1
|
Gu L, Hu H, Wu S, Li F, Li Z, Xiao Y, Li C, Zhang H, Wang Q, Li W, Fan Y. Machine learning predictors of risk of death within 7 days in patients with non-traumatic subarachnoid hemorrhage in the intensive care unit: A multicenter retrospective study. Heliyon 2024; 10:e23943. [PMID: 38192749 PMCID: PMC10772257 DOI: 10.1016/j.heliyon.2023.e23943] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/10/2023] [Revised: 11/04/2023] [Accepted: 12/15/2023] [Indexed: 01/10/2024] Open
Abstract
Non-traumatic subarachnoid hemorrhage (SAH) is a critical neurosurgical emergency with a high mortality rate, imposing a significant burden on both society and families. Accurate prediction of the risk of death within 7 days in SAH patients can provide valuable information for clinicians, enabling them to make better-informed medical decisions. In this study, we developed six machine learning models using the MIMIC III database and data collected at our institution. These models include Logistic Regression (LR), AdaBoosting (AB), Multilayer Perceptron (MLP), Bagging (BAG), Gradient Boosting Machines (GBM), and Extreme Gradient Boosting (XGB). The primary objective was to identify predictors of death within 7 days in SAH patients admitted to intensive care units. We employed univariate and multivariate logistic regression as well as Pearson correlation analysis to screen the clinical variables of the patients. The initially screened variables were then incorporated into the machine learning models, and the performance of these models was evaluated. Furthermore, we compared the performance differences among the six models and found that the MLP model exhibited the highest performance with an AUC of 0.913. In this study, we conducted risk factor analysis using Shapley values to identify the factors associated with death within 7 days in patients with SAH. The risk factors we identified include Gcsmotor, bicarbonate, wbc, spo2, heartrate, age, nely, glucose, aniongap, GCS, rbc, sysbp, sodium, and gcseys. To provide clinicians with a useful tool for assessing the risk of death within 7 days in SAH patients, we developed a web calculator based on the MLP machine learning model.
Collapse
Affiliation(s)
- Longyuan Gu
- Department of Neurosurgery, The Affiliated Hospital of Xuzhou Medical University, Xuzhou, Jiangsu, China
| | - Hongwei Hu
- Department of Neurosurgery, The Affiliated Hospital of Xuzhou Medical University, Xuzhou, Jiangsu, China
| | - Shinan Wu
- Xiamen University affiliated Xiamen Eye Center; Fujian Provincial Key Laboratory of Ophthalmology and Visual Science; Fujian Engineering and Research Center of Eye Regenerative Medicine; Eye Institute of Xiamen University; School of Medicine, Xiamen University, Xiamen, Fujian, China
| | - Fengda Li
- Department of Neurosurgery, Changshu Hospital Affiliated to Soochow University, Changshu, China
| | - Zeyi Li
- School of Computer Science, Nanjing University of Posts and Telecommunications, Nanjing, Jiangsu, China
| | - Yaodong Xiao
- Department of Neurosurgery, The Affiliated Hospital of Xuzhou Medical University, Xuzhou, Jiangsu, China
| | - Chuanqing Li
- Department of Neurosurgery, The Affiliated Hospital of Xuzhou Medical University, Xuzhou, Jiangsu, China
| | - Hui Zhang
- Department of Neurosurgery, The Affiliated Hospital of Xuzhou Medical University, Xuzhou, Jiangsu, China
| | - Qiang Wang
- Department of Neurosurgery, The Affiliated Hospital of Xuzhou Medical University, Xuzhou, Jiangsu, China
| | - Wenle Li
- The State Key Laboratory of Molecular Vaccinology and Molecular Diagnostics & Center for Molecular Imaging and Translational Medicine, School of Public Health, Xiamen University, Xiamen, China
| | - Yuechao Fan
- Department of Neurosurgery, The Affiliated Hospital of Xuzhou Medical University, Xuzhou, Jiangsu, China
| |
Collapse
|
2
|
