Clinical Nomogram Predicting Intracranial Injury in Pediatric Traumatic Brain Injury

Prognostic factors and clinical nomogram for in-hospital mortality in traumatic brain injury

BACKGROUND

Traumatic brain injury (TBI) is a major cause of death and functional disability in the general population. The nomogram is a clinical prediction tool that has been researched for a wide range of medical conditions. The purpose of this study was to identify prognostic factors associated with in-hospital mortality. The secondary objective was to develop a clinical nomogram for TBI patients' in-hospital mortality based on prognostic factors.

METHODS

A retrospective cohort study was conducted to analyze 14,075 TBI patients who were admitted to a tertiary hospital in southern Thailand. The total dataset was divided into the training and validation datasets. Several clinical characteristics and imaging findings were analyzed for in-hospital mortality in both univariate and multivariable analyses using the training dataset. Based on binary logistic regression, the nomogram was developed and internally validated using the final predictive model. Therefore, the predictive performances of the nomogram were estimated by the validation dataset.

RESULTS

Prognostic factors associated with in-hospital mortality comprised age, hypotension, antiplatelet, Glasgow coma scale score, pupillary light reflex, basilar skull fracture, acute subdural hematoma, subarachnoid hemorrhage, midline shift, and basal cistern obliteration that were used for building nomogram. The predictive performance of the nomogram was estimated by the training dataset; the area under the receiver operating characteristic curve (AUC) was 0.981. In addition, the AUCs of bootstrapping and cross-validation methods were 0.980 and 0.981, respectively. For the temporal validation with an unseen dataset, the sensitivity, specificity, accuracy, and AUC of the nomogram were 0.90, 0.88, 0.88, and 0.89, respectively.

CONCLUSION

A nomogram developed from prognostic factors had excellent performance; thus, the tool had the potential to serve as a screening tool for prognostication in TBI patients. Furthermore, future research should involve geographic validation to examine the predictive performances of the clinical prediction tool.

Comparison of intracranial pressure prediction in hydrocephalus patients among linear, non-linear, and machine learning regression models in Thailand

BACKGROUND

Hydrocephalus (HCP) is one of the most significant concerns in neurosurgical patients because it can cause increased intracranial pressure (ICP), resulting in mortality and morbidity. To date, machine learning (ML) has been helpful in predicting continuous outcomes. The primary objective of the present study was to identify the factors correlated with ICP, while the secondary objective was to compare the predictive performances among linear, non-linear, and ML regression models for ICP prediction.

METHODS

A total of 412 patients with various types of HCP who had undergone ventriculostomy was retrospectively included in the present study, and intraoperative ICP was recorded following ventricular catheter insertion. Several clinical factors and imaging parameters were analyzed for the relationship with ICP by linear correlation. The predictive performance of ICP was compared among linear, non-linear, and ML regression models.

RESULTS

Optic nerve sheath diameter (ONSD) had a moderately positive correlation with ICP (r=0.530, P<0.001), while several ventricular indexes were not statistically significant in correlation with ICP. For prediction of ICP, random forest (RF) and extreme gradient boosting (XGBoost) algorithms had low mean absolute error and root mean square error values and high R2 values compared to linear and non-linear regression when the predictive model included ONSD and ventricular indexes.

CONCLUSIONS

The XGBoost and RF algorithms are advantageous for predicting preoperative ICP and establishing prognoses for HCP patients. Furthermore, ML-based prediction could be used as a non-invasive method.

The establishment and validation of a prediction model for traumatic intracranial injury patients: a reliable nomogram

Objective

Traumatic brain injury (TBI) leads to death and disability. This study developed an effective prognostic nomogram for assessing the risk factors for TBI mortality.

Method

Data were extracted from an online database called "Multiparameter Intelligent Monitoring in Intensive Care IV" (MIMIC IV). The ICD code obtained data from 2,551 TBI persons (first ICU stay, >18 years old) from this database. R divided samples into 7:3 training and testing cohorts. The univariate analysis determined whether the two cohorts differed statistically in baseline data. This research used forward stepwise logistic regression after independent prognostic factors for these TBI patients. The optimal variables were selected for the model by the optimal subset method. The optimal feature subsets in pattern recognition improved the model prediction, and the minimum BIC forest of the high-dimensional mixed graph model achieved a better prediction effect. A nomogram-labeled TBI-IHM model containing these risk factors was made by nomology in State software. Least Squares OLS was used to build linear models, and then the Receiver Operating Characteristic (ROC) curve was plotted. The TBI-IHM nomogram model's validity was determined by receiver operating characteristic curves (AUCs), correction curve, Hosmer-Lemeshow test, integrated discrimination improvement (IDI), net reclassification improvement (NRI), and decision-curve analysis (DCA).

