambroxol, piperacillin/tazobactam, acetaminophen, ceftriaxone, and propacetamol), and sodium ion concentration (all p < 0.05).

Conclusions

We developed and internally validated an ML model to predict delirium in COVID-19 inpatients. The model identified modifiable factors associated with the development of delirium and could be clinically useful for the prediction and prevention of delirium in COVID-19 inpatients.

Evaluation of crowdsourced mortality prediction models as a framework for assessing artificial intelligence in medicine

OBJECTIVE

Applications of machine learning in healthcare are of high interest and have the potential to improve patient care. Yet, the real-world accuracy of these models in clinical practice and on different patient subpopulations remains unclear. To address these important questions, we hosted a community challenge to evaluate methods that predict healthcare outcomes. We focused on the prediction of all-cause mortality as the community challenge question.

MATERIALS AND METHODS

Using a Model-to-Data framework, 345 registered participants, coalescing into 25 independent teams, spread over 3 continents and 10 countries, generated 25 accurate models all trained on a dataset of over 1.1 million patients and evaluated on patients prospectively collected over a 1-year observation of a large health system.

RESULTS

The top performing team achieved a final area under the receiver operator curve of 0.947 (95% CI, 0.942-0.951) and an area under the precision-recall curve of 0.487 (95% CI, 0.458-0.499) on a prospectively collected patient cohort.

DISCUSSION

Post hoc analysis after the challenge revealed that models differ in accuracy on subpopulations, delineated by race or gender, even when they are trained on the same data.

CONCLUSION

This is the largest community challenge focused on the evaluation of state-of-the-art machine learning methods in a healthcare system performed to date, revealing both opportunities and pitfalls of clinical AI.

Effects of machine learning-based clinical decision support systems on decision-making, care delivery, and patient outcomes: a scoping review

OBJECTIVE

This study aims to summarize the research literature evaluating machine learning (ML)-based clinical decision support (CDS) systems in healthcare settings.

MATERIALS AND METHODS

We conducted a review in accordance with the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta Analyses extension for Scoping Review). Four databases, including PubMed, Medline, Embase, and Scopus were searched for studies published from January 2016 to April 2021 evaluating the use of ML-based CDS in clinical settings. We extracted the study design, care setting, clinical task, CDS task, and ML method. The level of CDS autonomy was examined using a previously published 3-level classification based on the division of clinical tasks between the clinician and CDS; effects on decision-making, care delivery, and patient outcomes were summarized.

RESULTS

Thirty-two studies evaluating the use of ML-based CDS in clinical settings were identified. All were undertaken in developed countries and largely in secondary and tertiary care settings. The most common clinical tasks supported by ML-based CDS were image recognition and interpretation (n = 12) and risk assessment (n = 9). The majority of studies examined assistive CDS (n = 23) which required clinicians to confirm or approve CDS recommendations for risk assessment in sepsis and for interpreting cancerous lesions in colonoscopy. Effects on decision-making, care delivery, and patient outcomes were mixed.

CONCLUSION

ML-based CDS are being evaluated in many clinical areas. There remain many opportunities to apply and evaluate effects of ML-based CDS on decision-making, care delivery, and patient outcomes, particularly in resource-constrained settings.

Clinical Prediction Models for Hospital-Induced Delirium Using Structured and Unstructured Electronic Health Record Data: Protocol for a Development and Validation Study

BACKGROUND

Hospital-induced delirium is one of the most common and costly iatrogenic conditions, and its incidence is predicted to increase as the population of the United States ages. An academic and clinical interdisciplinary systems approach is needed to reduce the frequency and impact of hospital-induced delirium.

OBJECTIVE

The long-term goal of our research is to enhance the safety of hospitalized older adults by reducing iatrogenic conditions through an effective learning health system. In this study, we will develop models for predicting hospital-induced delirium. In order to accomplish this objective, we will create a computable phenotype for our outcome (hospital-induced delirium), design an expert-based traditional logistic regression model, leverage machine learning techniques to generate a model using structured data, and use machine learning and natural language processing to produce an integrated model with components from both structured data and text data.

METHODS

This study will explore text-based data, such as nursing notes, to improve the predictive capability of prognostic models for hospital-induced delirium. By using supervised and unsupervised text mining in addition to structured data, we will examine multiple types of information in electronic health record data to predict medical-surgical patient risk of developing delirium. Development and validation will be compliant to the Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis (TRIPOD) statement.