Khan IR, Dar IA, Johnson TW, Loose E, Xu YY, Santiago E, Donohue KL, Marinescu MA, Gosev I, Schifitto G, Maddox RK, Busch DR, Choe R, Selioutski O. Correlations Between Quantitative EEG Parameters and Cortical Blood Flow in Patients Undergoing Extracorporeal Membrane Oxygenation With and Without Encephalopathy. J Clin Neurophysiol 2023:00004691-990000000-00108. [PMID: 37934074 DOI: 10.1097/wnp.0000000000001035] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/08/2023] Open
Abstract
PURPOSE The neurologic examination of patients undergoing extracorporeal membrane oxygenation (ECMO) is crucial for evaluating irreversible encephalopathy but is often obscured by sedation or neuromuscular blockade. Noninvasive neuromonitoring modalities including diffuse correlation spectroscopy and EEG measure cerebral perfusion and neuronal function, respectively. We hypothesized that encephalopathic ECMO patients with greater degree of irreversible cerebral injury demonstrate less correlation between electrographic activity and cerebral perfusion than those whose encephalopathy is attributable to medications. METHODS We performed a prospective observational study of adults undergoing ECMO who underwent simultaneous continuous EEG and diffuse correlation spectroscopy monitoring. (Alpha + beta)/delta ratio and alpha/delta Rartio derived from quantitative EEG analysis were correlated with frontal cortical blood flow index. Patients who awakened and followed commands during sedation pauses were included in group 1, whereas patients who could not follow commands for most neuromonitoring were placed in group 2. (Alpha + beta)/delta ratio-blood flow index and ADR-BFI correlations were compared between the groups. RESULTS Ten patients (five in each group) underwent 39 concomitant continuous EEG and diffuse correlation spectroscopy monitoring sessions. Four patients (80%) in each group received some form of analgosedation during neuromonitoring. (Alpha + beta)/delta ratio-blood flow index correlation was significantly lower in group 2 than group 1 (left: 0.05 vs. 0.52, P = 0.03; right: -0.12 vs. 0.39, P = 0.04). Group 2 ADR-BFI correlation was lower only over the right hemisphere (-0.06 vs. 0.47, P = 0.04). CONCLUSIONS Correlation between (alpha + beta)/delta ratio and blood flow index were decreased in encephalopathic ECMO patients compared with awake ones, regardless of the analgosedation use. The combined use of EEG and diffuse correlation spectroscopy may have utility in monitoring cerebral function in ECMO patients.
Collapse
Affiliation(s)
| | - Irfaan A Dar
- Biomedical Engineering, University of Rochester Medical Center, Rochester, New York, U.S.A
| | | | - Emily Loose
- School of Arts and Sciences, University of Rochester, Rochester, New York, U.S.A
| | - Yama Y Xu
- School of Arts and Sciences, University of Rochester, Rochester, New York, U.S.A
| | - Esmeralda Santiago
- School of Arts and Sciences, University of Rochester, Rochester, New York, U.S.A
| | - Kelly L Donohue
- Department of Neurology, Northwestern University Feinberg School of Medicine, Chicago, Illinois, U.S.A
| | - Mark A Marinescu
- Department of Medicine, University of Rochester Medical Center, Rochester, New York, U.S.A
| | - Igor Gosev
- Division of Cardiac Surgery, Department of Surgery, University of Rochester Medical Center, Rochester, New York, U.S.A
| | | | - Ross K Maddox
- Biomedical Engineering, University of Rochester Medical Center, Rochester, New York, U.S.A
- Department of Neuroscience, University of Rochester Medical Center, Rochester, New York, U.S.A
| | - David R Busch
- Department of Anesthesiology and Pain Management, University of Texas Southwestern Medical Center, Dallas, Texas, U.S.A
| | - Regine Choe
- Biomedical Engineering, University of Rochester Medical Center, Rochester, New York, U.S.A
- Department of Electrical and Computer Engineering, University of Rochester, Rochester, New York, U.S.A.; and
| | - Olga Selioutski
- Departments of Neurology and
- Department of Neurology, University of Mississippi, Jackson, Mississippi, U.S.A
| |
Collapse
|