Machine Learning to Predict Three Types of Outcomes After Traumatic Brain Injury Using Data at Admission: A Multi-Center Study for Development and Validation

A meta-analysis of the diagnostic test accuracy of artificial intelligence predicting emergency department dispositions

BACKGROUND

The rapid advancement of Artificial Intelligence (AI) has led to its widespread application across various domains, showing encouraging outcomes. Many studies have utilized AI to forecast emergency department (ED) disposition, aiming to forecast patient outcomes earlier and to allocate resources better; however, a dearth of comprehensive review literature exists to assess the objective performance standards of these predictive models using quantitative evaluations. This study aims to conduct a meta-analysis to assess the diagnostic accuracy of AI in predicting ED disposition, encompassing admission, critical care, and mortality.

METHODS

Multiple databases, including Scopus, Springer, ScienceDirect, PubMed, Wiley, Sage, and Google Scholar, were searched until December 31, 2023, to gather relevant literature. Risk of bias was assessed using the Prediction Model Risk of Bias Assessment Tool. Pooled estimates of sensitivity, specificity, and area under the receiver operating characteristic curve (AUROC) were calculated to evaluate AI's predictive performance. Sub-group analyses were performed to explore covariates affecting AI predictive model performance.

RESULTS

The study included 88 articles possessed with 117 AI models, among which 39, 45, and 33 models predicted admission, critical care, and mortality, respectively. The reported statistics for sensitivity, specificity, and AUROC represent pooled summary measures derived from the component studies included in this meta-analysis. AI's summary sensitivity, specificity, and AUROC for predicting admission were 0.81 (95% Confidence Interval [CI] 0.74-0.86), 0.87 (95% CI 0.81-0.91), and 0.87 (95% CI 0.84-0.93), respectively. For critical care, the values were 0.86 (95% CI 0.79-0.91), 0.89 (95% CI 0.83-0.93), and 0.93 (95% CI 0.89-0.95), respectively, and for mortality, they were 0.85 (95% CI 0.80-0.89), 0.94 (95% CI 0.90-0.96), and 0.93 (95% CI 0.89-0.96), respectively. Emergent sample characteristics and AI techniques showed evidence of significant covariates influencing the heterogeneity of AI predictive models for ED disposition.

CONCLUSIONS

The meta-analysis indicates promising performance of AI in predicting ED disposition, with certain potential for improvement, especially in sensitivity. Future research could explore advanced AI techniques such as ensemble learning and cross-validation with hyper-parameter tuning to enhance predictive model efficacy.

TRIAL REGISTRATION

This systematic review was not registered with PROSPERO or any other similar registry because the review was completed prior to the opportunity for registration, and PROSPERO currently does not accept registrations for reviews that are already completed. We are committed to transparency and have adhered to best practices in systematic review methodology throughout this study.

Prediction Model for Unfavorable Outcome in Primary Decompressive Craniectomy for Isolated Moderate to Severe Traumatic Brain Injury in India: A Prospective Observational Study

OBJECTIVE

Traumatic brain injury (TBI) prediction models have gained significant attention in recent years because of their potential to aid in clinical decision making. Existing models, such as Corticosteroid Randomization after Significant Head Injury and International Mission for Prognosis and Analysis of Clinical Trials, are currently losing external validity and performance, probably because of their diverse inclusion criteria and changes in treatment modalities over the years. There is a lack of models that predict outcomes strictly pertaining to primary decompression after TBI. In this study, we aimed to develop an easy-to-use prediction model for predicting the risk of poor functional outcomes at 3 months after hospital discharge in adult patients who had undergone primary decompressive craniectomy for isolated moderate-to-severe TBI.

METHODS

We conducted a prospective observational study at our tertiary care hospital. We trained and tested multiple prognostic logistic regression models with ten-fold cross validation to choose the model with the lowest Akaike information criterion, high sensitivity, and positive predictive value (PPV). Using the final model, we generated a nomogram to predict the risk of having a Glasgow outcome scale-extended (GOSE) 1-4 at three months after hospital discharge.

RESULTS

A total of 215 patients were included in this study. Variables with an absolute standardized difference >0·25 when grouped by GOSE 1-4/5-8 at three months were included in multivariable modeling. The model of choice had an accuracy of 87·91% (95% confidence interval of 82·78%-91·95%), a sensitivity of 84·42%, specificity of 89·86%, PPV of 82·28% (72·06%-89·96%), negative predictive value of 91·18% (85·09%-95·36%), LR+ of 8·32 (5·02-13·80), and LR-of 0·17 (0·10-0·29).

CONCLUSIONS

Our study provides a ready-to-use prognostic nomogram derived from prospective data that can predict the risk of having a GOSE of 1-4 at three months following primary decompressive craniectomy with high sensitivity, PPV, and low LR-.

