Performance of a Machine Learning Algorithm Using Electronic Health Record Data to Predict Postoperative Complications and Report on a Mobile Platform

Predicting postoperative complications after pneumonectomy using machine learning: a 10-year study

BACKGROUND

Reducing postoperative cardiovascular and neurological complications (PCNC) during thoracic surgery is the key to improving postoperative survival.

OBJECTIVE

We aimed to investigate independent predictors of PCNC, develop machine learning models, and construct a predictive nomogram for PCNC in patients undergoing thoracic surgery for lung cancer.

METHODS

This study used data from a previous retrospective study of 16,368 patients with lung cancer (training set: 11,458; validation set: 4,910) with American Standards Association physical statuses I-IV who underwent surgery. Postoperative information was collected from electronic medical records to help build models based on cause-and-effect and statistical data, potentially revealing hidden dependencies between factors and diseases in a big data environment. The optimal model was analyzed and filtered using multiple machine-learning models (Logistic regression, eXtreme Gradient Boosting, Random forest, Light Gradient Boosting Machine and Naïve Bayes). A predictive nomogram was built and receiver operating characteristics were used to assess the validity of the model. The discriminative power and clinical validity were assessed using calibration and decision-making curve analyses.

RESULTS

Multivariate logistic regression analysis revealed that age, surgery duration, intraoperative intercostal nerve block, postoperative patient-controlled analgesia, bronchial blocker use and sufentanil use were independent predictors of PCNC. Random forest was identified as the optimal model with an area under the curve of 0.898 in the training set and 0.752 in the validation set, confirming the excellent prediction accuracy of the nomogram. All the net benefits of the five machine-learning models in the training and validation sets demonstrated excellent clinical applicability, and the calibration curves showed good agreement between the predicted and observed risks.

CONCLUSION

The combination of machine-learning models and nomograms may contribute to the early prediction and reduction in the incidence of PCNC.

A Risk Prediction Model (CMC-AKIX) for Postoperative Acute Kidney Injury Using Machine Learning: Algorithm Development and Validation

BACKGROUND

Postoperative acute kidney injury (AKI) is a significant risk associated with surgeries under general anesthesia, often leading to increased mortality and morbidity. Existing predictive models for postoperative AKI are usually limited to specific surgical areas or require external validation.

OBJECTIVE

We proposed to build a prediction model for postoperative AKI using several machine learning methods.

METHODS

We conducted a retrospective cohort analysis of noncardiac surgeries from 2009 to 2019 at seven university hospitals in South Korea. We evaluated six machine learning models: deep neural network, logistic regression, decision tree, random forest, light gradient boosting machine, and naïve Bayes for predicting postoperative AKI, defined as a significant increase in serum creatinine or the initiation of renal replacement therapy within 30 days after surgery. The performance of the models was analyzed using the area under the curve (AUC) of the receiver operating characteristic curve, accuracy, precision, sensitivity (recall), specificity, and F1-score.

RESULTS

Among the 239,267 surgeries analyzed, 7935 cases of postoperative AKI were identified. The models, using 38 preoperative predictors, showed that deep neural network (AUC=0.832), light gradient boosting machine (AUC=0.836), and logistic regression (AUC=0.825) demonstrated superior performance in predicting AKI risk. The deep neural network model was then developed into a user-friendly website for clinical use.

CONCLUSIONS

Our study introduces a robust, high-performance AKI risk prediction system that is applicable in clinical settings using preoperative data. This model's integration into a user-friendly website enhances its clinical utility, offering a significant step forward in personalized patient care and risk management.

Management of lower extremity traumas: Comparing appropriate use criteria ChatGPT recommendations

BACKGROUND

High-energy lower extremity injury presents with difficult clinical decisions because successful limb salvage is the best scenario for complex traumas, but early amputation may be necessary to limit complications. Artificial Intelligence is a tool rising in popularity to help make clinical judgements.

PURPOSE/QUESTIONS

The aim of this study is to determine whether ChatGPT-4 can produce accurate recommendations for limb salvage or amputation given various patient scenarios.

METHODS

Various lower leg trauma scenarios were given to the appropriate use criteria for limb salvage made by AAOS or ChatGPT-4. A recommendation score for limb salvage and early amputation were collected. Tests to determine statistical significance between AAOS and ChatGPT-4 were performed.

RESULTS

A total of 196 patient scenario combinations were utilized. The mean error for limb salvage and early amputation were -0.3 and -0.2 respectively. AAOS and ChatGPT had significant positive correlations when predicting limb salvage and early amputation scores. The effect size of limb salvage and early amputation was -0.094 and -0.14, respectively.

CONCLUSION

ChatGPT-4 generally under-estimates appropriateness scores for both limb salvage and early amputation treatment options, but produces similar scores. ChatGPT-4 may be used to aid physicians in choosing between limb salvage and early amputation, though with caution.

Prognostic model for log odds of negative lymph node in locally advanced rectal cancer via interpretable machine learning

No studies have examined the prognostic value of the log odds of negative lymph nodes/T stage (LONT) in locally advanced rectal cancer (LARC) treated with neoadjuvant chemoradiotherapy (nCRT). We aimed to assess the prognostic value of LONT and develop a machine learning model to predict overall survival (OS) and disease-free survival (DFS) in LARC patients treated with nCRT. The study included 820 LARC patients who received nCRT between September 2010 and October 2017. Univariate and multivariate Cox regression analyses identified prognostic factors, which were then used to develop risk assessment models with 9 machine learning algorithms. Model hyperparameters were optimized using random search and 10-fold cross-validation. The models were evaluated using metrics such as the area under the receiver operating characteristic curves (AUC), decision curve analysis, calibration curves, and precision and accuracy for predicting OS and DFS. Shapley's additive explanations (SHAP) was also used for model interpretation. The study included 820 patients, identifying LONT as a significant independent prognostic factor for both OS and DFS. Nine machine learning algorithms were used to create predictive models based on these factors. The extreme gradient boosting (XGB) model showed the best performance, with a mean AUC of 0.89 for OS and 0.83 for DFS in 10-fold cross-validation. Additionally, the predictions generated by the XGB model were analyzed using SHAP. Finally, we developed an online web-based calculator utilizing the XGB model to enhance the model's generalizability and to provide improved support for physicians in their decision-making processes. The study developed an XGB model utilizing LONT to predict OS and DFS in patients with LARC undergoing nCRT. Furthermore, an online web calculator was constructed using the XGB model to facilitate the model's generalization and to enhance physician decision-making.

Introduction to Artificial Intelligence for General Surgeons: A Narrative Review

Artificial intelligence (AI) has rapidly progressed in the last decade and will inevitably become incorporated into trauma and surgical systems. In such settings, surgeons often need to make high-stakes, time-sensitive, and complex decisions with limited or uncertain information. AI has great potential to augment the pre-operative, intra-operative, and post-operative phases of trauma care. Despite the expeditious advancement of AI, many surgeons lack a foundational understanding of AI terminology, its processes, and potential applications in clinical practice. This narrative review aims to educate general surgeons about the basics of AI, highlight its applications in thoraco-abdominal trauma, and discuss the implications of incorporating its use into the Australian health care system. This review found that studies of AI in trauma care have predominantly focused on machine learning and deep learning applied to diagnostics, risk prediction, and decision-making. Other subfields of AI include natural language processing and computer vision. While AI tools have many potential applications in trauma care, current clinical use is limited. Future prospective, locally validated research is required prior to incorporating AI into clinical practice.

GAPPA: Enhancing prognosis prediction in primary aldosteronism post-adrenalectomy using graph-based modeling

BACKGROUND AND OBJECTIVE

Predicting postoperative prognosis is vital for clinical decision making in patients undergoing adrenalectomy (ADX). This study introduced GAPPA, a novel GNN-based approach, to predict post-ADX outcomes in patients with unilateral primary aldosteronism (UPA). The objective was to leverage the intricate dependencies between clinico-biochemical features and clinical outcomes using GNNs integrated into a bipartite graph structure to enhance prognostic prediction accuracy.

METHODS

We conceptualized prognostic prediction as a link prediction task on a bipartite graph, with nodes representing patients, clinico-biochemical features, and clinical outcomes, and edges denoting the connections between them. GAPPA utilizes GNNs to capture these dependencies and seamlessly integrates the outcome predictions into a graph structure. This approach was evaluated using a dataset of 640 patients with UPA who underwent unilateral ADX (uADX) between 1990 and 2022. We conducted a comparative analysis using repeated stratified five-fold cross-validation and paired t-tests to evaluate the performance of GAPPA against conventional machine learning methods and previous studies across various metrics.

RESULTS

GAPPA significantly outperformed conventional machine learning methods and previous studies (p < 0.05) across various metrics. It achieved F1-score, accuracy, sensitivity, and specificity of 71.3 % ± 3.1 %, 71.1 % ± 3.4 %, 69.9 % ± 4.3 %, and 72.4 % ± 7.2 %, respectively, with an AUC of 0.775 ± 0.030. We also investigated the impact of different initialization schemes on GAPPA outcome-edge embeddings, highlighting their robustness and stability.

CONCLUSION

GAPPA aids in preoperative prognosis assessment and facilitates patient counseling, contributing to prognostic prediction and advancing the applications of GNNs in the biomedical domain.

Leveraging artificial intelligence for predicting spontaneous closure of perimembranous ventricular septal defect in children: a multicentre, retrospective study in China

BACKGROUND

Perimembranous ventricular septal defect (PMVSD) is a prevalent congenital heart disease, presenting challenges in predicting spontaneous closure, which is crucial for therapeutic decisions. Existing models mainly rely on structured echocardiographic parameters or restricted data. This study introduces an artificial intelligence (AI)-based model, which uses natural language processing (NLP) and machine learning with the aim of improving spontaneous closure predictability in PMVSD.