RESULTS

Work on this project will take place through March 2024. For this study, we will use data from approximately 332,230 encounters that occurred between January 2012 to May 2021. Findings from this project will be disseminated at scientific conferences and in peer-reviewed journals.

CONCLUSIONS

Success in this study will yield a durable, high-performing research-data infrastructure that will process, extract, and analyze clinical text data in near real time. This model has the potential to be integrated into the electronic health record and provide point-of-care decision support to prevent harm and improve quality of care.

INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID)

DERR1-10.2196/48521.

Risk of bias in prognostic models of hospital-induced delirium for medical-surgical units: A systematic review

PURPOSE

The purpose of this systematic review was to assess risk of bias in existing prognostic models of hospital-induced delirium for medical-surgical units.

METHODS

APA PsycInfo, CINAHL, MEDLINE, and Web of Science Core Collection were searched on July 8, 2022, to identify original studies which developed and validated prognostic models of hospital-induced delirium for adult patients who were hospitalized in medical-surgical units. The Checklist for Critical Appraisal and Data Extraction for Systematic Reviews of Prediction Modelling Studies was used for data extraction. The Prediction Model Risk of Bias Assessment Tool was used to assess risk of bias. Risk of bias was assessed across four domains: participants, predictors, outcome, and analysis.

RESULTS

Thirteen studies were included in the qualitative synthesis, including ten model development and validation studies and three model validation only studies. The methods in all of the studies were rated to be at high overall risk of bias. The methods of statistical analysis were the greatest source of bias. External validity of models in the included studies was tested at low levels of transportability.

CONCLUSIONS

Our findings highlight the ongoing scientific challenge of developing a valid prognostic model of hospital-induced delirium for medical-surgical units to tailor preventive interventions to patients who are at high risk of this iatrogenic condition. With limited knowledge about generalizable prognosis of hospital-induced delirium in medical-surgical units, existing prognostic models should be used with caution when creating clinical practice policies. Future research protocols must include robust study designs which take into account the perspectives of clinicians to identify and validate risk factors of hospital-induced delirium for accurate and generalizable prognosis in medical-surgical units.

Comparing risk prediction models aimed at predicting hospitalizations for adverse drug events in community dwelling older adults: a protocol paper

Background

The objective of this paper is to describe the creation, validation, and comparison of two risk prediction modeling approaches for community-dwelling older adults to identify individuals at highest risk for adverse drug event-related hospitalizations. One approach will use traditional statistical methods, the second will use a machine learning approach.

Methods

We will construct medication, clinical, health care utilization, and other variables known to be associated with adverse drug event-related hospitalizations. To create the cohort, we will include older adults (≥ 65 years of age) empaneled to a primary care physician within the Cedars-Sinai Health System primary care clinics with polypharmacy (≥ 5 medications) or at least 1 medication commonly implicated in ADEs (certain oral hypoglycemics, anti-coagulants, anti-platelets, and insulins). We will use a Fine-Gray Cox proportional hazards model for one risk modeling approach and DataRobot, a data science and analytics platform, to run and compare several widely used supervised machine learning algorithms, including Random Forest, Support Vector Machine, Extreme Gradient Boosting (XGBoost), Decision Tree, Naïve Bayes, and K-Nearest Neighbors. We will use a variety of metrics to compare model performance and to assess the risk of algorithmic bias.

Discussion

In conclusion, we hope to develop a pragmatic model that can be implemented in the primary care setting to risk stratify older adults to further optimize medication management.

Evaluation of a Machine Learning-Based Dysphagia Prediction Tool in Clinical Routine: A Prospective Observational Cohort Study

Based on a large number of pre-existing documented electronic health records (EHR), we developed a machine learning (ML) algorithm for detection of dysphagia and aspiration pneumonia. The aim of our study was to prospectively apply this algorithm in two large patient cohorts. The tool was integrated in the hospital information system of a secondary care hospital in Austria. Based on existing data such as diagnoses, laboratory, and medication, dysphagia risk was predicted automatically, and patients were stratified into three risk groups. Patients' risk groups and risk factors were visualized in a web application. Prospective predictions of 1270 admissions to geriatric or internal medicine departments were compared with the occurrence of dysphagia or aspiration pneumonia of routinely documented events. The discriminative performance for internal medicine patients (n = 885) was excellent with an AUROC of 0.841, a sensitivity of 74.2%, and a specificity of 84.1%. For the smaller geriatric cohort (n = 221), the AUROC was 0.758, sensitivity 44.4%, and specificity 93.0%. For both cohorts, calibration plots showed a slight overestimation of the risk. This is the first study to evaluate the performance of a ML-based prediction tool for dysphagia in a prospective clinical setting. Future studies should validate the predictions on data of systematic dysphagia screening by specialists and evaluate user satisfaction and acceptance. The ML-based dysphagia prediction tool achieved an excellent performance in the internal medicine cohort. More data are needed to determine the performance in geriatric patients.

Integrating machine learning predictions for perioperative risk management: Towards an empirical design of a flexible-standardized risk assessment tool

BACKGROUND

Surgical patients are complex, vulnerable, and prone to postoperative complications that can potentially be mitigated with quality perioperative risk assessment and management. Several institutions have incorporated machine learning (ML) into their patient care to improve awareness and support clinician decision-making along the perioperative spectrum. Recent research suggests that ML risk prediction can support perioperative patient risk monitoring and management across several situations, including the operating room (OR) to intensive care unit (ICU) handoffs.

OBJECTIVES

Our study objectives were threefold: (1) evaluate whether ML-generated postoperative predictions are concordant with clinician-generated risk rankings for acute kidney injury, delirium, pneumonia, deep vein thrombosis, and pulmonary embolism, and establish their associated risk factors; (2) ascertain clinician end-user suggestions to improve adoption of ML-generated risks and their integration into the perioperative workflow; and (3) develop a user-friendly visualization format for a tool to display ML-generated risks and risk factors to support postoperative care planning, for example, within the context of OR-ICU handoffs.

METHODS

Graphical user interfaces for postoperative risk prediction models were assessed for end-user usability through cognitive walkthroughs and interviews with anesthesiologists, surgeons, certified registered nurse anesthetists, registered nurses, and critical care physicians. Thematic analysis relying on an explanation design framework was used to identify feedback and suggestions for improvement.

RESULTS

17 clinicians participated in the evaluation. ML estimates of complication risks aligned with clinicians' independent rankings, and related displays were perceived as valuable for decision-making and care planning for postoperative care. During OR-ICU handoffs, the tool could speed up report preparation and remind clinicians to address patient-specific complications, thus providing more tailored care information. Suggestions for improvement centered on electronic tool delivery; methods to build trust in ML models; modifiable risks and risk mitigation strategies; and additional patient information based on individual preferences (e.g., surgical procedure).

CONCLUSIONS

ML estimates of postoperative complication risks can provide anticipatory guidance, potentially increasing the efficiency of care planning. We have offered an ML visualization framework for designing future ML-augmented tools and anticipate the development of tools that recommend specific actions to the user based on ML model output.

Boosting Delirium Identification Accuracy With Sentiment-Based Natural Language Processing: Mixed Methods Study

BACKGROUND

Delirium is an acute neurocognitive disorder that affects up to half of older hospitalized medical patients and can lead to dementia, longer hospital stays, increased health costs, and death. Although delirium can be prevented and treated, it is difficult to identify and predict.

OBJECTIVE

This study aimed to improve machine learning models that retrospectively identify the presence of delirium during hospital stays (eg, to measure the effectiveness of delirium prevention interventions) by using the natural language processing (NLP) technique of sentiment analysis (in this case a feature that identifies sentiment toward, or away from, a delirium diagnosis).

METHODS

Using data from the General Medicine Inpatient Initiative, a Canadian hospital data and analytics network, a detailed manual review of medical records was conducted from nearly 4000 admissions at 6 Toronto area hospitals. Furthermore, 25.74% (994/3862) of the eligible hospital admissions were labeled as having delirium. Using the data set collected from this study, we developed machine learning models with, and without, the benefit of NLP methods applied to diagnostic imaging reports, and we asked the question "can NLP improve machine learning identification of delirium?"

