1
|
Vakitbilir N, Bergmann T, Froese L, Gomez A, Sainbhi AS, Stein KY, Islam A, Zeiler FA. Multivariate modeling and prediction of cerebral physiology in acute traumatic neural injury: A scoping review. Comput Biol Med 2024; 178:108766. [PMID: 38905893 DOI: 10.1016/j.compbiomed.2024.108766] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/24/2024] [Revised: 06/14/2024] [Accepted: 06/14/2024] [Indexed: 06/23/2024]
Abstract
Traumatic brain injury (TBI) poses a significant global public health challenge necessitating a profound understanding of cerebral physiology. The dynamic nature of TBI demands sophisticated methodologies for modeling and predicting cerebral signals to unravel intricate pathophysiology and predict secondary injury mechanisms prior to their occurrence. In this comprehensive scoping review, we focus specifically on multivariate cerebral physiologic signal analysis in the context of multi-modal monitoring (MMM) in TBI, exploring a range of techniques including multivariate statistical time-series models and machine learning algorithms. Conducting a comprehensive search across databases yielded 7 studies for evaluation, encompassing diverse cerebral physiologic signals and parameters from TBI patients. Among these, five studies concentrated on modeling cerebral physiologic signals using statistical time-series models, while the remaining two studies primarily delved into intracranial pressure (ICP) prediction through machine learning models. Autoregressive models were predominantly utilized in the modeling studies. In the context of prediction studies, logistic regression and Gaussian processes (GP) emerged as the predominant choice in both research endeavors, with their performance being evaluated against each other in one study and other models such as random forest, and decision tree in the other study. Notably among these models, random forest model, an ensemble learning approach, demonstrated superior performance across various metrics. Additionally, a notable gap was identified concerning the absence of studies focusing on prediction for multivariate outcomes. This review addresses existing knowledge gaps and sets the stage for future research in advancing cerebral physiologic signal analysis for neurocritical care improvement.
Collapse
Affiliation(s)
- Nuray Vakitbilir
- Biomedical Engineering, Price Faculty of Engineering, University of Manitoba, Winnipeg, Canada.
| | - Tobias Bergmann
- Undergraduate Engineering, Price Faculty of Engineering, University of Manitoba, Winnipeg, Canada
| | - Logan Froese
- Biomedical Engineering, Price Faculty of Engineering, University of Manitoba, Winnipeg, Canada
| | - Alwyn Gomez
- Section of Neurosurgery, Department of Surgery, Rady Faculty of Health Sciences, University of Manitoba, Winnipeg, Canada; Department of Human Anatomy and Cell Science, Rady Faculty of Health Sciences, University of Manitoba, Winnipeg, Canada
| | - Amanjyot Singh Sainbhi
- Biomedical Engineering, Price Faculty of Engineering, University of Manitoba, Winnipeg, Canada
| | - Kevin Y Stein
- Biomedical Engineering, Price Faculty of Engineering, University of Manitoba, Winnipeg, Canada
| | - Abrar Islam
- Biomedical Engineering, Price Faculty of Engineering, University of Manitoba, Winnipeg, Canada
| | - Frederick A Zeiler
- Biomedical Engineering, Price Faculty of Engineering, University of Manitoba, Winnipeg, Canada; Section of Neurosurgery, Department of Surgery, Rady Faculty of Health Sciences, University of Manitoba, Winnipeg, Canada; Department of Clinical Neuroscience, Karolinska Institutet, Stockholm, Sweden; Division of Anaesthesia, Department of Medicine, Addenbrooke's Hospital, University of Cambridge, Cambridge, UK; Pan Clinic Foundation, Winnipeg, Manitoba, Canada
| |
Collapse
|
2
|
Powers AY, Pinto MB, Tang OY, Chen JS, Doberstein C, Asaad WF. Predicting mortality in traumatic intracranial hemorrhage. J Neurosurg 2020; 132:552-559. [PMID: 30797192 DOI: 10.3171/2018.11.jns182199] [Citation(s) in RCA: 17] [Impact Index Per Article: 4.3] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/30/2018] [Accepted: 11/08/2018] [Indexed: 11/06/2022]
Abstract