METHODS

We did a multicentre, retrospective analysis using data from 29 142 PMVSD patients across six tertiary centres in China from May, 2004, to September, 2022, for training (70%) and validation (30%; dataset 1, 27 269 patients), and from September, 2001, to December, 2009 for testing (dataset 2, 1873 patients). NLP extracted structured data from echocardiography reports and medical records, which were used to develop machine learning models. Models were evaluated for spontaneous closure occurrence and timing by use of area under the receiver operating characteristic curve (AUC), decision curve analysis, and calibration index.

FINDINGS

Spontaneous closure occurred in 3520 patients (12·1%) at a median of 31 months (IQR 16-56). Eleven NLP-derived predictors, identified via least absolute shrinkage and selection operator, highlighted the importance of defect morphology and patient age. The random survival forest algorithm, selected for its superior concordance indexes, showed excellent predictive performance with validation set AUCs (95% CI) of 0·95 (0·94-0·96) for 1-year and 3-year predictions, and 0·95 (0·95-0·96) for 5-year predictions; testing set AUCs were 0·95 (0·94-0·97) for 1-year predictions, 0·97 (0·96-0·98) for 3-year predictions, and 0·98 (0·97-0·99) for 5-year predictions. The model showed high clinical utility through decision curve analysis, calibration, and risk stratification, maintaining consistent accuracy across centres and subgroups.

INTERPRETATION

This AI-based model for predicting spontaneous closure in PMVSD patients represents a substantial advancement, potentially improving patient management, reducing risks of delayed or inappropriate treatment, and enhancing clinical outcomes.

FUNDING

National Natural Science Foundation of China, Shanghai Municipal Hospital Clinical Technology Project, Shanghai Municipal Health Commission, and Clinical Research Unit of XinHua Hospital.

The present and future of digital health, digital medicine, and digital therapeutics for allergic diseases

BACKGROUND

Digital health, digital medicine, and digital therapeutics integrate advanced computer technologies into healthcare, aiming to improve efficiency and patient outcomes. These technologies offer innovative solutions for the management of allergic diseases, which affect a significant proportion of the global population and are increasing in prevalence. BODY: This review examines the current progress and future potential of digital health in allergic disease management. It highlights key advancements, including telehealth, mobile health (mHealth), artificial intelligence, clinical decision support systems (CDSS), and digital biomarkers, with a focus on their relevance to allergic disease management. The role of digital tools in improving treatment adherence, enabling remote care, and integrating environmental and patient data into personalized care models is discussed. Challenges such as data privacy, interoperability, and equitable access are addressed, alongside potential strategies to overcome these barriers.

CONCLUSION

Digital therapy will play an important role in allergic diseases, and the further development of digital therapies will effectively promote the development of clinical research, digital biomarkers, hypoallergenic environments and digital twins. More research is needed to support the progress of digital therapy for allergic diseases.

Protocol for the impact of machine learning-based clinician decision support algorithims in perioperative care (IMAGINATIVE) in Singapore general hospital : a large prospective randomised controlled trial

INTRODUCTION

As surgical accessibility improves, the incidence of postoperative complications is expected to rise. The implementation of a precise and objective risk stratification tool holds the potential to mitigate these complications by early identification of high-risk patients. Moreover, it could address the escalating costs from resource misallocation. In Singapore General Hospital (SGH), we introduced the Combined Assessment of Risk Encountered in Surgery-Machine Learning (CARES-ML) in June 2023, focusing on predicting 30-day postoperative mortality and the need for post-surgery intensive care unit (ICU) stays. The IMAGINATIVE Trial aims to evaluate the efficacy of such systems in a large academic medical centre.

METHODS AND ANALYSIS

This study adopts type 1 effectiveness-implementation study design within a randomised controlled trial framework. Patients will be randomly assigned in a 1:1 ratio to either the CARES-guided group (unblinded to risk level) or the unguided group (blinded to the risk level). A total of 9200 patients will be enrolled in the study, with the inclusion criteria encompassing individuals aged 21-100 years old undergoing elective surgeries except for neurology and cardiology surgeries at SGH. The primary outcome is to evaluate the effectiveness of the Machine Learning Clinical Decision Support (ML-CDS) algorithm in improving perioperative mortality rates when integrated into the clinical workflow.

ETHICS AND DISSEMINATION

The study has been approved by the SingHealth Centralised Institutional Review Board (CIRB Ref: 2023:2114) and is registered on ClinicalTrials.gov (trial number: NCT05809232). All patients will sign an informed consent form before recruitment and translators will be made available to non-English-speaking participants. This study is funded by the National Medical Research Council, Singapore (HCSAINV22jul-0002) and the findings will be published in peer-reviewed journals and presented at academic conferences.

TRIAL REGISTRATION NUMBER

NCT05809232.

Prediction and validation of pathologic complete response for locally advanced rectal cancer under neoadjuvant chemoradiotherapy based on a novel predictor using interpretable machine learning

BACKGROUND

Precise evaluation of pathological complete response (pCR) is essential for determining the prognosis of patients with locally advanced rectal cancer (LARC) undergoing neoadjuvant chemoradiotherapy (NCRT) and can offer clues for the selection of subsequent treatment strategies. Most current predictive models for pCR focus primarily on pre-treatment factors, neglecting the dynamic systemic changes that occur during neoadjuvant chemoradiotherapy, and are constrained by low accuracy and lack of integrity.

PURPOSE

This study devised a novel predictor of pCR using dynamic alterations in systemic inflammation-nutritional marker indexes (SINI) during neoadjuvant therapy and developed a machine-learning model to predict pCR.

METHODS

Two cohorts of patients with LARC from center one from 2012 to 2017 and from center two from 2020 to 2023 were integrated for analysis. This study compared dynamic changes in blood indexes before and after neoadjuvant therapy and surgical operation. A least absolute shrinkage and selection operator (LASSO) regression analysis was conducted to mitigate collinearity and identify key indexes, constructing the SINI. Univariate and multiple logistic regression analyses were used to identify the independent risk factors associated with pCR. Additionally, 10 machine learning algorithms were employed to develop predictive models to assess risk. The hyperparameters of the machine learning models were optimized using a random search and 10-fold cross-validation. The models were assessed by examining various metrics, including the area under the receiver operating characteristic curves (AUC), the area under the precision-recall curve (AUPRC), decision curve analysis, calibration curves, and the precision and accuracy of the internal and external validation cohorts. Additionally, Shapley's additive explanations (SHAP) were employed to interpret the machine learning models.

RESULTS

The study cohort comprised 677 patients from the center one and 224 patients from the center two. Six key indexes were identified, and a predictive index, SINI, was constructed. Univariate and multiple logistic regression analyses revealed that SINI, clinical T-stage, clinical N-stage, tumor size, and the distance from the anal verge were independent risk factors for pCR in patients with LARC following NCRT. The mean AUC value of the extreme gradient boosting (XGB) model in the 10-fold cross-validation of the training set was 0.877. The XGB model demonstrated superior performance in the internal and external validation sets. Specifically, in the internal test set, the XGB model achieved an AUC of 0.86, AUPRC of 0.707, accuracy of 0.82, and precision of 0.80. In the external validation set, the XGB model exhibited an AUC of 0.83, AUPRC of 0.702, accuracy of 0.81, and precision of 0.81. Additionally, the predictions generated by the XGB model were analyzed using SHAP.

CONCLUSION

This study involved developing and validating an XGB model using SINI to predict pCR in patients with LARC. Besides, a SINI-based machine learning model shows promise in accurately predicting pCR following NCRT in patients with resectable LARC, offering valuable insights for personalized treatment approaches.

Risk-Specific Training Cohorts to Address Class Imbalance in Surgical Risk Prediction

Importance

Machine learning tools are increasingly deployed for risk prediction and clinical decision support in surgery. Class imbalance adversely impacts predictive performance, especially for low-incidence complications.

Objective

To evaluate risk-prediction model performance when trained on risk-specific cohorts.

Design, Setting, and Participants

This cross-sectional study performed from February 2024 to July 2024 deployed a deep learning model, which generated risk scores for common postoperative complications. A total of 109 445 inpatient operations performed at 2 University of Florida Health hospitals from June 1, 2014, to May 5, 2021 were examined.

Exposures

The model was trained de novo on separate cohorts for high-risk, medium-risk, and low-risk Common Procedure Terminology codes defined empirically by incidence of 5 postoperative complications: (1) in-hospital mortality; (2) prolonged intensive care unit (ICU) stay (≥48 hours); (3) prolonged mechanical ventilation (≥48 hours); (4) sepsis; and (5) acute kidney injury (AKI). Low-risk and high-risk cutoffs for complications were defined by the lower-third and upper-third prevalence in the dataset, except for mortality, cutoffs for which were set at 1% or less and greater than 3%, respectively.

Main Outcomes and Measures

Model performance metrics were assessed for each risk-specific cohort alongside the baseline model. Metrics included area under the receiver operating characteristic curve (AUROC), area under the precision-recall curve (AUPRC), F1 scores, and accuracy for each model.

Results

A total of 109 445 inpatient operations were examined among patients treated at 2 University of Florida Health hospitals in Gainesville (77 921 procedures [71.2%]) and Jacksonville (31 524 procedures [28.8%]). Median (IQR) patient age was 58 (43-68) years, and median (IQR) Charlson Comorbidity Index score was 2 (0-4). Among 109 445 operations, 55 646 patients were male (50.8%), and 66 495 patients (60.8%) underwent a nonemergent, inpatient operation. Training on the high-risk cohort had variable impact on AUROC, but significantly improved AUPRC (as assessed by nonoverlapping 95% confidence intervals) for predicting mortality (0.53; 95% CI, 0.43-0.64), AKI (0.61; 95% CI, 0.58-0.65), and prolonged ICU stay (0.91; 95% CI, 0.89-0.92). It also significantly improved F1 score for mortality (0.42; 95% CI, 0.36-0.49), prolonged mechanical ventilation (0.55; 95% CI, 0.52-0.58), sepsis (0.46; 95% CI, 0.43-0.49), and AKI (0.57; 95% CI, 0.54-0.59). After controlling for baseline model performance on high-risk cohorts, AUPRC increased significantly for in-hospital mortality only (0.53; 95% CI, 0.42-0.65 vs 0.29; 95% CI, 0.21-0.40).

