The detection of mild traumatic brain injury in paediatrics using artificial neural networks
Hanem Ellethy1, Shekhar S Chandra2, Fatima A Nasrallah1
1Queensland Brain Institute, The University of Queensland, Brisbane, QLD, Australia.
Insights
Machine learning models accurately diagnose mild traumatic brain injury (mTBI) in children using head CT data. Artificial neural networks show high accuracy, aiding faster TBI diagnosis in emergency settings.
Area of Science:
- Pediatric Emergency Medicine
- Medical Imaging Analysis
- Artificial Intelligence in Healthcare
Background:
- Head computed tomography (CT) is standard for evaluating mild traumatic brain injury (mTBI) in pediatric emergency departments (EDs).
- Developing data-driven models can improve the timeliness and cost-effectiveness of TBI diagnosis.
- Existing diagnostic methods can be burdensome in busy EDs.
Purpose of the Study:
- To apply and compare two machine learning (ML) models for diagnosing mTBI in a pediatric population.
- To evaluate the diagnostic performance of a hybrid Random Forest-Artificial Neural Network (RF-ANN) model and a deep ANN model.
- To assess the feasibility of using ANN for mTBI diagnosis in children using clinical and non-imaging data.
Main Methods:
- Utilized a dataset of 15,271 pediatric patients (<18 years) with mTBI and head CT reports from the PECARN study (2004-2006).
- Developed a hybrid RF-ANN model using top-ranked clinical and CT features and a deep ANN model using all available features.
- Employed five-fold cross-validation with an 80% training and 20% testing data split; calculated accuracy, sensitivity, precision, and specificity.
Main Results:
- The hybrid RF-ANN model achieved high performance: 99.96% specificity, 95.98% sensitivity, 99.25% precision, and 99.74% accuracy.
- The deep ANN model demonstrated excellent results: 99.9% specificity, 99.2% sensitivity, 99.9% precision, and 99.9% accuracy.
- Both models showed strong capabilities in classifying mTBI in the pediatric cohort.
Conclusions:
- Artificial neural networks (ANNs), including deep learning, are feasible for diagnosing pediatric mTBI using clinical and non-imaging data.
- These ML models offer a potential to reduce the evaluation burden in EDs and support clinical decision-making.
- This study is the first to investigate deep ANN for mTBI diagnosis in a pediatric cohort with balanced sensitivity and specificity.
Abstract:
Head computed tomography (CT) is the gold standard in emergency departments (EDs) to evaluate mild traumatic brain injury (mTBI) patients, especially for paediatrics. Data-driven models for successfully classifying head CT scans that have mTBI will be valuable in terms of timeliness and cost-effectiveness for TBI diagnosis. This study applied two different machine learning (ML) models to diagnose mTBI in a paediatric population collected as part of the paediatric emergency care applied research network (PECARN) study between 2004 and 2006. The models were conducted using 15,271 patients under the age of 18 years with mTBI and had a head CT report. In the conventional model, random forest (RF) ranked the features to reduce data dimensionality and the top ranked features were used to train a shallow artificial neural network (ANN) model. In the second model, a deep ANN applied to classify positive and negative mTBI patients using the entirety of the features available. The dataset was divided into two subsets: 80% for training and 20% for testing using five-fold cross-validation. Accuracy, sensitivity, precision, and specificity were calculated by comparing the model's prediction outcome to the actual diagnosis for each patient. RF ranked ten clinical demographic features and twelve CT-findings; the hybrid RF-ANN model achieved an average specificity of 99.96%, sensitivity of 95.98%, precision of 99.25%, and accuracy of 99.74% in identifying positive mTBI from negative mTBI subjects. The deep ANN proved its ability to carry out the task efficiently with an average specificity of 99.9%, sensitivity of 99.2%, precision of 99.9%, and accuracy of 99.9%. The performance of the two proposed models demonstrated the feasibility of using ANN to diagnose mTBI in a paediatric population. This is the first study to investigate deep ANN in a paediatric cohort with mTBI using clinical and non-imaging data and diagnose mTBI with balanced sensitivity and specificity using shallow and deep ML models. This method, if validated, would have the potential to reduce the burden of TBI evaluation in EDs and aide clinicians in the decision-making process.
