Dataset Information

Prediction Model for 30-Day Mortality after Non-Cardiac Surgery Using Machine-Learning Techniques Based on Preoperative Evaluation of Electronic Medical Records.

ABSTRACT:

Background

Machine-learning techniques are useful for creating prediction models in clinical practice. This study aimed to construct a prediction model of postoperative 30-day mortality based on an automatically extracted electronic preoperative evaluation sheet.

Methods

We used data from 276,341 consecutive adult patients who underwent non-cardiac surgery between January 2011 and December 2020 at a tertiary center for model development and internal validation, and another dataset from 63,384 patients between January 2011 and October 2021 at another center for external validation. Postoperative 30-day mortality was 0.16%. We developed an extreme gradient boosting (XGB) prediction model using only variables from preoperative evaluation sheets.

Results

The model yielded an area under the curve of 0.960 and an area under the precision and recall curve of 0.216, which were 0.932 and 0.122, respectively, in the external validation set. The optimal threshold calculated by Youden's J statistic had a sensitivity of 0.885 and specificity of 0.914. In an additional analysis with balanced distribution, the model showed a similar predictive value.

Conclusion

We presented a machine-learning prediction model for 30-day mortality after non-cardiac surgery using preoperative variables automatically extracted from electronic medical records and validated the model in a multi-center setting. Our model may help clinicians predict postoperative outcomes.

SUBMITTER: Choi B

PROVIDER: S-EPMC9659244 | biostudies-literature | 2022 Nov

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Prediction Model for 30-Day Mortality after Non-Cardiac Surgery Using Machine-Learning Techniques Based on Preoperative Evaluation of Electronic Medical Records.

Choi Byungjin B Oh Ah Ran AR Lee Seung-Hwa SH Lee Dong Yun DY Lee Jong-Hwan JH Yang Kwangmo K Kim Ha Yeon HY Park Rae Woong RW Park Jungchan J

Journal of clinical medicine 20221101 21

<h4>Background</h4>Machine-learning techniques are useful for creating prediction models in clinical practice. This study aimed to construct a prediction model of postoperative 30-day mortality based on an automatically extracted electronic preoperative evaluation sheet.<h4>Methods</h4>We used data from 276,341 consecutive adult patients who underwent non-cardiac surgery between January 2011 and December 2020 at a tertiary center for model development and internal validation, and another dataset ...[more]

PMID: 36362715

Similar Datasets

Project description:BackgroundHeart failure is one of the leading causes of hospitalization in the United States. Advances in big data solutions allow for storage, management, and mining of large volumes of structured and semi-structured data, such as complex healthcare data. Applying these advances to complex healthcare data has led to the development of risk prediction models to help identify patients who would benefit most from disease management programs in an effort to reduce readmissions and healthcare cost, but the results of these efforts have been varied. The primary aim of this study was to develop a 30-day readmission risk prediction model for heart failure patients discharged from a hospital admission.MethodsWe used longitudinal electronic medical record data of heart failure patients admitted within a large healthcare system. Feature vectors included structured demographic, utilization, and clinical data, as well as selected extracts of un-structured data from clinician-authored notes. The risk prediction model was developed using deep unified networks (DUNs), a new mesh-like network structure of deep learning designed to avoid over-fitting. The model was validated with 10-fold cross-validation and results compared to models based on logistic regression, gradient boosting, and maxout networks. Overall model performance was assessed using concordance statistic. We also selected a discrimination threshold based on maximum projected cost saving to the Partners Healthcare system.ResultsData from 11,510 patients with 27,334 admissions and 6369 30-day readmissions were used to train the model. After data processing, the final model included 3512 variables. The DUNs model had the best performance after 10-fold cross-validation. AUCs for prediction models were 0.664 ± 0.015, 0.650 ± 0.011, 0.695 ± 0.016 and 0.705 ± 0.015 for logistic regression, gradient boosting, maxout networks, and DUNs respectively. The DUNs model had an accuracy of 76.4% at the classification threshold that corresponded with maximum cost saving to the hospital.ConclusionsDeep learning techniques performed better than other traditional techniques in developing this EMR-based prediction model for 30-day readmissions in heart failure patients. Such models can be used to identify heart failure patients with impending hospitalization, enabling care teams to target interventions at their most high-risk patients and improving overall clinical outcomes.