Conclusion and Relevance

In this cross-sectional study, by training separate models using a priori knowledge for procedure-specific risk classes, improved performance in standard evaluation metrics was observed, especially for low-prevalence complications like in-hospital mortality. Used cautiously, this approach may represent an optimal training strategy for surgical risk-prediction models.

Feature selection integrating Shapley values and mutual information in reinforcement learning: An application in the prediction of post-operative outcomes in patients with end-stage renal disease

BACKGROUND

In predicting post-operative outcomes for patients with end-stage renal disease, our study faced challenges related to class imbalance and a high-dimensional feature space. Therefore, with a focus on overcoming class imbalance and improving interpretability, we propose a novel feature selection approach using multi-agent reinforcement learning.

METHODS

We proposed a multi-agent feature selection model based on a comprehensive reward function that combines classification model performance, Shapley additive explanations values, and the mutual information. The definition of rewards in reinforcement learning is crucial for model convergence and performance improvement. Initially, we set a deterministic reward based on the mutual information between variables and the target class, selecting variables that are highly dependent on the class, thus accelerating convergence. We then prioritized variables that influence the minority class on a sample basis and introduced a dynamic reward distribution strategy using Shapley additive explanations values to improve interpretability and solve the class imbalance problem.

RESULTS

Involving the integration of electronic medical records, anesthesia records, operating room vital signs, and pre-operative anesthesia evaluations, our approach effectively mitigated class imbalance and demonstrated superior performance in ablation analysis. Our model achieved a 16% increase in the minority class F1 score and an 8.2% increase in the overall F1 score compared to the baseline model without feature selection.

CONCLUSION

This study contributes important research findings that show that the multi-agent-based feature selection method can be a promising approach for solving the class imbalance problem.

Integrating AI into Breast Reconstruction Surgery: Exploring Opportunities, Applications, and Challenges

Background: Artificial intelligence (AI) has significantly influenced various sectors, including healthcare, by enhancing machine capabilities in assisting with human tasks. In surgical fields, where precision and timely decision-making are crucial, AI's integration could revolutionize clinical quality and health resource optimization. This study explores the current and future applications of AI technologies in reconstructive breast surgery, aiming for broader implementation. Methods: We conducted systematic reviews through PubMed, Web of Science, and Google Scholar using relevant keywords and MeSH terms. The focus was on the main AI subdisciplines: machine learning, computer vision, natural language processing, and robotics. This review includes studies discussing AI applications across preoperative, intraoperative, postoperative, and academic settings in breast plastic surgery. Results: AI is currently utilized preoperatively to predict surgical risks and outcomes, enhancing patient counseling and informed consent processes. During surgery, AI supports the identification of anatomical landmarks and dissection strategies and provides 3-dimensional visualizations. Robotic applications are promising for procedures like microsurgical anastomoses, flap harvesting, and dermal matrix anchoring. Postoperatively, AI predicts discharge times and customizes follow-up schedules, which improves resource allocation and patient management at home. Academically, AI offers personalized training feedback to surgical trainees and aids research in breast reconstruction. Despite these advancements, concerns regarding privacy, costs, and operational efficacy persist and are critically examined in this review. Conclusions: The application of AI in breast plastic and reconstructive surgery presents substantial benefits and diverse potentials. However, much remains to be explored and developed. This study aims to consolidate knowledge and encourage ongoing research and development within the field, thereby empowering the plastic surgery community to leverage AI technologies effectively and responsibly for advancing breast reconstruction surgery.

Effect of machine learning models on clinician prediction of postoperative complications: the Perioperative ORACLE randomised clinical trial

BACKGROUND

Anaesthesiologists might be able to mitigate risk if they know which patients are at greatest risk for postoperative complications. This trial examined the impact of machine learning models on clinician risk assessment.

METHODS

This single-centre, prospective, randomised clinical trial enrolled surgical patients aged ≥18 yr. Anaesthesiologists and nurse anaesthetists providing remote telemedicine support reviewed electronic health records with (assisted group) or without (unassisted group) reviewing machine learning predictions. Clinicians predicted the likelihood of postoperative 30-day all-cause mortality and postoperative acute kidney injury (AKI) within 7 days. The primary outcome was area under the receiver operating characteristic curve (AUROC) for clinician predictions of mortality and AKI, comparing AUROCs between assisted and unassisted assessments.

RESULTS

We analysed 5071 patients (mean [range] age: 58 [18-100] yr; 52% female) assessed by 89 clinicians. Of these, 98 (2.2%) patients died within 30 days of surgery and 450 (11.1%) patients sustained AKI. Clinician predictions agreed with the models more strongly in the assisted vs unassisted group (weighted kappa 0.75 vs 0.62 for death, mean difference: 0.13 [95% CI 0.10-0.17]; and 0.79 vs 0.54 for AKI, mean difference: 0.25 [95% CI 0.21-0.29]). Clinical prediction of death was similar between the assisted (AUROC 0.793) and unassisted (AUROC 0.780) groups (mean difference: 0.013 [95% CI -0.070 to 0.097]; P=0.76). Prediction of AKI had an AUROC of 0.734 in the assisted group vs 0.688 in the unassisted group (difference 0.046 [95% CI -0.003 to 0.091]; P=0.06).

CONCLUSIONS

Clinician performance was not improved by machine learning assistance. Further work is needed to clarify the role of machine learning in real-time perioperative risk stratification.

CLINICAL TRIAL REGISTRATION

NCT05042804.

Machine Learning-Based Predictive Models for Patients with Venous Thromboembolism: A Systematic Review

BACKGROUND

 Venous thromboembolism (VTE) is a chronic disorder with a significant health and economic burden. Several VTE-specific clinical prediction models (CPMs) have been used to assist physicians in decision-making but have several limitations. This systematic review explores if machine learning (ML) can enhance CPMs by analyzing extensive patient data derived from electronic health records. We aimed to explore ML-CPMs' applications in VTE for risk stratification, outcome prediction, diagnosis, and treatment.

METHODS

 Three databases were searched: PubMed, Google Scholar, and IEEE electronic library. Inclusion criteria focused on studies using structured data, excluding non-English publications, studies on non-humans, and certain data types such as natural language processing and image processing. Studies involving pregnant women, cancer patients, and children were also excluded. After excluding irrelevant studies, a total of 77 studies were included.

RESULTS

 Most studies report that ML-CPMs outperformed traditional CPMs in terms of receiver operating area under the curve in the four clinical domains that were explored. However, the majority of the studies were retrospective, monocentric, and lacked detailed model architecture description and external validation, which are essential for quality audit. This review identified research gaps and highlighted challenges related to standardized reporting, reproducibility, and model comparison.

CONCLUSION

 ML-CPMs show promise in improving risk assessment and individualized treatment recommendations in VTE. Apparently, there is an urgent need for standardized reporting and methodology for ML models, external validation, prospective and real-world data studies, as well as interventional studies to evaluate the impact of artificial intelligence in VTE.

Venous thromboembolism in the era of machine learning and artificial intelligence in medicine

In this review, we embark on a comprehensive exploration of venous thromboembolism (VTE) in the context of medical history and its current practice within medicine. We delve into the landscape of artificial intelligence (AI), exploring its present utility and envisioning its transformative roles within VTE management, from prevention to screening and beyond. Central to our discourse is a forward-looking perspective on the integration of AI within VTE in medicine, advocating for rigorous study design, robust validation processes, and meticulous statistical analysis to gauge the efficacy of AI applications. We further illuminate the potential of large language models and generative AI in revolutionizing VTE care, while acknowledging their inherent limitations and proposing innovative solutions to overcome challenges related to data availability and integrity, including the strategic use of synthetic data. The critical importance of navigating ethical, legal, and privacy concerns associated with AI is underscored, alongside the imperative for comprehensive governance and policy frameworks to regulate its deployment in VTE treatment. We conclude on a note of cautious optimism, where we highlight the significance of proactively addressing the myriad challenges that accompany the advent of AI in healthcare. Through diligent design, stringent validation, extensive education, and prudent regulation, we can harness AI's potential to significantly enhance our understanding and management of VTE. As we stand on the cusp of a new era, our commitment to these principles will be instrumental in ensuring that the promise of AI is fully realized within the realm of VTE care.

Clinical Decision Support for Surgery: A Mixed Methods Study on Design and Implementation Perspectives From Urologists

OBJECTIVE

To assess urologist attitudes toward clinical decision support (CDS) embedded into the electronic health record (EHR) and define design needs to facilitate implementation and impact. With recent advances in big data and artificial intelligence (AI), enthusiasm for personalized, data-driven tools to improve surgical decision-making has grown, but the impact of current tools remains limited.

METHODS

A sequential explanatory mixed methods study from 2019 to 2020 was performed. First, survey responses from the 2019 American Urological Association Annual Census evaluated attitudes toward an automatic CDS tool that would display risk/benefit data. This was followed by the purposeful sampling of 25 urologists and qualitative interviews assessing perspectives on CDS impact and design needs. Bivariable, multivariable, and coding-based thematic analysis were applied and integrated.