RESULTS

Among the eligible 3862 hospital admissions, 994 (25.74%) admissions were labeled as having delirium. Identification and calibration of the models were satisfactory. The accuracy and area under the receiver operating characteristic curve of the main model with NLP in the independent testing data set were 0.807 and 0.930, respectively. The accuracy and area under the receiver operating characteristic curve of the main model without NLP in the independent testing data set were 0.811 and 0.869, respectively. Model performance was also found to be stable over the 5-year period used in the experiment, with identification for a likely future holdout test set being no worse than identification for retrospective holdout test sets.

CONCLUSIONS

Our machine learning model that included NLP (ie, sentiment analysis in medical image description text mining) produced valid identification of delirium with the sentiment analysis, providing significant additional benefit over the model without NLP.

Delirium screening in an acute care setting with a machine learning classifier based on routinely collected nursing data: A model development study

Delirium screening in acute care settings is a resource intensive process with frequent deviations from screening protocols. A predictive model relying only on daily collected nursing data for delirium screening could expand the populations covered by such screening programs. Here, we present the results of the development and validation of a series of machine-learning based delirium prediction models. For this purpose, we used data of all patients 18 years or older which were hospitalized for more than a day between January 1, 2014, and December 31, 2018, at a single tertiary teaching hospital in Zurich, Switzerland. A total of 48,840 patients met inclusion criteria. 18,873 (38.6%) were excluded due to missing data. Mean age (SD) of the included 29,967 patients was 71.1 (12.2) years and 12,231 (40.8%) were women. Delirium was assessed with the Delirium Observation Scale (DOS) with a total score of 3 or greater indicating that a patient is at risk for delirium. Additional measures included structured data collected for nursing process planning and demographic characteristics. The performance of the machine learning models was assessed using the area under the receiver operating characteristic curve (AUC). The training set consisted of 21,147 patients (mean age 71.1 (12.1) years; 8,630 (40.8%) women|) including 233,024 observations with 16,167 (6.9%) positive DOS screens. The test set comprised 8,820 patients (median age 71.1 (12.4) years; 3,601 (40.8%) women) with 91,026 observations with 5,445 (6.0%) positive DOS screens. Overall, the gradient boosting machine model performed best with an AUC of 0.933 (95% CI, 0.929 - 0.936). In conclusion, machine learning models based only on structured nursing data can reliably predict patients at risk for delirium in an acute care setting. Prediction models, using existing data collection processes, could reduce the resources required for delirium screening procedures in clinical practice.

New onset delirium prediction using machine learning and long short-term memory (LSTM) in electronic health record

OBJECTIVE

To develop and test an accurate deep learning model for predicting new onset delirium in hospitalized adult patients.

METHODS

Using electronic health record (EHR) data extracted from a large academic medical center, we developed a model combining long short-term memory (LSTM) and machine learning to predict new onset delirium and compared its performance with machine-learning-only models (logistic regression, random forest, support vector machine, neural network, and LightGBM). The labels of models were confusion assessment method (CAM) assessments. We evaluated models on a hold-out dataset. We calculated Shapley additive explanations (SHAP) measures to gauge the feature impact on the model.

RESULTS

A total of 331 489 CAM assessments with 896 features from 34 035 patients were included. The LightGBM model achieved the best performance (AUC 0.927 [0.924, 0.929] and F1 0.626 [0.618, 0.634]) among the machine learning models. When combined with the LSTM model, the final model's performance improved significantly (P = .001) with AUC 0.952 [0.950, 0.955] and F1 0.759 [0.755, 0.765]. The precision value of the combined model improved from 0.497 to 0.751 with a fixed recall of 0.8. Using the mean absolute SHAP values, we identified the top 20 features, including age, heart rate, Richmond Agitation-Sedation Scale score, Morse fall risk score, pulse, respiratory rate, and level of care.

CONCLUSION

Leveraging LSTM to capture temporal trends and combining it with the LightGBM model can significantly improve the prediction of new onset delirium, providing an algorithmic basis for the subsequent development of clinical decision support tools for proactive delirium interventions.

The development of a web-based app employing machine learning for delirium prevention in long-term care facilities in South Korea

Background

Long-term care facilities (LCFs) in South Korea have limited knowledge of and capability to care for patients with delirium. They also often lack an electronic medical record system. These barriers hinder systematic approaches to delirium monitoring and intervention. Therefore, this study aims to develop a web-based app for delirium prevention in LCFs and analyse its feasibility and usability.