Predictive Modeling of Long-Term Care Needs in Traumatic Brain Injury Patients Using Machine Learning

BACKGROUND

Traumatic brain injury (TBI) research often focuses on mortality rates or functional recovery, yet the critical need for long-term care among patients dependent on institutional or Respiratory Care Ward (RCW) support remains underexplored. This study aims to address this gap by employing machine learning techniques to develop and validate predictive models that analyze the prognosis of this patient population.

METHOD

Retrospective data from electronic medical records at Chi Mei Medical Center, encompassing 2020 TBI patients admitted to the ICU between January 2016 and December 2021, were collected. A total of 44 features were included, utilizing four machine learning models and various feature combinations based on clinical significance and Spearman correlation coefficients. Predictive performance was evaluated using the area under the curve (AUC) of the receiver operating characteristic (ROC) curve and validated with the DeLong test and SHAP (SHapley Additive exPlanations) analysis.

RESULT

Notably, 236 patients (11.68%) were transferred to long-term care centers. XGBoost with 27 features achieved the highest AUC (0.823), followed by Random Forest with 11 features (0.817), and LightGBM with 44 features (0.813). The DeLong test revealed no significant differences among the best predictive models under various feature combinations. SHAP analysis illustrated a similar distribution of feature importance for the top 11 features in XGBoost, with 27 features, and Random Forest with 11 features.

CONCLUSIONS

Random Forest, with an 11-feature combination, provided clinically meaningful predictive capability, offering early insights into long-term care trends for TBI patients. This model supports proactive planning for institutional or RCW resources, addressing a critical yet often overlooked aspect of TBI care.

Predicting the Outcome and Survival of Patients with Spinal Cord Injury Using Machine Learning Algorithms: A Systematic Review

BACKGROUND

Spinal cord injury (SCI) is a significant public health issue, leading to physical, psychological, and social complications. Machine learning (ML) algorithms have shown potential in diagnosing and predicting the functional and neurologic outcomes of subjects with SCI. ML algorithms can predict scores for SCI classification systems and accurately predict outcomes by analyzing large amounts of data. This systematic review aimed to examine the performance of ML algorithms for diagnosing and predicting the outcomes of subjects with SCI.

METHODS

The literature was comprehensively searched for the pertinent studies from inception to May 25, 2023. Therefore, electronic databases of PubMed, Embase, Scopus, and Web of Science were systematically searched with individual search syntax.

RESULTS

A total of 9424 individuals diagnosed with SCI across multiple studies were analyzed. Among the 21 studies included, 5 specifically aimed to evaluate diagnostic accuracy, while the remaining 16 focused on exploring prognostic factors or management strategies.

CONCLUSIONS

ML and deep learning (DL) have shown great potential in various aspects of SCI. ML and DL algorithms have been employed multiple times in predicting and diagnosing patients with SCI. While there are studies on diagnosing acute SCI using DL algorithms, further research is required in this area.

Computational Prognostic Modeling in Traumatic Brain Injury

Traumatic brain injury is the leading cause of death and disability worldwide. Despite this large impact, no predictive models are in widespread use due to tedious data collection requirements, lack of provider trust, and poor performance. Furthermore, these models use simple, often binary, data elements that fail to capture the complex heterogeneity of traumatic brain injury. Recent advances in computational modeling efforts have demonstrated promising results for capturing imaging, clinical, electroencephalographic, and other biomarkers for powerful predictive models. In this review, we provide an overview of efforts in computational modeling in neurotrauma and provide insights into future directions.

Machine learning-based model for predicting inpatient mortality in adults with traumatic brain injury: a systematic review and meta-analysis

Background and objective

Predicting mortality from traumatic brain injury facilitates early data-driven treatment decisions. Machine learning has predicted mortality from traumatic brain injury in a growing number of studies, and the aim of this study was to conduct a meta-analysis of machine learning models in predicting mortality from traumatic brain injury.

Methods

This systematic review and meta-analysis included searches of PubMed, Web of Science and Embase from inception to June 2023, supplemented by manual searches of study references and review articles. Data were analyzed using Stata 16.0 software. This study is registered with PROSPERO (CRD2023440875).

Results

A total of 14 studies were included. The studies showed significant differences in the overall sample, model type and model validation. Predictive models performed well with a pooled AUC of 0.90 (95% CI: 0.87 to 0.92).

Conclusion

Overall, this study highlights the excellent predictive capabilities of machine learning models in determining mortality following traumatic brain injury. However, it is important to note that the optimal machine learning modeling approach has not yet been identified.

Systematic review registration

https://www.crd.york.ac.uk/PROSPERO/display_record.php?RecordID=440875, identifier CRD2023440875.