RESULTS

Among a weighted sample of 12,366 practicing urologists, the majority agreed CDS would help decision-making (70.9%, 95% CI 68.7%-73.2%), aid patient counseling (78.5%, 95% CI 76.5%-80.5%), save time (58.1%, 95% CI 55.7%-60.5%), and improve patient outcomes (42.9%, 95% CI 40.5%-45.4%). More years in practice was negatively associated with agreement (P <.001). Urologists described how CDS could bolster evidence-based care, personalized medicine, resource utilization, and patient experience. They also identified multiple implementation barriers and provided suggestions on form, functionality, and visual design to improve usefulness and ease of use.

CONCLUSION

Urologists have favorable attitudes toward the potential for clinical decision support in the EHR. Smart design will be critical to ensure effective implementation and impact.

A Novel Machine-Learning Algorithm to Predict Stone Recurrence with 24-Hour Urine Data

Objectives: The absence of predictive markers for kidney stone recurrence poses a challenge for the clinical management of stone disease. The unpredictability of stone events is also a significant limitation for clinical trials, where many patients must be enrolled to obtain sufficient stone events for analysis. In this study, we sought to use machine learning methods to identify a novel algorithm to predict stone recurrence. Subjects/Patients and Methods: Patients enrolled in the Registry for Stones of the Kidney and Ureter (ReSKU), a registry of nephrolithiasis patients collected between 2015-2020, with at least one prospectively collected 24-hour urine test (Litholink 24-hour urine test; Labcorp) were included in the training set. A validation set was obtained from chart review of stone patients not enrolled in ReSKU with 24-hour urine data. Stone events were defined as either an office visit where a patient reports symptomatic passage of stones or a surgical procedure for stone removal. Seven prediction classification methods were evaluated. Predictive analyses and receiver operator characteristics (ROC) curve generation were performed in R. Results: A training set of 423 kidney stone patients with stone event data and 24-hour urine samples were trained using the prediction classification methods. The highest performing prediction model was a Logistic Regression with ElasticNet machine learning model (area under curve [AUC] = 0.65). Restricting analysis to high confidence predictions significantly improved model accuracy (AUC = 0.82). The prediction model was validated on a validation set of 172 stone patients with stone event data and 24-hour urine samples. Prediction accuracy in the validation set demonstrated moderate discriminative ability (AUC = 0.64). Repeat modeling was performed with four of the highest scoring features, and ROC analyses demonstrated minimal loss in accuracy (AUC = 0.63). Conclusion: Machine-learning models based on 24-hour urine data can predict stone recurrences with a moderate degree of accuracy.

Development of a machine learning approach for prediction of red blood cell transfusion in patients undergoing Cesarean section at a single institution

Despite recent advances in surgical techniques and perinatal management in obstetrics for reducing intraoperative bleeding, blood transfusion may occur during a cesarean section (CS). This study aims to identify machine learning models with an optimal diagnostic performance for intraoperative transfusion prediction in parturients undergoing a CS. Additionally, to address model performance degradation due to data imbalance, this study further investigated the variation in predictive model performance depending on the ratio of event to non-event data (1:1, 1:2, 1:3, and 1:4 model datasets and raw data).The area under the receiver operating characteristic curve (AUROC) and area under the precision-recall curve (AUPRC) were evaluated to compare the predictive accuracy of different machine learning algorithms, including XGBoost, K-nearest neighbor, decision tree, support vector machine, multilayer perceptron, logistic regression, random forest, and deep neural network. We compared the predictive performance of eight prediction algorithms that were applied to five types of datasets. The intraoperative transfusion in maternal CS was 7.2% (1020/14,254). XGBoost showed the highest AUROC (0.8257) and AUPRC (0.4825) among the models. The most significant predictors for transfusion in maternal CS as per machine learning models were placenta previa totalis, haemoglobin, placenta previa partialis, and platelets. In all eight prediction algorithms, the change in predictive performance based on the AUROC and AUPRC according to the resampling ratio was insignificant. The XGBoost algorithm exhibited optimal performance for predicting intraoperative transfusion. Data balancing techniques employed to alter the event data composition ratio of the training data failed to improve the performance of the prediction model.

KAMLN: A Knowledge-aware Multi-label Network for Lung Cancer Complication Prediction

Surgical resection is now the only curative approach for early stage lung cancer patients. However, postoperative complications pose a significant threat to the health and life of patients. The current complication prediction methods usually ignore the potential causal relationships between different complications, which may limit their predictive performances. To exploit this knowledge, we propose a knowledge-aware multi-label network (KAMLN) for complication prediction. In this approach, we first construct a knowledge graph to describe the potential causal relationships between different complications. Then, we design a neural network based on this knowledge graph to learn the relationships between different complications to achieve better predictive performances. Experiments using 593 lung cancer patients' data show that the KAMLN achieves a micro-AUC value of 0.664±0.100, which is better than the baseline methods. The SHAP analysis indicates lymph node dissection has a significant impact on multiple complications. Based on the experimental results, the proposed KAMLN can effectively utilize prior knowledge between different complications to achieve more accurate and fine-grained complication prediction.

An explainable long short-term memory network for surgical site infection identification

BACKGROUND

Currently, surgical site infection surveillance relies on labor-intensive manual chart review. Recently suggested solutions involve machine learning to identify surgical site infections directly from the medical record. Deep learning is a form of machine learning that has historically performed better than traditional methods while being harder to interpret. We propose a deep learning model, a long short-term memory network, for the identification of surgical site infection from the medical record with an attention layer for explainability.

METHODS

We retrieved structured data and clinical notes from the University of Utah Health System's electronic health care record for operative events randomly selected for manual chart review from January 2016 to June 2021. Surgical site infection occurring within 30 days of surgery was determined according to the National Surgical Quality Improvement Program definition. We trained the long short-term memory model along with traditional machine learning models for comparison. We calculated several performance metrics from a holdout test set and performed additional analyses to understand the performance of the long short-term memory, including an explainability analysis.

RESULTS

Surgical site infection was present in 4.7% of the total 9,185 operative events. The area under the receiver operating characteristic curve and sensitivity of the long short-term memory was higher (area under the receiver operating characteristic curve: 0.954, sensitivity: 0.920) compared to the top traditional model (area under the receiver operating characteristic curve: 0.937, sensitivity: 0.736). The top 5 features of the long short-term memory included 2 procedure codes and 3 laboratory values.

CONCLUSION

Surgical site infection surveillance is vital for the reduction of surgical site infection rates. Our explainable long short-term memory achieved a comparable area under the receiver operating characteristic curve and greater sensitivity when compared to traditional machine learning methods. With explainable deep learning, automated surgical site infection surveillance could replace burdensome manual chart review processes.

Diagnosis to dissection: AI's role in early detection and surgical intervention for gastric cancer

Gastric cancer remains a formidable health challenge worldwide; early detection and effective surgical intervention are critical for improving patient outcomes. This comprehensive review explores the evolving landscape of gastric cancer management, emphasizing the significant contributions of artificial intelligence (AI) in revolutionizing both diagnostic and therapeutic approaches. Despite advancements in the medical field, the subtle nature of early gastric cancer symptoms often leads to late-stage diagnoses, where survival rates are notably decreased. Historically, the treatment of gastric cancer has transitioned from palliative care to surgical resection, evolving further with the introduction of minimally invasive surgical (MIS) techniques. In the current era, AI has emerged as a transformative force, enhancing the precision of early gastric cancer detection through sophisticated image analysis, and supporting surgical decision-making with predictive modeling and real-time preop-, intraop-, and postoperative guidance. However, the deployment of AI in healthcare raises significant ethical, legal, and practical challenges, including the necessity for ongoing professional education and the development of standardized protocols to ensure patient safety and the effective use of AI technologies. Future directions point toward a synergistic integration of AI with clinical best practices, promising a new era of personalized, efficient, and safer gastric cancer management.

Large Language Models for Intraoperative Decision Support in Plastic Surgery: A Comparison between ChatGPT-4 and Gemini

Background and Objectives: Large language models (LLMs) are emerging as valuable tools in plastic surgery, potentially reducing surgeons' cognitive loads and improving patients' outcomes. This study aimed to assess and compare the current state of the two most common and readily available LLMs, Open AI's ChatGPT-4 and Google's Gemini Pro (1.0 Pro), in providing intraoperative decision support in plastic and reconstructive surgery procedures. Materials and Methods: We presented each LLM with 32 independent intraoperative scenarios spanning 5 procedures. We utilized a 5-point and a 3-point Likert scale for medical accuracy and relevance, respectively. We determined the readability of the responses using the Flesch-Kincaid Grade Level (FKGL) and Flesch Reading Ease (FRE) score. Additionally, we measured the models' response time. We compared the performance using the Mann-Whitney U test and Student's t-test. Results: ChatGPT-4 significantly outperformed Gemini in providing accurate (3.59 ± 0.84 vs. 3.13 ± 0.83, p-value = 0.022) and relevant (2.28 ± 0.77 vs. 1.88 ± 0.83, p-value = 0.032) responses. Alternatively, Gemini provided more concise and readable responses, with an average FKGL (12.80 ± 1.56) significantly lower than ChatGPT-4's (15.00 ± 1.89) (p < 0.0001). However, there was no difference in the FRE scores (p = 0.174). Moreover, Gemini's average response time was significantly faster (8.15 ± 1.42 s) than ChatGPT'-4's (13.70 ± 2.87 s) (p < 0.0001). Conclusions: Although ChatGPT-4 provided more accurate and relevant responses, both models demonstrated potential as intraoperative tools. Nevertheless, their performance inconsistency across the different procedures underscores the need for further training and optimization to ensure their reliability as intraoperative decision-support tools.

Development and validation of an ensemble learning risk model for sepsis after abdominal surgery

Introduction

Although their importance has gained attention, the clinical applications of methods for screening patients at high risk of sepsis after abdominal surgery have been restricted. Therefore, we aimed to develop and validate models for screening patients at high risk of sepsis after abdominal surgery based on machine learning with routine variables.