Methods

The app was developed based on the validity of the AI prediction model algorithm. A total of 173 participants were selected from LCFs to participate in a study to determine the predictive risk factors for delerium.

The app was developed in five phases: (1) the identification of risk factors and preventive intervention strategies from a review of evidence-based literature, (2) the iterative design of the app and components of delirium prevention, (3) the development of a delirium prediction algorithm and cloud platform, (4) a pilot test and validation conducted with 33 patients living in a LCF, and (5) an evaluation of the usability and feasibility of the app, completed by nurses (Main users).

Results

A web-based app was developed to predict high risk of delirium and apply preventive interventions accordingly. Moreover, its validity, usability, and feasibility were confirmed after app development. By employing machine learning, the app can predict the degree of delirium risk and issue a warning alarm. Therefore, it can be used to support clinical decision-making, help initiate the assessment of delirium, and assist in applying preventive interventions.

Conclusions

This web-based app is evidence-based and can be easily mobilised to support care for patients with delirium in LCFs. This app can improve the recognition of delirium and predict the degree of delirium risk, thereby helping develop initiatives for delirium prevention and providing interventions. Moreover, this app can be extended to predict various risk factors of LCF and apply preventive interventions. Its use can ultimately improve patient safety and quality of care.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12911-022-01966-8.

Artificial intelligence in perioperative medicine - a narrative review

Recent advancements in artificial intelligence (AI) techniques have enabled the development of accurate prediction models using clinical big data. AI models for perioperative risk stratification, intraoperative event prediction, biosignal analyses, and intensive care medicine have been developed in the field of perioperative medicine. Some of these models have been validated using external datasets and randomized controlled trials. Once these models are implemented in electronic health record systems or software medical devices, they could help anesthesiologists improve clinical outcomes by accurately predicting complications and suggesting optimal treatment strategies in real-time. This review provides an overview of the AI techniques used in perioperative medicine and a summary of the studies that have been published using these techniques. Understanding these techniques will aid in their appropriate application in clinical practice.

Evaluating live performance of machine learning based prediction models for different clinical risks: a study of live systems in different hospitals (Preprint)

Background

Machine learning algorithms are currently used in a wide array of clinical domains to produce models that can predict clinical risk events. Most models are developed and evaluated with retrospective data, very few are evaluated in a clinical workflow, and even fewer report performances in different hospitals. In this study, we provide detailed evaluations of clinical risk prediction models in live clinical workflows for three different use cases in three different hospitals.

Objective

The main objective of this study was to evaluate clinical risk prediction models in live clinical workflows and compare their performance in these setting with their performance when using retrospective data. We also aimed at generalizing the results by applying our investigation to three different use cases in three different hospitals.

Methods

We trained clinical risk prediction models for three use cases (ie, delirium, sepsis, and acute kidney injury) in three different hospitals with retrospective data. We used machine learning and, specifically, deep learning to train models that were based on the Transformer model. The models were trained using a calibration tool that is common for all hospitals and use cases. The models had a common design but were calibrated using each hospital’s specific data. The models were deployed in these three hospitals and used in daily clinical practice. The predictions made by these models were logged and correlated with the diagnosis at discharge. We compared their performance with evaluations on retrospective data and conducted cross-hospital evaluations.

Results

The performance of the prediction models with data from live clinical workflows was similar to the performance with retrospective data. The average value of the area under the receiver operating characteristic curve (AUROC) decreased slightly by 0.6 percentage points (from 94.8% to 94.2% at discharge). The cross-hospital evaluations exhibited severely reduced performance: the average AUROC decreased by 8 percentage points (from 94.2% to 86.3% at discharge), which indicates the importance of model calibration with data from the deployment hospital.

Conclusions

Calibrating the prediction model with data from different deployment hospitals led to good performance in live settings. The performance degradation in the cross-hospital evaluation identified limitations in developing a generic model for different hospitals. Designing a generic process for model development to generate specialized prediction models for each hospital guarantees model performance in different hospitals.