Result

The eight features with a minimal BIC model were mannitol use, mechanical ventilation, vasopressor use, international normalized ratio, urea nitrogen, respiratory rate, and cerebrovascular disease. The proposed nomogram (TBI-IHM model) was the best mortality prediction model, with better discrimination and superior model fitting for severely ill TBI patients staying in ICU. The model's receiver operating characteristic curve (ROC) was the best compared to the seven other models. It might be clinically helpful for doctors to make clinical decisions.

Conclusion

The proposed nomogram (TBI-IHM model) has significant potential as a clinical utility in predicting mortality in TBI patients.

Economic impact of a machine learning-based strategy for preparation of blood products in brain tumor surgery

Background

Globally, blood donation has been disturbed due to the pandemic. Consequently, the optimization of preoperative blood preparation should be a point of concern. Machine learning (ML) is one of the modern approaches that have been applied by physicians to help decision-making. The main objective of this study was to identify the cost differences of the ML-based strategy compared with other strategies in preoperative blood products preparation. A secondary objective was to compare the effectiveness indexes of blood products preparation among strategies.

Methods

The study utilized a retrospective cohort design conducted on brain tumor patients who had undergone surgery between January 2014 and December 2021. Overall data were divided into two cohorts. The first cohort was used for the development and deployment of the ML-based web application, while validation, comparison of the effectiveness indexes, and economic evaluation were performed using the second cohort. Therefore, the effectiveness indexes of blood preparation and cost difference were compared among the ML-based strategy, clinical trial-based strategy, and routine-based strategy.

Results

Over a 2-year period, the crossmatch to transfusion (C/T) ratio, transfusion probability (Tp), and transfusion index (Ti) of the ML-based strategy were 1.10, 57.0%, and 1.62, respectively, while the routine-based strategy had a C/T ratio of 4.67%, Tp of 27.9%%, and Ti of 0.79. The overall costs of blood products preparation among the ML-based strategy, clinical trial-based strategy, and routine-based strategy were 30, 061.56$, 57,313.92$, and 136,292.94$, respectively. From the cost difference between the ML-based strategy and routine-based strategy, we observed cost savings of 92,519.97$ (67.88%) for the 2-year period.

Conclusion

The ML-based strategy is one of the most effective strategies to balance the unnecessary workloads at blood banks and reduce the cost of unnecessary blood products preparation from low C/T ratio as well as high Tp and Ti. Further studies should be performed to confirm the generalizability and applicability of the ML-based strategy.

Prognostic Impact of the Combination of MGMT Methylation and TERT Promoter Mutation in Glioblastoma

Background  The concept of combinational analysis between the methylation of O ⁶ -methylguanine-DNA methyltransferase ( MGMT ) and telomerase reverse transcriptase promoter ( pTERT ) mutation in glioblastoma (GBM) has been reported. The main study objective was to determine the prognosis of patients with GBM based on MGMT/pTERT classification, while the secondary objective was to estimate the temozolomide effect on the survival time of GBM with MGMT/pTERT classification. Methods  A total of 50 GBM specimens were collected after tumor resection and were selected for investigating MGMT methylation and pTERT mutation. Clinical imaging and pathological characteristics were retrospectively analyzed. Patients with MGMT/pTERT classification were analyzed using survival analysis to develop the nomogram for forecasting and individual prognosis. Results  All patients underwent resection (total resection: 28%, partial resection: 64%, biopsy: 8%). Thirty-two percent of all cases received adjuvant temozolomide with radiotherapy. Sixty-four percent of the case was found methylated MGMT , and 56% of the present cohort found pTERT mutation. Following combinational analysis of biomarkers, results showed that the GBMs with methylated MGMT and wild-type pTERT had a superior prognosis compared with other subtypes. Using Cox regression analysis with multivariable analysis, the extent of resection, postoperative chemoradiotherapy, MGMT/pTERT classification were associated with a favorable prognosis. Hence, a web-based nomogram was developed for deploying individual prognostication. Conclusions  The interaction of MGMT methylation and pTERT mutation was confirmed for predicting prognosis. The results from the present study could help physicians create treatment strategies for GBM patients in real-world situations.

Comparison of intracranial injury predictability between machine learning algorithms and the nomogram in pediatric traumatic brain injury

OBJECTIVE

The overuse of head CT examinations has been much discussed, especially those for minor traumatic brain injury (TBI). In the disruptive era, machine learning (ML) is one of the prediction tools that has been used and applied in various fields of neurosurgery. The objective of this study was to compare the predictive performance between ML and a nomogram, which is the other prediction tool for intracranial injury following cranial CT in children with TBI.

METHODS

Data from 964 pediatric patients with TBI were randomly divided into a training data set (75%) for hyperparameter tuning and supervised learning from 14 clinical parameters, while the remaining data (25%) were used for validation purposes. Moreover, a nomogram was developed from the training data set with similar parameters. Therefore, models from various ML algorithms and the nomogram were built and deployed via web-based application.

RESULTS

A random forest classifier (RFC) algorithm established the best performance for predicting intracranial injury following cranial CT of the brain. The area under the receiver operating characteristic curve for the performance of RFC algorithms was 0.80, with 0.34 sensitivity, 0.95 specificity, 0.73 positive predictive value, 0.80 negative predictive value, and 0.79 accuracy.

CONCLUSIONS

The ML algorithms, particularly the RFC, indicated relatively excellent predictive performance that would have the ability to support physicians in balancing the overuse of head CT scans and reducing the treatment costs of pediatric TBI in general practice.

Application of machine learning to predict the outcome of pediatric traumatic brain injury

PURPOSE

Traumatic brain injury (TBI) generally causes mortality and disability, particularly in children. Machine learning (ML) is a computer algorithm, applied as a clinical prediction tool. The present study aims to assess the predictability of ML for the functional outcomes of pediatric TBI.

METHODS

A retrospective cohort study was performed targeting children with TBI who were admitted to the trauma center of southern Thailand between January 2009 and July 2020. The patient was excluded if he/she (1) did not undergo a CT scan of the brain, (2) died within the first 24 h, (3) had unavailable complete medical records during admission, or (4) was unable to provide updated outcomes. Clinical and radiologic characteristics were collected such as vital signs, Glasgow coma scale score, and characteristics of intracranial injuries. The functional outcome was assessed using the King's Outcome Scale for Childhood Head Injury, which was thus dichotomized into favourable outcomes and unfavourable outcomes: good recovery and moderate disability were categorized as the former, whereas death, vegetative state, and severe disability were categorized as the latter. The prognostic factors were estimated using traditional binary logistic regression. By data splitting, 70% of data were used for training the ML models and the remaining 30% were used for testing the ML models. The supervised algorithms including support vector machines, neural networks, random forest, logistic regression, naive Bayes and k-nearest neighbor were performed for training of the ML models. Therefore, the ML models were tested for the predictive performances by the testing datasets.

RESULTS

There were 828 patients in the cohort. The median age was 72 months (interquartile range 104.7 months, range 2-179 months). Road traffic accident was the most common mechanism of injury, accounting for 68.7%. At hospital discharge, favourable outcomes were achieved in 97.0% of patients, while the mortality rate was 2.2%. Glasgow coma scale score, hypotension, pupillary light reflex, and subarachnoid haemorrhage were associated with TBI outcomes following traditional binary logistic regression; hence, the 4 prognostic factors were used for building ML models and testing performance. The support vector machine model had the best performance for predicting pediatric TBI outcomes: sensitivity 0.95, specificity 0.60, positive predicted value 0.99, negative predictive value 1.0; accuracy 0.94, and area under the receiver operating characteristic curve 0.78.

CONCLUSION

The ML algorithms of the present study have a high sensitivity; therefore they have the potential to be screening tools for predicting functional outcomes and counselling prognosis in general practice of pediatric TBIs.