Material and methods

The whole dataset was composed of three representative academic hospitals in China and the Medical Information Mart for Intensive Care IV (MIMIC-IV) database. Routine clinical variables were implemented for model development. The Boruta algorithm was applied for feature selection. Afterwards, ensemble learning and eight other conventional algorithms were used for model fitting and validation based on all features and selected features. The area under the receiver operating characteristic curves (ROC AUC), sensitivity, specificity, F1 score, accuracy, net reclassification index (NRI), integrated discrimination improvement (IDI), decision curve analysis (DCA), and calibration curves were used for model evaluation.

Results

A total of 955 patients undergoing abdominal surgery were finally analyzed (sepsis: 285, non-sepsis: 670). After feature selection, the ensemble learning model constructed by integrating k-Nearest Neighbor (KNN) and Support Vector Machine (SVM) yielded the ROC AUC of 0.892 (0.841-0.944) and accuracy of 85.0% on the test data, and the ROC AUC of 0.782 (0.727-0.838) and accuracy of 68.1% on the validation data, which performed best. Albumin, ASA score, neutrophil-lymphocyte ratio, age, and glucose were the top features associated with postoperative sepsis by KNN and SVM.

Conclusions

We developed a new and potential generalizable model to preoperatively screen patients at high risk of sepsis after abdominal surgery, with the advantages of a representative training cohort and routine variables.

Association of Sociodemographic Factors With Overtriage, Undertriage, and Value of Care After Major Surgery

Objective

To determine whether certain patients are vulnerable to errant triage decisions immediately after major surgery and whether there are unique sociodemographic phenotypes within overtriaged and undertriaged cohorts.

Background

In a fair system, overtriage of low-acuity patients to intensive care units (ICUs) and undertriage of high-acuity patients to general wards would affect all sociodemographic subgroups equally.

Methods

This multicenter, longitudinal cohort study of hospital admissions immediately after major surgery compared hospital mortality and value of care (risk-adjusted mortality/total costs) across 4 cohorts: overtriage (N = 660), risk-matched overtriage controls admitted to general wards (N = 3077), undertriage (N = 2335), and risk-matched undertriage controls admitted to ICUs (N = 4774). K-means clustering identified sociodemographic phenotypes within overtriage and undertriage cohorts.

Results

Compared with controls, overtriaged admissions had a predominance of male patients (56.2% vs 43.1%, P < 0.001) and commercial insurance (6.4% vs 2.5%, P < 0.001); undertriaged admissions had a predominance of Black patients (28.4% vs 24.4%, P < 0.001) and greater socioeconomic deprivation. Overtriage was associated with increased total direct costs [$16.2K ($11.4K-$23.5K) vs $14.1K ($9.1K-$20.7K), P < 0.001] and low value of care; undertriage was associated with increased hospital mortality (1.5% vs 0.7%, P = 0.002) and hospice care (2.2% vs 0.6%, P < 0.001) and low value of care. Unique sociodemographic phenotypes within both overtriage and undertriage cohorts had similar outcomes and value of care, suggesting that triage decisions, rather than patient characteristics, drive outcomes and value of care.

Conclusions

Postoperative triage decisions should ensure equality across sociodemographic groups by anchoring triage decisions to objective patient acuity assessments, circumventing cognitive shortcuts and mitigating bias.

Effect of Machine Learning on Anaesthesiology Clinician Prediction of Postoperative Complications: The Perioperative ORACLE Randomised Clinical Trial

Background

Anaesthesiology clinicians can implement risk mitigation strategies if they know which patients are at greatest risk for postoperative complications. Although machine learning models predicting complications exist, their impact on clinician risk assessment is unknown.

Methods

This single-centre randomised clinical trial enrolled patients age ≥18 undergoing surgery with anaesthesiology services. Anaesthesiology clinicians providing remote intraoperative telemedicine support reviewed electronic health records with (assisted group) or without (unassisted group) also reviewing machine learning predictions. Clinicians predicted the likelihood of postoperative 30-day all-cause mortality and postoperative acute kidney injury within 7 days. Area under the receiver operating characteristic curve (AUROC) for the clinician predictions was determined.

Results

Among 5,071 patient cases reviewed by 89 clinicians, the observed incidence was 2% for postoperative death and 11% for acute kidney injury. Clinician predictions agreed with the models more strongly in the assisted versus unassisted group (weighted kappa 0.75 versus 0.62 for death [difference 0.13, 95%CI 0.10-0.17] and 0.79 versus 0.54 for kidney injury [difference 0.25, 95%CI 0.21-0.29]). Clinicians predicted death with AUROC of 0.793 in the assisted group and 0.780 in the unassisted group (difference 0.013, 95%CI -0.070 to 0.097). Clinicians predicted kidney injury with AUROC of 0.734 in the assisted group and 0.688 in the unassisted group (difference 0.046, 95%CI -0.003 to 0.091).

Conclusions

Although there was evidence that the models influenced clinician predictions, clinician performance was not statistically significantly different with and without machine learning assistance. Further work is needed to clarify the role of machine learning in real-time perioperative risk stratification.

Trial Registration

ClinicalTrials.gov NCT05042804.

Artificial intelligence in surgery

Artificial intelligence (AI) is rapidly emerging in healthcare, yet applications in surgery remain relatively nascent. Here we review the integration of AI in the field of surgery, centering our discussion on multifaceted improvements in surgical care in the preoperative, intraoperative and postoperative space. The emergence of foundation model architectures, wearable technologies and improving surgical data infrastructures is enabling rapid advances in AI interventions and utility. We discuss how maturing AI methods hold the potential to improve patient outcomes, facilitate surgical education and optimize surgical care. We review the current applications of deep learning approaches and outline a vision for future advances through multimodal foundation models.

Development of interpretable machine learning models for prediction of acute kidney injury after noncardiac surgery: a retrospective cohort study

BACKGROUND

Early identification of patients at high-risk of postoperative acute kidney injury (AKI) can facilitate the development of preventive approaches. This study aimed to develop prediction models for postoperative AKI in noncardiac surgery using machine learning algorithms. The authors also evaluated the predictive performance of models that included only preoperative variables or only important predictors.

MATERIALS AND METHODS

Adult patients undergoing noncardiac surgery were retrospectively included in the study (76 457 patients in the discovery cohort and 11 910 patients in the validation cohort). AKI was determined using the KDIGO criteria. The prediction model was developed using 87 variables (56 preoperative variables and 31 intraoperative variables). A variety of machine learning algorithms were employed to develop the model, including logistic regression, random forest, extreme gradient boosting, and gradient boosting decision trees. The performance of different models was compared using the area under the receiver operating characteristic curve (AUROC). Shapley Additive Explanations (SHAP) analysis was employed for model interpretation.

RESULTS

The patients in the discovery cohort had a median age of 52 years (IQR: 42-61 years), and 1179 patients (1.5%) developed AKI after surgery. The gradient boosting decision trees algorithm showed the best predictive performance using all available variables, or only preoperative variables. The AUROCs were 0.849 (95% CI: 0.835-0.863) and 0.828 (95% CI: 0.813-0.843), respectively. The SHAP analysis showed that age, surgical duration, preoperative serum creatinine, and gamma-glutamyltransferase, as well as American Society of Anesthesiologists physical status III were the most important five features. When gradually reducing the features, the AUROCs decreased from 0.852 (including the top 40 features) to 0.839 (including the top 10 features). In the validation cohort, the authors observed a similar pattern regarding the models' predictive performance.

CONCLUSIONS

The machine learning models the authors developed had satisfactory predictive performance for identifying high-risk postoperative AKI patients. Furthermore, the authors found that model performance was only slightly affected when only preoperative variables or only the most important predictive features were included.

Explainable Machine Learning Model to Preoperatively Predict Postoperative Complications in Inpatients With Cancer Undergoing Major Operations

PURPOSE

Preoperative prediction of postoperative complications (PCs) in inpatients with cancer is challenging. We developed an explainable machine learning (ML) model to predict PCs in a heterogenous population of inpatients with cancer undergoing same-hospitalization major operations.

METHODS

Consecutive inpatients who underwent same-hospitalization operations from December 2017 to June 2021 at a single institution were retrospectively reviewed. The ML model was developed and tested using electronic health record (EHR) data to predict 30-day PCs for patients with Clavien-Dindo grade 3 or higher (CD 3+) per the CD classification system. Model performance was assessed using area under the receiver operating characteristic curve (AUROC), area under the precision recall curve (AUPRC), and calibration plots. Model explanation was performed using the Shapley additive explanations (SHAP) method at cohort and individual operation levels.

RESULTS

A total of 988 operations in 827 inpatients were included. The ML model was trained using 788 operations and tested using a holdout set of 200 operations. The CD 3+ complication rates were 28.6% and 27.5% in the training and holdout test sets, respectively. Training and holdout test sets' model performance in predicting CD 3+ complications yielded an AUROC of 0.77 and 0.73 and an AUPRC of 0.56 and 0.52, respectively. Calibration plots demonstrated good reliability. The SHAP method identified features and the contributions of the features to the risk of PCs.

CONCLUSION

We trained and tested an explainable ML model to predict the risk of developing PCs in patients with cancer. Using patient-specific EHR data, the ML model accurately discriminated the risk of developing CD 3+ complications and displayed top features at the individual operation and cohort level.

ChatGPT Yields a Passing Score on a Pediatric Board Preparatory Exam but Raises Red Flags

Objectives

We aimed to evaluate the performance of a publicly-available online artificial intelligence program (OpenAI's ChatGPT-3.5 and -4.0, August 3 versions) on a pediatric board preparatory examination, 2021 and 2022 PREP® Self-Assessment, American Academy of Pediatrics (AAP).

Methods

We entered 245 questions and answer choices from the Pediatrics 2021 PREP® Self-Assessment and 247 questions and answer choices from the Pediatrics 2022 PREP® Self-Assessment into OpenAI's ChatGPT-3.5 and ChatGPT-4.0, August 3 versions, in September 2023. The ChatGPT-3.5 and 4.0 scores were compared with the advertised passing scores (70%+) for the PREP® exams and the average scores (74.09%) and (75.71%) for all 10 715 and 6825 first-time human test takers.

Results

For the AAP 2021 and 2022 PREP® Self-Assessments, ChatGPT-3.5 answered 143 of 243 (58.85%) and 137 of 247 (55.46%) questions correctly on a single attempt. ChatGPT-4.0 answered 193 of 243 (79.84%) and 208 of 247 (84.21%) questions correctly.

Conclusion

Using a publicly-available online chatbot to answer pediatric board preparatory examination questions yielded a passing score but demonstrated significant limitations in the chatbot's ability to assess some complex medical situations in children, posing a potential risk to this vulnerable population.

Risk Perception, Acceptance, and Trust of Using AI in Gastroenterology Practice in the Asia-Pacific Region: Web-Based Survey Study

BACKGROUND

The use of artificial intelligence (AI) can revolutionize health care, but this raises risk concerns. It is therefore crucial to understand how clinicians trust and accept AI technology. Gastroenterology, by its nature of being an image-based and intervention-heavy specialty, is an area where AI-assisted diagnosis and management can be applied extensively.

OBJECTIVE

This study aimed to study how gastroenterologists or gastrointestinal surgeons accept and trust the use of AI in computer-aided detection (CADe), computer-aided characterization (CADx), and computer-aided intervention (CADi) of colorectal polyps in colonoscopy.

METHODS

We conducted a web-based questionnaire from November 2022 to January 2023, involving 5 countries or areas in the Asia-Pacific region. The questionnaire included variables such as background and demography of users; intention to use AI, perceived risk; acceptance; and trust in AI-assisted detection, characterization, and intervention. We presented participants with 3 AI scenarios related to colonoscopy and the management of colorectal polyps. These scenarios reflect existing AI applications in colonoscopy, namely the detection of polyps (CADe), characterization of polyps (CADx), and AI-assisted polypectomy (CADi).

RESULTS

In total, 165 gastroenterologists and gastrointestinal surgeons responded to a web-based survey using the structured questionnaire designed by experts in medical communications. Participants had a mean age of 44 (SD 9.65) years, were mostly male (n=116, 70.3%), and mostly worked in publicly funded hospitals (n=110, 66.67%). Participants reported relatively high exposure to AI, with 111 (67.27%) reporting having used AI for clinical diagnosis or treatment of digestive diseases. Gastroenterologists are highly interested to use AI in diagnosis but show different levels of reservations in risk prediction and acceptance of AI. Most participants (n=112, 72.72%) also expressed interest to use AI in their future practice. CADe was accepted by 83.03% (n=137) of respondents, CADx was accepted by 78.79% (n=130), and CADi was accepted by 72.12% (n=119). CADe and CADx were trusted by 85.45% (n=141) of respondents and CADi was trusted by 72.12% (n=119). There were no application-specific differences in risk perceptions, but more experienced clinicians gave lesser risk ratings.

CONCLUSIONS

Gastroenterologists reported overall high acceptance and trust levels of using AI-assisted colonoscopy in the management of colorectal polyps. However, this level of trust depends on the application scenario. Moreover, the relationship among risk perception, acceptance, and trust in using AI in gastroenterology practice is not straightforward.

Prediction of Pathologic Complete Response for Rectal Cancer Based on Pretreatment Factors Using Machine Learning

BACKGROUND

Pathologic complete response after neoadjuvant therapy is an important prognostic indicator for locally advanced rectal cancer and may give insights into which patients might be treated nonoperatively in the future. Existing models for predicting pathologic complete response in the pretreatment setting are limited by small data sets and low accuracy.

OBJECTIVE

We sought to use machine learning to develop a more generalizable predictive model for pathologic complete response for locally advanced rectal cancer.

DESIGN

Patients with locally advanced rectal cancer who underwent neoadjuvant therapy followed by surgical resection were identified in the National Cancer Database from years 2010 to 2019 and were split into training, validation, and test sets. Machine learning techniques included random forest, gradient boosting, and artificial neural network. A logistic regression model was also created. Model performance was assessed using an area under the receiver operating characteristic curve.

SETTINGS

This study used a national, multicenter data set.

PATIENTS

Patients with locally advanced rectal cancer who underwent neoadjuvant therapy and proctectomy.

MAIN OUTCOME MEASURES

Pathologic complete response defined as T0/xN0/x.

RESULTS

The data set included 53,684 patients. Pathologic complete response was experienced by 22.9% of patients. Gradient boosting showed the best performance with an area under the receiver operating characteristic curve of 0.777 (95% CI, 0.773-0.781), compared with 0.684 (95% CI, 0.68-0.688) for logistic regression. The strongest predictors of pathologic complete response were no lymphovascular invasion, no perineural invasion, lower CEA, smaller size of tumor, and microsatellite stability. A concise model including the top 5 variables showed preserved performance.

LIMITATIONS

The models were not externally validated.

CONCLUSIONS

Machine learning techniques can be used to accurately predict pathologic complete response for locally advanced rectal cancer in the pretreatment setting. After fine-tuning a data set including patients treated nonoperatively, these models could help clinicians identify the appropriate candidates for a watch-and-wait strategy. See Video Abstract .

EL CNCER DE RECTO BASADA EN FACTORES PREVIOS AL TRATAMIENTO MEDIANTE EL APRENDIZAJE AUTOMTICO

ANTECEDENTES:La respuesta patológica completa después de la terapia neoadyuvante es un indicador pronóstico importante para el cáncer de recto localmente avanzado y puede dar información sobre qué pacientes podrían ser tratados de forma no quirúrgica en el futuro. Los modelos existentes para predecir la respuesta patológica completa en el entorno previo al tratamiento están limitados por conjuntos de datos pequeños y baja precisión.OBJETIVO:Intentamos utilizar el aprendizaje automático para desarrollar un modelo predictivo más generalizable para la respuesta patológica completa para el cáncer de recto localmente avanzado.DISEÑO:Los pacientes con cáncer de recto localmente avanzado que se sometieron a terapia neoadyuvante seguida de resección quirúrgica se identificaron en la Base de Datos Nacional del Cáncer de los años 2010 a 2019 y se dividieron en conjuntos de capacitación, validación y prueba. Las técnicas de aprendizaje automático incluyeron bosque aleatorio, aumento de gradiente y red neuronal artificial. También se creó un modelo de regresión logística. El rendimiento del modelo se evaluó utilizando el área bajo la curva característica operativa del receptor.ÁMBITO:Este estudio utilizó un conjunto de datos nacional multicéntrico.PACIENTES:Pacientes con cáncer de recto localmente avanzado sometidos a terapia neoadyuvante y proctectomía.PRINCIPALES MEDIDAS DE VALORACIÓN:Respuesta patológica completa definida como T0/xN0/x.RESULTADOS:El conjunto de datos incluyó 53.684 pacientes. El 22,9% de los pacientes experimentaron una respuesta patológica completa. El refuerzo de gradiente mostró el mejor rendimiento con un área bajo la curva característica operativa del receptor de 0,777 (IC del 95%: 0,773 - 0,781), en comparación con 0,684 (IC del 95%: 0,68 - 0,688) para la regresión logística. Los predictores más fuertes de respuesta patológica completa fueron la ausencia de invasión linfovascular, la ausencia de invasión perineural, un CEA más bajo, un tamaño más pequeño del tumor y la estabilidad de los microsatélites. Un modelo conciso que incluye las cinco variables principales mostró un rendimiento preservado.LIMITACIONES:Los modelos no fueron validados externamente.CONCLUSIONES:Las técnicas de aprendizaje automático se pueden utilizar para predecir con precisión la respuesta patológica completa para el cáncer de recto localmente avanzado en el entorno previo al tratamiento. Después de realizar ajustes en un conjunto de datos que incluye pacientes tratados de forma no quirúrgica, estos modelos podrían ayudar a los médicos a identificar a los candidatos adecuados para una estrategia de observar y esperar. (Traducción-Dr. Ingrid Melo ).

Prediction models for postoperative delirium in elderly patients with machine-learning algorithms and SHapley Additive exPlanations

Postoperative delirium (POD) is a common and severe complication in elderly patients with hip fractures. Identifying high-risk patients with POD can help improve the outcome of patients with hip fractures. We conducted a retrospective study on elderly patients (≥65 years of age) who underwent orthopedic surgery with hip fracture between January 2014 and August 2019. Conventional logistic regression and five machine-learning algorithms were used to construct prediction models of POD. A nomogram for POD prediction was built with the logistic regression method. The area under the receiver operating characteristic curve (AUC-ROC), accuracy, sensitivity, and precision were calculated to evaluate different models. Feature importance of individuals was interpreted using Shapley Additive Explanations (SHAP). About 797 patients were enrolled in the study, with the incidence of POD at 9.28% (74/797). The age, renal insufficiency, chronic obstructive pulmonary disease (COPD), use of antipsychotics, lactate dehydrogenase (LDH), and C-reactive protein are used to build a nomogram for POD with an AUC of 0.71. The AUCs of five machine-learning models are 0.81 (Random Forest), 0.80 (GBM), 0.68 (AdaBoost), 0.77 (XGBoost), and 0.70 (SVM). The sensitivities of the six models range from 68.8% (logistic regression and SVM) to 91.9% (Random Forest). The precisions of the six machine-learning models range from 18.3% (logistic regression) to 67.8% (SVM). Six prediction models of POD in patients with hip fractures were constructed using logistic regression and five machine-learning algorithms. The application of machine-learning algorithms could provide convenient POD risk stratification to benefit elderly hip fracture patients.