OBJECTIVE Traumatic intracranial hemorrhage (tICH) is a significant source of morbidity and mortality in trauma patients. While prognostic models for tICH outcomes may assist in alerting clinicians to high-risk patients, previously developed models face limitations, including low accuracy, poor generalizability, and the use of more prognostic variables than is practical. This study aimed to construct a simpler and more accurate method of risk stratification for all tICH patients. METHODS The authors retrospectively identified a consecutive series of 4110 patients admitted to their institution's level 1 trauma center between 2003 and 2013. For each admission, they collected the patient's sex, age, systolic blood pressure, blood alcohol concentration, antiplatelet/anticoagulant use, Glasgow Coma Scale (GCS) score, Injury Severity Score, presence of epidural hemorrhage, presence of subdural hemorrhage, presence of subarachnoid hemorrhage, and presence of intraparenchymal hemorrhage. The final study population comprised 3564 patients following exclusion of records with missing data. The dependent variable under study was patient death. A k-fold cross-validation was carried out with the best models selected via the Akaike Information Criterion. These models risk stratified the study partitions into grade I (< 1% predicted mortality), grade II (1%-10% predicted mortality), grade III (10%-40% predicted mortality), or grade IV (> 40% predicted mortality) tICH. Predicted mortalities were compared with actual mortalities within grades to assess calibration. Concordance was also evaluated. A final model was constructed using the entire data set. Subgroup analysis was conducted for each hemorrhage type. RESULTS Cross-validation demonstrated good calibration (p < 0.001 for all grades) with a mean concordance of 0.881 (95% CI 0.865-0.898). In the authors' final model, older age, lower blood alcohol concentration, antiplatelet/anticoagulant use, lower GCS score, and higher Injury Severity Score were all associated with greater mortality. Subgroup analysis showed successful stratification for subarachnoid, intraparenchymal, grade II-IV subdural, and grade I epidural hemorrhages. CONCLUSIONS The authors developed a risk stratification model for tICH of any GCS score with concordance comparable to prior models and excellent calibration. These findings are applicable to multiple hemorrhage subtypes and can assist in identifying low-risk patients for more efficient resource allocation, facilitate family conversations regarding goals of care, and stratify patients for research purposes. Future work will include testing of more variables, validation of this model across institutions, as well as creation of a simplified model whose outputs can be calculated mentally.
Collapse
Affiliation(s)
- Andrew Y Powers
- 1Department of Neurosurgery, Warren Alpert Medical School of Brown University
| | - Mauricio B Pinto
- 1Department of Neurosurgery, Warren Alpert Medical School of Brown University
| | - Oliver Y Tang
- 1Department of Neurosurgery, Warren Alpert Medical School of Brown University
| | - Jia-Shu Chen
- 1Department of Neurosurgery, Warren Alpert Medical School of Brown University
| | - Cody Doberstein
- 1Department of Neurosurgery, Warren Alpert Medical School of Brown University
| | - Wael F Asaad
- 1Department of Neurosurgery, Warren Alpert Medical School of Brown University
- 2Carney Institute for Brain Science, Brown University
- 3Department of Neuroscience, Brown University; and
- 4Norman Prince Neurosciences Institute and
- 5Department of Neurosurgery, Rhode Island Hospital, Providence, Rhode Island
| |
Collapse
|
3
|
Jiang C, Cao J, Williamson C, Farzaneh N, Rajajee V, Gryak J, Najarian K, Soroushmehr SMR. Midline Shift vs. Mid-Surface Shift: Correlation with Outcome of Traumatic Brain Injuries. PROCEEDINGS. IEEE INTERNATIONAL CONFERENCE ON BIOINFORMATICS AND BIOMEDICINE 2019; 2019:1083-1086. [PMID: 33569243 PMCID: PMC7871460 DOI: 10.1109/bibm47256.2019.8983159] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 11/09/2022]
Abstract
Traumatic brain injury (TBI) is a major health and socioeconomic problem globally that is associated with a high level of mortality. Early and accurate diagnosis and prognosis of TBI is important in patient management and preventing any secondary injuries. Computer tomography (CT) imaging assists physicians in diagnosing injury and guiding treatment. One of the clinical parameters extracted from CT images is midline shift, a measure of linear displacement in brain structure, which is correlated with TBI patient outcomes. However, only a tiny fraction of the overall tissue displacement is quantified through this parameter. In this paper, a novel measurement of overall mid-surface shift is proposed that quantifies the total volume of brain tissue shifted across the midline. When compared to traditional midline shift, mid-surface shift has a stronger correlation with TBI patient outcomes.