Machine Learning to Identify Psychomotor Behaviors of Delirium for Patients in Long-Term Care Facility

This study aimed to develop accurate and explainable machine learning models for three psychomotor behaviors of delirium for hospitalized adult patients. A prospective pilot study was conducted with 33 participants admitted to a long-term care facility between August 10 and 25, 2020. During the pilot study, we collected 560 cases that included 33 clinical variables and the survey items from the short confusion assessment method (S-CAM), and developed a mobile-based application. Multiple machine learning algorithms, including four rule-mining algorithms (C4.5, CBA, MCAR, and LEM2) and four other statistical learning algorithms (LR, ANNs, SVMs with three kernel functions, and random forest), were validated by paired Wilcoxon signed-rank tests on both macro-averaged F1 and weighted average F1-measures during the 10-times stratified 2-fold cross-validation. The LEM2 algorithm achieved the best prediction performance (macro-averaged F1-measure of 49.35%; weighted average F1-measure of 96.55%), correctly identifying adult patients at delirium risk. In the pairwise comparison between predictive powers observed from independent models, the LEM2 model showed a medium or large effect size between 0.4925 and 0.8766 when compared with LR, ANN, SVM with RBF, and MCAR models. We have confirmed that acute consciousness in S-CAM assessment is closely associated with different predictors for screening three psychomotor behaviors of delirium: 1) education level, dementia type or its level, sleep disorder, dehydration, and infection in mixed-type delirium; 2) gender, education level, dementia type, dehydration, bedsores, and foley catheter in hyperactive delirium; and 3) pain, sleep disorder, and haloperidol use in hypoactive delirium.

Identifying patients with delirium based on unstructured clinical notes. (Preprint)

Background

Delirium in hospitalized patients is a syndrome of acute brain dysfunction. Diagnostic (International Classification of Diseases [ICD]) codes are often used in studies using electronic health records (EHRs), but they are inaccurate.

Objective

We sought to develop a more accurate method using natural language processing (NLP) to detect delirium episodes on the basis of unstructured clinical notes.

Methods

We collected 1.5 million notes from >10,000 patients from among 9 hospitals. Seven experts iteratively labeled 200,471 sentences. Using these, we trained three NLP classifiers: Support Vector Machine, Recurrent Neural Networks, and Transformer. Testing was performed using an external data set. We also evaluated associations with delirium billing (ICD) codes, medications, orders for restraints and sitters, direct assessments (Confusion Assessment Method [CAM] scores), and in-hospital mortality. F1 scores, confusion matrices, and areas under the receiver operating characteristic curve (AUCs) were used to compare NLP models. We used the φ coefficient to measure associations with other delirium indicators.

Results

The transformer NLP performed best on the following parameters: micro F1=0.978, macro F1=0.918, positive AUC=0.984, and negative AUC=0.992. NLP detections exhibited higher correlations (φ) than ICD codes with deliriogenic medications (0.194 vs 0.073 for ICD codes), restraints and sitter orders (0.358 vs 0.177), mortality (0.216 vs 0.000), and CAM scores (0.256 vs –0.028).

Conclusions

Clinical notes are an attractive alternative to ICD codes for EHR delirium studies but require automated methods. Our NLP model detects delirium with high accuracy, similar to manual chart review. Our NLP approach can provide more accurate determination of delirium for large-scale EHR-based studies regarding delirium, quality improvement, and clinical trails.

Use of Machine Learning and Artificial Intelligence Methods in Geriatric Mental Health Research Involving Electronic Health Record or Administrative Claims Data: A Systematic Review

Introduction: Electronic health records (EHR) and administrative healthcare data (AHD) are frequently used in geriatric mental health research to answer various health research questions. However, there is an increasing amount and complexity of data available that may lend itself to alternative analytic approaches using machine learning (ML) or artificial intelligence (AI) methods. We performed a systematic review of the current application of ML or AI approaches to the analysis of EHR and AHD in geriatric mental health. Methods: We searched MEDLINE, Embase, and PsycINFO to identify potential studies. We included all articles that used ML or AI methods on topics related to geriatric mental health utilizing EHR or AHD data. We assessed study quality either by Prediction model Risk OF Bias ASsessment Tool (PROBAST) or Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2) checklist. Results: We initially identified 391 articles through an electronic database and reference search, and 21 articles met inclusion criteria. Among the selected studies, EHR was the most used data type, and the datasets were mainly structured. A variety of ML and AI methods were used, with prediction or classification being the main application of ML or AI with the random forest as the most common ML technique. Dementia was the most common mental health condition observed. The relative advantages of ML or AI techniques compared to biostatistical methods were generally not assessed. Only in three studies, low risk of bias (ROB) was observed according to all the PROBAST domains but in none according to QUADAS-2 domains. The quality of study reporting could be further improved. Conclusion: There are currently relatively few studies using ML and AI in geriatric mental health research using EHR and AHD methods, although this field is expanding. Aside from dementia, there are few studies of other geriatric mental health conditions. The lack of consistent information in the selected studies precludes precise comparisons between them. Improving the quality of reporting of ML and AI work in the future would help improve research in the field. Other courses of improvement include using common data models to collect/organize data, and common datasets for ML model validation.