Development and validation of an interpretable Markov-embedded multilabel model for predicting risks of multiple postoperative complications among surgical inpatients: a multicenter prospective cohort study

BACKGROUND

When they encounter various highly related postoperative complications, existing risk evaluation tools that focus on single or any complications are inadequate in clinical practice. This seriously hinders complication management because of the lack of a quantitative basis. An interpretable multilabel model framework that predicts multiple complications simultaneously is urgently needed.

MATERIALS AND METHODS

The authors included 50 325 inpatients from a large multicenter cohort (2014-2017). The authors separated patients from one hospital for external validation and randomly split the remaining patients into training and internal validation sets. A MARKov-EmbeDded (MARKED) multilabel model was proposed, and three models were trained for comparison: binary relevance, a fully connected network (FULLNET), and a deep neural network. Performance was mainly evaluated using the area under the receiver operating characteristic curve (AUC). The authors interpreted the model using Shapley Additive Explanations. Complication-specific risk and risk source inference were provided at the individual level.

RESULTS

There were 26 292, 6574, and 17 459 inpatients in the training, internal validation, and external validation sets, respectively. For the external validation set, MARKED achieved the highest average AUC (0.818, 95% CI: 0.771-0.864) across eight outcomes [compared with binary relevance, 0.799 (0.748-0.849), FULLNET, 0.806 (0.756-0.856), and deep neural network, 0.815 (0.765-0.866)]. Specifically, the AUCs of MARKED were above 0.9 for cardiac complications [0.927 (0.894-0.960)], neurological complications [0.905 (0.870-0.941)], and mortality [0.902 (0.867-0.937)]. Serum albumin, surgical specialties, emergency case, American Society of Anesthesiologists score, age, and sex were the six most important preoperative variables. The interaction between complications contributed more than the preoperative variables, and formed a hierarchical chain of risk factors, mild complications, and severe complications.

CONCLUSION

The authors demonstrated the advantage of MARKED in terms of performance and interpretability. The authors expect that the identification of high-risk patients and the inference of the risk source for specific complications will be valuable for clinical decision-making.

How anesthesiology can deal with innovation and new technologies?

Innovation and new technologies have always impacted significantly the anesthesiology practice all along the perioperative course, as it is recognized as one of the most transformative medical specialties specifically regarding patient's safety. Beside a number of major changes in procedures, equipment, training, and organization that aggregated to establish a strong safety culture with effective practices, anesthesiology is also a stakeholder in disruptive innovation. The present review is not exhaustive and aims to provide an overview on how innovation could change and improve anesthesiology practices through some examples as telemedicine (TM), machine learning and artificial intelligence (AI). For example, postoperative complications can be accurately predicted by AI from automated real-time electronic health record data, matching physicians' predictive accuracy. Clinical workflow could be facilitated and accelerated with mobile devices and applications, assuming that these tools should remain at the service of patients and care providers. Care providers and patients connections have improved, thanks to these digital and innovative transformations, without replacing existing relationships between them. It also should give time back to physicians and nurses to better spend it in the perioperative care, and to provide "personalized" medicine keeping a high level of standard of care.

Modeling the influence of attitudes, trust, and beliefs on endoscopists' acceptance of artificial intelligence applications in medical practice

Introduction

The potential for deployment of Artificial Intelligence (AI) technologies in various fields of medicine is vast, yet acceptance of AI amongst clinicians has been patchy. This research therefore examines the role of antecedents, namely trust, attitude, and beliefs in driving AI acceptance in clinical practice.

Methods

We utilized online surveys to gather data from clinicians in the field of gastroenterology.

Results

A total of 164 participants responded to the survey. Participants had a mean age of 44.49 (SD = 9.65). Most participants were male (n = 116, 70.30%) and specialized in gastroenterology (n = 153, 92.73%). Based on the results collected, we proposed and tested a model of AI acceptance in medical practice. Our findings showed that while the proposed drivers had a positive impact on AI tools' acceptance, not all effects were direct. Trust and belief were found to fully mediate the effects of attitude on AI acceptance by clinicians.

Discussion

The role of trust and beliefs as primary mediators of the acceptance of AI in medical practice suggest that these should be areas of focus in AI education, engagement and training. This has implications for how AI systems can gain greater clinician acceptance to engender greater trust and adoption amongst public health systems and professional networks which in turn would impact how populations interface with AI. Implications for policy and practice, as well as future research in this nascent field, are discussed.

Risk factors of major bleeding detected by machine learning method in patients undergoing liver resection with controlled low central venous pressure technique

BACKGROUND

Controlled low central venous pressure (CLCVP) technique has been extensively validated in clinical practices to decrease intraoperative bleeding during liver resection process; however, no studies to date have attempted to propose a scoring method to better understand what risk factors might still be responsible for bleeding when CLCVP technique was implemented.

METHODS

We aimed to use machine learning to develop a model for detecting the risk factors of major bleeding in patients who underwent liver resection using CLCVP technique. We reviewed the medical records of 1077 patients who underwent liver surgery between January 2017 and June 2020. We evaluated the XGBoost model and logistic regression model using stratified K-fold cross-validation (K = 5), and the area under the receiver operating characteristic curve, the recall rate, precision rate, and accuracy score were calculated and compared. The SHapley Additive exPlanations was employed to identify the most influencing factors and their contribution to the prediction.

RESULTS

The XGBoost classifier with an accuracy of 0.80 and precision of 0.89 outperformed the logistic regression model with an accuracy of 0.76 and precision of 0.79. According to the SHapley Additive exPlanations summary plot, the top six variables ranked from most to least important included intraoperative hematocrit, surgery duration, intraoperative lactate, preoperative hemoglobin, preoperative aspartate transaminase, and Pringle maneuver duration.

CONCLUSIONS

Anesthesiologists should be aware of the potential impact of increased Pringle maneuver duration and lactate levels on intraoperative major bleeding in patients undergoing liver resection with CLCVP technique.   What is already known on this topic-Low central venous pressure technique has already been extensively validated in clinical practices, with no prediction model for major bleeding. What this study adds-The XGBoost classifier outperformed logistic regression model for the prediction of major bleeding during liver resection with low central venous pressure technique. How this study might affect research, practice, or policy-anesthesiologists should be aware of the potential impact of increased PM duration and lactate levels on intraoperative major bleeding in patients undergoing liver resection with CLCVP technique.

Temporal Generalizability of Machine Learning Models for Predicting Postoperative Delirium Using Electronic Health Record Data: Model Development and Validation Study

BACKGROUND

Although machine learning models demonstrate significant potential in predicting postoperative delirium, the advantages of their implementation in real-world settings remain unclear and require a comparison with conventional models in practical applications.

OBJECTIVE

The objective of this study was to validate the temporal generalizability of decision tree ensemble and sparse linear regression models for predicting delirium after surgery compared with that of the traditional logistic regression model.

METHODS

The health record data of patients hospitalized at an advanced emergency and critical care medical center in Kumamoto, Japan, were collected electronically. We developed a decision tree ensemble model using extreme gradient boosting (XGBoost) and a sparse linear regression model using least absolute shrinkage and selection operator (LASSO) regression. To evaluate the predictive performance of the model, we used the area under the receiver operating characteristic curve (AUROC) and the Matthews correlation coefficient (MCC) to measure discrimination and the slope and intercept of the regression between predicted and observed probabilities to measure calibration. The Brier score was evaluated as an overall performance metric. We included 11,863 consecutive patients who underwent surgery with general anesthesia between December 2017 and February 2022. The patients were divided into a derivation cohort before the COVID-19 pandemic and a validation cohort during the COVID-19 pandemic. Postoperative delirium was diagnosed according to the confusion assessment method.

RESULTS

A total of 6497 patients (68.5, SD 14.4 years, women n=2627, 40.4%) were included in the derivation cohort, and 5366 patients (67.8, SD 14.6 years, women n=2105, 39.2%) were included in the validation cohort. Regarding discrimination, the XGBoost model (AUROC 0.87-0.90 and MCC 0.34-0.44) did not significantly outperform the LASSO model (AUROC 0.86-0.89 and MCC 0.34-0.41). The logistic regression model (AUROC 0.84-0.88, MCC 0.33-0.40, slope 1.01-1.19, intercept -0.16 to 0.06, and Brier score 0.06-0.07), with 8 predictors (age, intensive care unit, neurosurgery, emergency admission, anesthesia time, BMI, blood loss during surgery, and use of an ambulance) achieved good predictive performance.

CONCLUSIONS

The XGBoost model did not significantly outperform the LASSO model in predicting postoperative delirium. Furthermore, a parsimonious logistic model with a few important predictors achieved comparable performance to machine learning models in predicting postoperative delirium.

Predictive Modeling Using Artificial Intelligence and Machine Learning Algorithms on Electronic Health Record Data: Advantages and Challenges

The rapid adoption of electronic health record (EHR) systems in US hospitals from 2008 to 2014 produced novel data elements for analysis. Concurrent innovations in computing architecture and machine learning (ML) algorithms have made rapid consumption of health data feasible and a powerful engine for clinical innovation. In critical care research, the net convergence of these trends has resulted in an exponential increase in outcome prediction research. In the following article, we explore the history of outcome prediction in the intensive care unit (ICU), the growing use of EHR data, and the rise of artificial intelligence and ML (AI) in critical care.