Collapse
Affiliation(s)
- Cheng Jiang
- Department of Electrical Engineering and Computer Science, University of Michigan, Ann Arbor, MI
| | - Jie Cao
- Department of Computational Medicine and Bioinformatics, University of Michigan, Ann Arbor, MI
| | - Craig Williamson
- Department of Neurosurgery, University of Michigan, Ann Arbor, MI
- Michigan Center for Integrative Research in Critical Care, University of Michigan, Ann Arbor, MI
| | - Negar Farzaneh
- Department of Computational Medicine and Bioinformatics, University of Michigan, Ann Arbor, MI
| | - Venkatakrishna Rajajee
- Department of Neurosurgery, University of Michigan, Ann Arbor, MI
- Michigan Center for Integrative Research in Critical Care, University of Michigan, Ann Arbor, MI
| | - Jonathan Gryak
- Department of Computational Medicine and Bioinformatics, University of Michigan, Ann Arbor, MI
| | - Kayvan Najarian
- Department of Electrical Engineering and Computer Science, University of Michigan, Ann Arbor, MI
- Department of Computational Medicine and Bioinformatics, University of Michigan, Ann Arbor, MI
- Michigan Center for Integrative Research in Critical Care, University of Michigan, Ann Arbor, MI
- Department of Emergency Medicine, University of Michigan, Ann Arbor, MI
| | - S M Reza Soroushmehr
- Department of Computational Medicine and Bioinformatics, University of Michigan, Ann Arbor, MI
- Michigan Center for Integrative Research in Critical Care, University of Michigan, Ann Arbor, MI
| |
Collapse
|
4
|
Dijkland SA, Foks KA, Polinder S, Dippel DWJ, Maas AIR, Lingsma HF, Steyerberg EW. Prognosis in Moderate and Severe Traumatic Brain Injury: A Systematic Review of Contemporary Models and Validation Studies. J Neurotrauma 2019; 37:1-13. [PMID: 31099301 DOI: 10.1089/neu.2019.6401] [Citation(s) in RCA: 83] [Impact Index Per Article: 16.6] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 12/22/2022] Open
Abstract
Outcome prognostication in traumatic brain injury (TBI) is important but challenging due to heterogeneity of the disease. The aim of this systematic review is to present the current state-of-the-art on prognostic models for outcome after moderate and severe TBI and evidence on their validity. We searched for studies reporting on the development, validation or extension of prognostic models for functional outcome after TBI with Glasgow Coma Scale (GCS) ≤12 published between 2006-2018. Studies with patients age ≥14 years and evaluating a multi-variable prognostic model based on admission characteristics were included. Model discrimination was expressed with the area under the receiver operating characteristic curve (AUC), and model calibration with calibration slope and intercept. We included 58 studies describing 67 different prognostic models, comprising the development of 42 models, 149 external validations of 31 models, and 12 model extensions. The most common predictors were GCS (motor) score (n = 55), age (n = 54), and pupillary reactivity (n = 48). Model discrimination varied substantially between studies. The International Mission for Prognosis and Analysis of Clinical Trials (IMPACT) and Corticoid Randomisation After Significant Head injury (CRASH) models were developed on the largest cohorts (8509 and 10,008 patients, respectively) and were most often externally validated (n = 91), yielding AUCs ranging between 0.65-0.90 and 0.66-1.00, respectively. Model calibration was reported with a calibration intercept and slope for seven models in 53 validations, and was highly variable. In conclusion, the discriminatory validity of the IMPACT and CRASH prognostic models is supported across a range of settings. The variation in calibration, reflecting heterogeneity in reliability of predictions, motivates continuous validation and updating if clinical implementation is pursued.
Collapse
Affiliation(s)
- Simone A Dijkland
- Department of Public Health, Center for Medical Decision Making, Erasmus MC-University Medical Center Rotterdam, the Netherlands
| | - Kelly A Foks
- Department of Public Health, Center for Medical Decision Making, Erasmus MC-University Medical Center Rotterdam, the Netherlands.,Department of Neurology, Erasmus MC-University Medical Center Rotterdam, the Netherlands
| | - Suzanne Polinder
- Department of Public Health, Center for Medical Decision Making, Erasmus MC-University Medical Center Rotterdam, the Netherlands
| | - Diederik W J Dippel
- Department of Neurology, Erasmus MC-University Medical Center Rotterdam, the Netherlands
| | - Andrew I R Maas
- Department of Neurosurgery, Antwerp University Hospital, University of Antwerp, Edegem, Belgium
| | - Hester F Lingsma
- Department of Public Health, Center for Medical Decision Making, Erasmus MC-University Medical Center Rotterdam, the Netherlands
| | - Ewout W Steyerberg
- Department of Public Health, Center for Medical Decision Making, Erasmus MC-University Medical Center Rotterdam, the Netherlands.,Department of Biomedical Data Sciences, Leiden University Medical Center, Leiden, the Netherlands
| |
Collapse
|