Dataset Information

Comparison of Conventional Logistic Regression and Machine Learning Methods for Predicting Delayed Cerebral Ischemia After Aneurysmal Subarachnoid Hemorrhage: A Multicentric Observational Cohort Study.

ABSTRACT:

Background

Timely and accurate prediction of delayed cerebral ischemia is critical for improving the prognosis of patients with aneurysmal subarachnoid hemorrhage. Machine learning (ML) algorithms are increasingly regarded as having a higher prediction power than conventional logistic regression (LR). This study aims to construct LR and ML models and compare their prediction power on delayed cerebral ischemia (DCI) after aneurysmal subarachnoid hemorrhage (aSAH).

Methods

This was a multicenter, retrospective, observational cohort study that enrolled patients with aneurysmal subarachnoid hemorrhage from five hospitals in China. A total of 404 aSAH patients were prospectively enrolled. We randomly divided the patients into training (N = 303) and validation cohorts (N = 101) according to a ratio of 75-25%. One LR and six popular ML algorithms were used to construct models. The area under the receiver operating characteristic curve (AUC), accuracy, balanced accuracy, confusion matrix, sensitivity, specificity, calibration curve, and Hosmer-Lemeshow test were used to assess and compare the model performance. Finally, we calculated each feature of importance.

Results

A total of 112 (27.7%) patients developed DCI. Our results showed that conventional LR with an AUC value of 0.824 (95%CI: 0.73-0.91) in the validation cohort outperformed k-nearest neighbor, decision tree, support vector machine, and extreme gradient boosting model with the AUCs of 0.792 (95%CI: 0.68-0.9, P = 0.46), 0.675 (95%CI: 0.56-0.79, P < 0.01), 0.677 (95%CI: 0.57-0.77, P < 0.01), and 0.78 (95%CI: 0.68-0.87, P = 0.50). However, random forest (RF) and artificial neural network model with the same AUC (0.858, 95%CI: 0.78-0.93, P = 0.26) were better than the LR. The accuracy and the balanced accuracy of the RF were 20.8% and 11% higher than the latter, and the RF also showed good calibration in the validation cohort (Hosmer-Lemeshow: P = 0.203). We found that the CT value of subarachnoid hemorrhage, WBC count, neutrophil count, CT value of cerebral edema, and monocyte count were the five most important features for DCI prediction in the RF model. We then developed an online prediction tool (https://dynamic-nomogram.shinyapps.io/DynNomapp-DCI/) based on important features to calculate DCI risk precisely.

Conclusions

In this multicenter study, we found that several ML methods, particularly RF, outperformed conventional LR. Furthermore, an online prediction tool based on the RF model was developed to identify patients at high risk for DCI after SAH and facilitate timely interventions.

Clinical trial registration

http://www.chictr.org.cn, Unique identifier: ChiCTR2100044448.

SUBMITTER: Hu P

PROVIDER: S-EPMC9247265 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:Delayed cerebral ischemia (DCI) is a common secondary complication and an important cause of disability and mortality among patients who survive aneurysmal subarachnoid hemorrhage (aSAH). Knowledge on DCI pathogenesis, risk factors, and biomarkers are essential for early detection and improved prognosis. To investigate the role of DNA methylation in DCI risk, we conducted an epigenome-wide association study (EWAS) in 68 patients followed up to 1 year after the initial aneurysm rupture. Blood samples were collected within 48 h post hemorrhage and used for DNA methylation profiling at ~ 450k CpG sites. A separate cohort of 175 patients was sequenced for the top CpG sites from the discovery analysis for a replication of the EWAS findings. EWAS did not identify any epigenome-wide significant CpGs. The top signal, cg18031596, was annotated to ANGPT1, a gene with critical functions in angiogenesis after vascular injury. Post hoc power calculations indicated a well-powered discovery analysis for cg18031596. Analysis of the replication cohort showed that four out of the five CpG sites sequenced at the ANGPT1 locus passed a Bonferroni-adjusted significance threshold. In a pooled analysis of the entire sample, three out of five yielded a significant p-value, and the top association signal (p-value = 0.004) was seen for a CpG that was not originally measured in the discovery EWAS. However, four ANGPT1 CpG sites had an opposite effect direction in the replication analysis compared to the discovery EWAS, marking a failure of replication. We carefully examined this observed flip in directions and propose several possible explanations in addition to that it was a random chance that ANGPT1 ranked at the top in the discovery EWAS. We failed to demonstrate a significant and consistent effect of ANGPT1 methylation in DCI risk in two cohorts. Though the replication attempt to weaken the overall support of this gene, given its relevant function and top rank of significance in the EWAS, our results call for future studies of larger aSAH cohorts to determine its relevance for the occurrence of DCI.