Identifying Effective Biomarkers for Accurate Pancreatic Cancer Prognosis Using Statistical Machine Learning

Pancreatic cancer (PC) has one of the lowest survival rates among all major types of cancer. Consequently, it is one of the leading causes of mortality worldwide. Serum biomarkers historically correlate well with the early prognosis of post-surgical complications of PC. However, attempts to identify an effective biomarker panel for the successful prognosis of PC were almost non-existent in the current literature. The current study investigated the roles of various serum biomarkers including carbohydrate antigen 19-9 (CA19-9), chemokine (C-X-C motif) ligand 8 (CXCL-8), procalcitonin (PCT), and other relevant clinical data for identifying PC progression, classified into sepsis, recurrence, and other post-surgical complications, among PC patients. The most relevant biochemical and clinical markers for PC prognosis were identified using a random-forest-powered feature elimination method. Using this informative biomarker panel, the selected machine-learning (ML) classification models demonstrated highly accurate results for classifying PC patients into three complication groups on independent test data. The superiority of the combined biomarker panel (Max AUC-ROC = 100%) was further established over using CA19-9 features exclusively (Max AUC-ROC = 75%) for the task of classifying PC progression. This novel study demonstrates the effectiveness of the combined biomarker panel in successfully diagnosing PC progression and other relevant complications among Egyptian PC survivors.

Computer-aided shape features extraction and regression models for predicting the ascending aortic aneurysm growth rate

OBJECTIVE

ascending aortic aneurysm growth prediction is still challenging in clinics. In this study, we evaluate and compare the ability of local and global shape features to predict the ascending aortic aneurysm growth.

MATERIAL AND METHODS

70 patients with aneurysm, for which two 3D acquisitions were available, are included. Following segmentation, three local shape features are computed: (1) the ratio between maximum diameter and length of the ascending aorta centerline, (2) the ratio between the length of external and internal lines on the ascending aorta and (3) the tortuosity of the ascending tract. By exploiting longitudinal data, the aneurysm growth rate is derived. Using radial basis function mesh morphing, iso-topological surface meshes are created. Statistical shape analysis is performed through unsupervised principal component analysis (PCA) and supervised partial least squares (PLS). Two types of global shape features are identified: three PCA-derived and three PLS-based shape modes. Three regression models are set for growth prediction: two based on gaussian support vector machine using local and PCA-derived global shape features; the third is a PLS linear regression model based on the related global shape features. The prediction results are assessed and the aortic shapes most prone to growth are identified.

RESULTS

the prediction root mean square error from leave-one-out cross-validation is: 0.112 mm/month, 0.083 mm/month and 0.066 mm/month for local, PCA-based and PLS-derived shape features, respectively. Aneurysms close to the root with a large initial diameter report faster growth.

CONCLUSION

global shape features might provide an important contribution for predicting the aneurysm growth.

Artificial Intelligence-enabled Decision Support in Surgery: State-of-the-art and Future Directions

OBJECTIVE

To summarize state-of-the-art artificial intelligence-enabled decision support in surgery and to quantify deficiencies in scientific rigor and reporting.

BACKGROUND

To positively affect surgical care, decision-support models must exceed current reporting guideline requirements by performing external and real-time validation, enrolling adequate sample sizes, reporting model precision, assessing performance across vulnerable populations, and achieving clinical implementation; the degree to which published models meet these criteria is unknown.

METHODS

Embase, PubMed, and MEDLINE databases were searched from their inception to September 21, 2022 for articles describing artificial intelligence-enabled decision support in surgery that uses preoperative or intraoperative data elements to predict complications within 90 days of surgery. Scientific rigor and reporting criteria were assessed and reported according to Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews guidelines.

RESULTS

Sample size ranged from 163-2,882,526, with 8/36 articles (22.2%) featuring sample sizes of less than 2000; 7 of these 8 articles (87.5%) had below-average (<0.83) area under the receiver operating characteristic or accuracy. Overall, 29 articles (80.6%) performed internal validation only, 5 (13.8%) performed external validation, and 2 (5.6%) performed real-time validation. Twenty-three articles (63.9%) reported precision. No articles reported performance across sociodemographic categories. Thirteen articles (36.1%) presented a framework that could be used for clinical implementation; none assessed clinical implementation efficacy.

CONCLUSIONS

Artificial intelligence-enabled decision support in surgery is limited by reliance on internal validation, small sample sizes that risk overfitting and sacrifice predictive performance, and failure to report confidence intervals, precision, equity analyses, and clinical implementation. Researchers should strive to improve scientific quality.

Quantifying the Prognostic Value of Preoperative Surgeon Intuition: Comparing Surgeon Intuition and Clinical Risk Prediction as Derived from the American College of Surgeons NSQIP Risk Calculator

BACKGROUND

Surgical risk prediction models traditionally use patient attributes and measures of physiology to generate predictions about postoperative outcomes. However, the surgeon's assessment of the patient may be a valuable predictor, given the surgeon's ability to detect and incorporate factors that existing models cannot capture. We compare the predictive utility of surgeon intuition and a risk calculator derived from the American College of Surgeons (ACS) NSQIP.

STUDY DESIGN

From January 10, 2021 to January 9, 2022, surgeons were surveyed immediately before performing surgery to assess their perception of a patient's risk of developing any postoperative complication. Clinical data were abstracted from ACS NSQIP. Both sources of data were independently used to build models to predict the likelihood of a patient experiencing any 30-day postoperative complication as defined by ACS NSQIP.

RESULTS

Preoperative surgeon assessment was obtained for 216 patients. NSQIP data were available for 9,182 patients who underwent general surgery (January 1, 2017 to January 9, 2022). A binomial regression model trained on clinical data alone had an area under the receiver operating characteristic curve (AUC) of 0.83 (95% CI 0.80 to 0.85) in predicting any complication. A model trained on only preoperative surgeon intuition had an AUC of 0.70 (95% CI 0.63 to 0.78). A model trained on surgeon intuition and a subset of clinical predictors had an AUC of 0.83 (95% CI 0.77 to 0.89).

CONCLUSIONS

Preoperative surgeon intuition alone is an independent predictor of patient outcomes; however, a risk calculator derived from ACS NSQIP is a more robust predictor of postoperative complication. Combining intuition and clinical data did not strengthen prediction.

The digital transformation of surgery

Rapid advances in digital technology and artificial intelligence in recent years have already begun to transform many industries, and are beginning to make headway into healthcare. There is tremendous potential for new digital technologies to improve the care of surgical patients. In this piece, we highlight work being done to advance surgical care using machine learning, computer vision, wearable devices, remote patient monitoring, and virtual and augmented reality. We describe ways these technologies can be used to improve the practice of surgery, and discuss opportunities and challenges to their widespread adoption and use in operating rooms and at the bedside.

Electronic Health Record Use and Perceptions among Urologic Surgeons

OBJECTIVE

Electronic health records (EHRs) have become widely adopted with increasing emphasis on improving care delivery. Improvements in surgery may be limited by specialty-specific issues that impact EHR usability and engagement. Accordingly, we examined EHR use and perceptions in urology, a diverse surgical specialty.

METHODS

We conducted a national, sequential explanatory mixed methods study. Through the 2019 American Urological Association Census, we surveyed urologic surgeons on EHR use and perceptions and then identified associated characteristics through bivariable and multivariable analyses. Using purposeful sampling, we interviewed 25 urologists and applied coding-based thematic analysis, which was then integrated with survey findings.

RESULTS

Among 2,159 practicing urologic surgeons, 2,081 (96.4%) reported using an EHR. In the weighted sample (n = 12,366), over 90% used the EHR for charting, viewing results, and order entry with most using information exchange functions (59.0-79.6%). In contrast, only 35.8% felt the EHR increases clinical efficiency, whereas 43.1% agreed it improves patient care, which related thematically to information management, administrative burden, patient safety, and patient-surgeon interaction. Quantitatively and qualitatively, use and perceptions differed by years in practice and practice type with more use and better perceptions among more recent entrants into the urologic workforce and those in academic/multispecialty practices, who may have earlier EHR exposure, better infrastructure, and more support.

CONCLUSION

Despite wide and substantive usage, EHRs engender mixed feelings, especially among longer-practicing surgeons and those in lower-resourced settings (e.g., smaller and private practices). Beyond reducing administrative burden and simplifying information management, efforts to improve care delivery through the EHR should focus on surgeon engagement, particularly in the community, to boost implementation and user experience.

Re-Aiming Equity Evaluation in Clinical Decision Support: A Scoping Review of Equity Assessments in Surgical Decision Support Systems

OBJECTIVE

We critically evaluated the surgical literature to explore the prevalence and describe how equity assessments occur when using clinical decision support systems.

BACKGROUND

Clinical decision support (CDS) systems are increasingly used to facilitate surgical care delivery. Despite formal recommendations to do so, equity evaluations are not routinely performed on CDS systems and underrepresented populations are at risk of harm and further health disparities. We explored surgical literature to determine frequency and rigor of CDS equity assessments and offer recommendations to improve CDS equity by appending existing frameworks.

METHODS

We performed a scoping review up to Augus 25, 2021 using PubMed and Google Scholar for the following search terms: clinical decision support, implementation, RE-AIM, Proctor, Proctor's framework, equity, trauma, surgery, surgical. We identified 1415 citations and 229 abstracts met criteria for review. A total of 84 underwent full review after 145 were excluded if they did not assess outcomes of an electronic CDS tool or have a surgical use case.

RESULTS

Only 6% (5/84) of surgical CDS systems reported equity analyses, suggesting that current methods for optimizing equity in surgical CDS are inadequate. We propose revising the RE-AIM framework to include an Equity element (RE 2 -AIM) specifying that CDS foundational analyses and algorithms are performed or trained on balanced datasets with sociodemographic characteristics that accurately represent the CDS target population and are assessed by sensitivity analyses focused on vulnerable subpopulations.

CONCLUSION

Current surgical CDS literature reports little with respect to equity. Revising the RE-AIM framework to include an Equity element (RE 2 -AIM) promotes the development and implementation of CDS systems that, at minimum, do not worsen healthcare disparities and possibly improve their generalizability.