Predicting brain function status changes in critically ill patients via Machine learning

OBJECTIVE

In intensive care units (ICUs), a patient's brain function status can shift from a state of acute brain dysfunction (ABD) to one that is ABD-free and vice versa, which is challenging to forecast and, in turn, hampers the allocation of hospital resources. We aim to develop a machine learning model to predict next-day brain function status changes.

MATERIALS AND METHODS

Using multicenter prospective adult cohorts involving medical and surgical ICU patients from 2 civilian and 3 Veteran Affairs hospitals, we trained and externally validated a light gradient boosting machine to predict brain function status changes. We compared the performances of the boosting model against state-of-the-art models-an ABD predictive model and its variants. We applied Shapley additive explanations to identify influential factors to develop a compact model.

RESULTS

There were 1026 critically ill patients without evidence of prior major dementia, or structural brain diseases, from whom 12 295 daily transitions (ABD: 5847 days; ABD-free: 6448 days) were observed. The boosting model achieved an area under the receiver-operating characteristic curve (AUROC) of 0.824 (95% confidence interval [CI], 0.821-0.827), compared with the state-of-the-art models of 0.697 (95% CI, 0.693-0.701) with P < .001. Using 13 identified top influential factors, the compact model achieved 99.4% of the boosting model on AUROC. The boosting and the compact models demonstrated high generalizability in external validation by achieving an AUROC of 0.812 (95% CI, 0.812-0.813).

CONCLUSION

The inputs of the compact model are based on several simple questions that clinicians can quickly answer in practice, which demonstrates the model has direct prospective deployment potential into clinical practice, aiding in critical hospital resource allocation.

Prediction of delirium using data mining: A systematic review

Delirium remains a significant cause of morbidity, mortality and economic burden to society. "Big data" refers to data of significantly large volume, obtained from a variety of resources, which is created and processed at high velocity. We conducted a systematic review and meta-analysis exploring whether big data could predict the incidence of delirium of patients in the inpatient setting. Medline, Embase, the Cochrane Library, Web of Science, CINAHL, clinicaltrials.gov, who.int and IEEE Xplore were searched using MeSH terms "big data", "data mining", "delirium" and "confusion" up to 30th September 2019. We included both randomised and observational studies. The primary outcome of interest was development of delirium and the secondary outcomes of interest were type of statistical methods used, variables included in the mining algorithms and clinically important outcomes such as mortality and length of hospital stay. The quality of studies was graded using the CHARMs checklist. Six retrospective single centre observational studies were included (n = 178,091), of which 17, 574 participants developed delirium. Studies were of generally of low to moderate quality. The most commonly studied method was random forest, followed by support vector machine and artificial neural networks. The model with best performance for delirium prediction was random forest, with area under receiver operating curve (AUROC) ranging from 0.78 to 0.91. Sensitivity ranged from 0.59 to 0.81 and specificity ranged from 0.73 to 0.92. Our systematic review suggests that machine-learning techniques can be utilised to predict delirium.

A Machine Learning Approach for Investigating Delirium as a Multifactorial Syndrome

Delirium is a psycho-organic syndrome common in hospitalized patients, especially the elderly, and is associated with poor clinical outcomes. This study aims to identify the predictors that are mostly associated with the risk of delirium episodes using a machine learning technique (MLT). A random forest (RF) algorithm was used to evaluate the association between the subject’s characteristics and the 4AT (the 4 A’s test) score screening tool for delirium. RF algorithm was implemented using information based on demographic characteristics, comorbidities, drugs and procedures. Of the 78 patients enrolled in the study, 49 (63%) were at risk for delirium, 32 (41%) had at least one episode of delirium during the hospitalization (38% in orthopedics and 31% both in internal medicine and in the geriatric ward). The model explained 75.8% of the variability of the 4AT score with a root mean squared error of 3.29. Higher age, the presence of dementia, physical restraint, diabetes and a lower degree are the variables associated with an increase of the 4AT score. Random forest is a valid method for investigating the patients’ characteristics associated with delirium onset also in small case-series. The use of this model may allow for early detection of delirium onset to plan the proper adjustment in healthcare assistance.

Profiling Delirium Progression in Elderly Patients via Continuous-Time Markov Multi-State Transition Models

Poor recognition of delirium among hospitalized elderlies is a typical challenge for health care professionals. Considering methodological insufficiency for assessing time-varying diseases, a continuous-time Markov multi-state transition model (CTMMTM) was used to investigate delirium evolution in elderly patients. This is a longitudinal observational study performed in September 2016 in an Italian hospital. Change of delirium states was modeled according to the 4AT score. A Cox model (CM) and a CTMMTM were used for identifying factors affecting delirium onset both with a two-state and three-state model. In this study, 78 patients were enrolled and evaluated for 5 days. Both the CM and the CTMMTM show that urine catheter (UC), aging, drugs, and invasive devices (ID) are risk factors for delirium onset. The CTMMTM model shows that transition from no-delirium/cognitive impairment to delirium was associated with aging (HR = 1.14; 95%CI, 1.05, 1.23) and neuroleptics (HR = 4.3; 1.57, 11.77), dopaminergic drugs (HR = 3.89; 1.2, 12.6), UC (HR = 2.92; 1.09, 7.79) and ID (HR = 1.67; 103, 2.71). These results are confirmed by the multivariable model. Aging, ID, antibiotics, drugs affecting the central nervous system, and absence of moving ability are identified as the significant predictors of delirium. Additionally, it seems that modeling with CTMMTM may show associations that are not directly detectable with the traditional CM.

Technology Acceptance of a Machine Learning Algorithm Predicting Delirium in a Clinical Setting: a Mixed-Methods Study

Early identification of patients with life-threatening risks such as delirium is crucial in order to initiate preventive actions as quickly as possible. Despite intense research on machine learning for the prediction of clinical outcomes, the acceptance of the integration of such complex models in clinical routine remains unclear. The aim of this study was to evaluate user acceptance of an already implemented machine learning-based application predicting the risk of delirium for in-patients. We applied a mixed methods design to collect opinions and concerns from health care professionals including physicians and nurses who regularly used the application. The evaluation was framed by the Technology Acceptance Model assessing perceived ease of use, perceived usefulness, actual system use and output quality of the application. Questionnaire results from 47 nurses and physicians as well as qualitative results of four expert group meetings rated the overall usefulness of the delirium prediction positively. For healthcare professionals, the visualization and presented information was understandable, the application was easy to use and the additional information for delirium management was appreciated. The application did not increase their workload, but the actual system use was still low during the pilot study. Our study provides insights into the user acceptance of a machine learning-based application supporting delirium management in hospitals. In order to improve quality and safety in healthcare, computerized decision support should predict actionable events and be highly accepted by users.

Advancing care for acute gastrointestinal bleeding using artificial intelligence

The future of gastrointestinal bleeding will include the integration of machine learning algorithms to enhance clinician risk assessment and decision making. Machine learning algorithms have shown promise in outperforming existing clinical risk scores for both upper and lower gastrointestinal bleeding but have not been validated in any prospective clinical trials. The adoption of electronic health records provides an exciting opportunity to deploy risk prediction tools in real time and also to expand the data available to train predictive models. Machine learning algorithms can be used to identify patients with acute gastrointestinal bleeding using data extracted from the electronic health record. This can lead to an automated process to find patients with symptoms of acute gastrointestinal bleeding so that risk prediction tools can be then triggered to consistently provide decision support to the physician. Neural network models can be used to provide continuous risk predictions for patients who are at higher risk, which can be used to guide triage of patients to appropriate levels of care. Finally, the future will likely include neural network-based analysis of endoscopic stigmata of bleeding to help guide best practices for hemostasis during the endoscopic procedure. Machine learning will enhance the delivery of care at every level for patients with acute gastrointestinal bleeding through identifying very low risk patients for outpatient management, triaging high risk patients for higher levels of care, and guiding optimal intervention during endoscopy.