Project description:BackgroundAlthough there is a growing interest in prediction models based on electronic medical records (EMRs) to identify patients at risk of adverse cardiac events following invasive coronary treatment, robust models fully utilizing EMR data are limited.ObjectiveWe aimed to develop and validate machine learning (ML) models by using diverse fields of EMR to predict the risk of 30-day adverse cardiac events after percutaneous intervention or bypass surgery.MethodsEMR data of 5,184,565 records of 16,793 patients at a quaternary hospital between 2006 and 2016 were categorized into static basic (eg, demographics), dynamic time-series (eg, laboratory values), and cardiac-specific data (eg, coronary angiography). The data were randomly split into training, tuning, and testing sets in a ratio of 3:1:1. Each model was evaluated with 5-fold cross-validation and with an external EMR-based cohort at a tertiary hospital. Logistic regression (LR), random forest (RF), gradient boosting machine (GBM), and feedforward neural network (FNN) algorithms were applied. The primary outcome was 30-day mortality following invasive treatment.ResultsGBM showed the best performance with area under the receiver operating characteristic curve (AUROC) of 0.99; RF had a similar AUROC of 0.98. AUROCs of FNN and LR were 0.96 and 0.93, respectively. GBM had the highest area under the precision-recall curve (AUPRC) of 0.80, and the AUPRCs of RF, LR, and FNN were 0.73, 0.68, and 0.63, respectively. All models showed low Brier scores of <0.1 as well as highly fitted calibration plots, indicating a good fit of the ML-based models. On external validation, the GBM model demonstrated maximal performance with an AUROC of 0.90, while FNN had an AUROC of 0.85. The AUROCs of LR and RF were slightly lower at 0.80 and 0.79, respectively. The AUPRCs of GBM, LR, and FNN were similar at 0.47, 0.43, and 0.41, respectively, while that of RF was lower at 0.33. Among the categories in the GBM model, time-series dynamic data demonstrated a high AUROC of >0.95, contributing majorly to the excellent results.ConclusionsExploiting the diverse fields of the EMR data set, the ML-based 30-day adverse cardiac event prediction models demonstrated outstanding results, and the applied framework could be generalized for various health care prediction models.

Project description:ContextEnhanced recovery after surgery (ERAS) aims to improve surgical outcomes by integrating evidence-based practices across preoperative, intraoperative, and postoperative care. Data in electronic medical records (EMRs) provide insight on how ERAS is implemented and its impact on surgical outcomes. Because ERAS is a multimodal pathway provided by multiple physicians and health care providers over time, identifying ERAS cases in EMRs is not a trivial task. To better understand how EMRs can be used to study ERAS, we describe our experience with using current methodologies and the development and rationale of a new method for retrospectively identifying ERAS cases in EMRs.Case descriptionUsing EMR data from surgical departments at the University of North Carolina at Chapel Hill, we first identified ERAS cases using a protocol-based method, using basic information including the date of ERAS implementation, surgical procedure and date, and primary surgeon. We further examined two operational flags in the EMRs, a nursing order and a case request for OR order. Wide variation between the methods compelled us to consult with ERAS surgical staff and explore the EMRs to develop a more refined method for identifying ERAS cases.MethodWe developed a two-step method, with the first step based on the protocol definition and the second step based on an ERAS-specific medication definition. To test our method, we randomly sampled 150 general, gynecological, and urologic surgeries performed between January 1, 2016 and March 30, 2017. Surgical cases were classified as ERAS or not using the protocol definition, nursing order, case request for OR order, and our two-step method. To assess the accuracy of each method, two independent reviewers assessed the charts to determine whether cases were ERAS.FindingsOf the 150 charts reviewed, 74 were ERAS cases. The protocol only method and nursing order flag performed similarly, correctly identifying 74 percent and 73 percent of true ERAS cases, respectively. The case request for OR order flag performed less well, correctly identifying only 44 percent of the true ERAS cases. Our two-step method performed well, correctly identifying 98 percent of true ERAS cases.ConclusionERAS pathways are complex, making study of them from EMRs difficult. Current strategies for doing so are relatively easy to implement, but unreliable. We have developed a reproducible and observable ERAS computational phenotype that identifies ERAS cases reliably. This is a step forward in using the richness of EMR data to study ERAS implementation, efficacy, and how they can contribute to surgical care improvement.

Dataset Information

Prediction Model for 30-Day Mortality after Non-Cardiac Surgery Using Machine-Learning Techniques Based on Preoperative Evaluation of Electronic Medical Records.

Background

Methods

Results

Conclusion

Publications

Prediction Model for 30-Day Mortality after Non-Cardiac Surgery Using Machine-Learning Techniques Based on Preoperative Evaluation of Electronic Medical Records.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets