Random forest-based prediction of intracranial hypertension in patients with traumatic brain injury.

BACKGROUND: Treatment and prevention of intracranial hypertension (IH) to minimize secondary brain injury are central to the neurocritical care management of traumatic brain injury (TBI). Predicting the onset of IH in advance allows for a more aggressive prophylactic treatment. This study aimed to develop random forest (RF) models for predicting IH events in TBI patients. METHODS: We analyzed prospectively collected data from patients admitted to the intensive care unit with invasive intracranial pressure (ICP) monitoring. Patients with persistent ICP > 22 mmHg in the early postoperative period (first 6 h) were excluded to focus on IH events that had not yet occurred. ICP-related data from the initial 6 h were used to extract linear (ICP, cerebral perfusion pressure, pressure reactivity index, and cerebrospinal fluid compensatory reserve index) and nonlinear features (complexity of ICP and cerebral perfusion pressure). IH was defined as ICP > 22 mmHg for > 5 min, and severe IH (SIH) as ICP > 22 mmHg for > 1 h during the subsequent ICP monitoring period. RF models were then developed using baseline characteristics (age, sex, and initial Glasgow Coma Scale score) along with linear and nonlinear features. Fivefold cross-validation was performed to avoid overfitting. RESULTS: The study included 69 patients. Forty-three patients (62.3%) experienced an IH event, of whom 30 (43%) progressed to SIH. The median time to IH events was 9.83 h, and to SIH events, it was 11.22 h. The RF model showed acceptable performance in predicting IH with an area under the curve (AUC) of 0.76 and excellent performance in predicting SIH (AUC = 0.84). Cross-validation analysis confirmed the stability of the results. CONCLUSIONS: The presented RF model can forecast subsequent IH events, particularly severe ones, in TBI patients using ICP data from the early postoperative period. It provides researchers and clinicians with a potentially predictive pathway and framework that could help triage patients requiring more intensive neurological treatment at an early stage.

Predicting the Severity and Discharge Prognosis of Traumatic Brain Injury Based on Intracranial Pressure Data Using Machine Learning Algorithms.

OBJECTIVE: This study aimed to explore the potential of employing machine learning algorithms based on intracranial pressure (ICP), ICP-derived parameters, and their complexity to predict the severity and short-term prognosis of traumatic brain injury (TBI). METHODS: A single-center prospectively collected cohort of neurosurgical intensive care unit admissions was analyzed. We extracted ICP-related data within the first 6 hours and processed them using complex algorithms. To indicate TBI severity and short-term prognosis, Glasgow Coma Scale score on the first postoperative day and Glasgow Outcome Scale-Extended score at discharge were used as binary outcome variables. A univariate logistic regression model was developed to predict TBI severity using only mean ICP values. Subsequently, 3 multivariate Random Forest (RF) models were constructed using different combinations of mean and complexity metrics of ICP-related data. To avoid overfitting, five-fold cross-validations were performed. Finally, the best-performing multivariate RF model was used to predict patients' discharge Glasgow Outcome Scale-Extended score. RESULTS: The logistic regression model exhibited limited predictive ability with an area under the curve (AUC) of 0.558. Among multivariate models, the RF model, combining the mean and complexity metrics of ICP-related data, achieved the most robust ability with an AUC of 0.815. Finally, in terms of predicting discharge Glasgow Outcome Scale-Extended score, this model had a consistent performance with an AUC of 0.822. Cross-validation analysis confirmed the performance. CONCLUSIONS: This study demonstrates the clinical utility of the RF model, which integrates the mean and complexity metrics of ICP data, in accurately predicting the TBI severity and short-term prognosis.

Non-Invasive Assessment of Intracranial Hypertension in Patients with Traumatic Brain Injury Using Computed Tomography Radiomic Features: A Pilot Study.

This study aimed to assess intracranial hypertension in patients with traumatic brain injury non-invasively using computed tomography (CT) radiomic features. Fifty patients from the primary cohort were enrolled in this study. The clinical data, pre-operative cranial CT images, and initial intracranial pressure readings were collected and used to develop a prediction model. Data of 20 patients from another hospital were used to validate the model. Clinical features including age, sex, midline shift, basilar cistern status, and ventriculocranial ratio were measured. Radiomic features-i.e., 18 first-order and 40 second-order features- were extracted from the CT images. LASSO method was used for features filtration. Multi-variate logistic regression was used to develop three prediction models with clinical (CF model), first-order (FO model), and second-order features (SO model). The SO model achieved the most robust ability to predict intracranial hypertension. Internal validation showed that the C-statistic of the model was 0.811 (95% confidence interval [CI]: 0.691-0.931) with the bootstrapping method. The Hosmer Lemeshow test and calibration curve also showed that the SO model had excellent performance. The external validation results showed a good discrimination with an area under the curve of 0.725 (95% CI: 0.500-0.951). Although the FO model was inferior to the SO model, it had better prediction ability than the CF model. The study shows that the radiomic features analysis, especially second-order features, can be used to evaluate intracranial hypertension non-invasively compared with conventional clinical features, given its potential for clinical practice and further research.

The application value of CT radiomics features in predicting pressure amplitude correlation index in patients with severe traumatic brain injury.

Purpose: To explore the application value of a machine learning model based on CT radiomics features in predicting the pressure amplitude correlation index (RAP) in patients with severe traumatic brain injury (sTBI). Methods: Retrospectively analyzed the clinical and imaging data in 36 patients with sTBI. All patients underwent surgical treatment, continuous ICP monitoring, and invasive arterial pressure monitoring. The pressure amplitude correlation index (RAP) was collected within 1 h after surgery. Three volume of interest (VOI) was selected from the craniocerebral CT images of patients 1 h after surgery, and a total of 93 radiomics features were extracted from each VOI. Three models were established to be used to evaluate the patients' RAP levels. The accuracy, precision, recall rate, F1 score, receiver operating characteristic (ROC) curve, and area under the curve (AUC) were used to evaluate the predictive performance of each model. Results: The optimal number of features for three predicting models of RAP was five, respectively. The accuracy of predicting the model of the hippocampus was 77.78%, precision was 88.24%, recall rate was 60%, the F1 score was 0.6, and AUC was 0.88. The accuracy of predicting the model of the brainstem was 63.64%, precision was 58.33%, the recall rate was 60%, the F1 score was 0.54, and AUC was 0.82. The accuracy of predicting the model of the thalamus was 81.82%, precision was 88.89%, recall rate was 75%, the F1 score was 0.77, and AUC was 0.96. Conclusions: CT radiomics can predict RAP levels in patients with sTBI, which has the potential to establish a method of non-invasive intracranial pressure (NI-ICP) monitoring.

Evaluation of Intracranial Hypertension in Patients With Hypertensive Intracerebral Hemorrhage Using Texture Analysis.

Purpose: Texture analysis based on clinical images had been widely used in neurological diseases. This study aimed to achieve depth information of computed tomography (CT) images by texture analysis and to establish a model for noninvasive evaluation of intracranial pressure (ICP) in patients with hypertensive intracerebral hemorrhage (HICH). Methods: Forty-seven patients with HICH were selected. Related CT images and ICP value were collected. The morphological features of hematoma volume, midline shift, and ventriculocranial ratio were measured. Forty textural features were extracted from regions of interest. Four models were established to predict intracranial hypertension with morphological features, textural features of anterior horn, textural features of temporal lobe, and textural features of posterior horn. Results: Model of posterior horn had the highest ability to predict intracranial hypertension (AUC = 0.90, F1 score = 0.72), followed by model of anterior horn (AUC = 0.70, F1 score = 0.53) and model of temporal lobe (AUC = 0.70, F1 score = 0.58), and model of morphological features displayed the worst performance (AUC = 0.42, F1 score = 0.38). Conclusion: Texture analysis can realize interpretation of CT images in depth, which has great potential in noninvasive evaluation of intracranial hypertension.

Spindle wave in intracranial pressure signal analysis for patients with traumatic brain injury: A single-center prospective observational cohort study.

Objective: Intracranial pressure (ICP) monitoring is an integral part of the multimodality monitoring system in the neural intensive care unit. The present study aimed to describe the morphology of the spindle wave (a shuttle shape with wide middle and narrow ends) during ICP signal monitoring in TBI patients and to investigate its clinical significance. Methods: Sixty patients who received ICP sensor placement and admitted to the neurosurgical intensive care unit between January 2021 and September 2021 were prospectively enrolled. The patient's Glasgow Coma Scale (GCS) score on admission and at discharge and length of stay in hospital were recorded. ICP monitoring data were monitored continuously. The primary endpoint was 6-month Glasgow Outcome Scale-Extended (GOSE) score. Patients with ICP spindle waves were assigned to the spindle wave group and those without were assigned to the control group. The correlation between the spindle wave and 6-month GOSE was analyzed. Meanwhile, the mean ICP and two ICP waveform-derived indices, ICP pulse amplitude (AMP) and correlation coefficient between AMP and ICP (RAP) were comparatively analyzed. Results: There were no statistically significant differences between groups in terms of age (p = 0.89), gender composition (p = 0.62), and GCS score on admission (p = 0.73). Patients with spindle waves tended to have a higher GCS score at discharge (12.75 vs. 10.90, p = 0.01), a higher increment in GCS score during hospitalization (ΔGCS, the difference between discharge GCS score and admission GCS score) (4.95 vs. 2.80, p = 0.01), and a better 6-month GOSE score (4.90 vs. 3.68, p = 0.04) compared with the control group. And the total duration of the spindle wave was positively correlated with 6-month GOSE (r = 0.62, p = 0.004). Furthermore, the parameters evaluated during spindle waves, including mean ICP, AMP, and RAP, demonstrated significant decreases compared with the parameters before the occurrence of the spindle wave (all p < 0.025). Conclusion: The ICP spindle wave was associated with a better prognosis in TBI patients. Physiological parameters such as ICP, AMP, and RAP were significantly improved when spindle waves occurred, which may explain the enhancement of clinical outcomes. Further studies are needed to investigate the pathophysiological mechanisms behind this wave.

Evaluation of Intracranial Hypertension in Traumatic Brain Injury Patient: A Noninvasive Approach Based on Cranial Computed Tomography Features.

BACKGROUND: Our purpose was to establish a noninvasive quantitative method for assessing intracranial pressure (ICP) levels in patients with traumatic brain injury (TBI) through investigating the Hounsfield unit (HU) features of computed tomography (CT) images. METHODS: In this retrospective study, 47 patients with a closed TBI were recruited. Hounsfield unit features from the last cranial CT and the initial ICP value were collected. Three models were established to predict intracranial hypertension with Hounsfield unit (HU model), midline shift (MLS model), and clinical expertise (CE model) features. RESULTS: The HU model had the highest ability to predict intracranial hypertension. In 34 patients with unilateral injury, the HU model displayed the highest performance. In three classifications of intracranial hypertension (ICP ≤ 22, 23-29, and ≥30 mmHg), the HU model achieved the highest F1 score. CONCLUSIONS: This radiological feature-based noninvasive quantitative approach showed better performance compared with conventional methods, such as the degree of midline shift and clinical expertise. The results show its potential in clinical practice and further research.

Intra-Abdominal Pressure Measurements in Neurocritical Patients.

Intra-abdominal pressure (IAP) is increasingly being recognized as an indispensable and significant physiological parameter in intensive care units (ICU). IAP has been measured in a variety of ways with the development of many techniques in recent years. The level of intra-abdominal pressure under normal conditions is generally equal to or less than 12 mmHg. Accordingly, abdominal hypertension (IAH) is defined as two consecutive IAP measurements higher than 12 mmHg within 4-6 h. When IAH deteriorates further with IAP higher than 20 mmHg along with organ dysfunction and/or failure, this clinical manifestation can be diagnosed as abdominal compartment syndrome (ACS). IAH and ACS are associated with gastrointestinal ischemia, acute renal failure, and lung injury, leading to severe morbidity and mortality. Elevated IAP and IAH may affect the cerebral venous return and outflow of the cerebrospinal fluid by increasing the intrathoracic pressure (ITP), ultimately leading to increased intracranial pressure (ICP). Therefore, it is essential to monitor IAP in critically ill patients. The reproducibility and accuracy of intra-bladder pressure (IBP) measurements in previous studies need to be further improved, although the indirect measurement of IAP is now a widely used technique. To address these limitations, we recently used a set of IAP monitoring systems with advantages of convenience, continuous monitoring, digital visualization, and long-term IAP recording and data storage in critically ill patients. This IAP monitoring system can detect intra-abdominal hypertension and potentially analyze clinical status in real time. The recorded IAP data and other physiological indicators, such as intracranial pressure, can be further used for correlation analysis to guide treatment and predict a patient's possible prognosis.