Search

Abstract This study aimed to develop a machine learning (ML)-based tool for early and accurate prediction of in-hospital mortality risk in patients with spontaneous intracerebral hemorrhage (sICH) in the intensive care unit (ICU). We did a retrospective study in our study and identified cases of sICH from the MIMIC IV (n = 1486) and Zhejiang Hospital databases (n = 110). The model was constructed using features selected through LASSO regression. Among five well-known models, the selection of the best model was based on the area under the curve (AUC) in the validation cohort. We further analyzed calibration and decision curves to assess prediction results and visualized the impact of each variable on the model through SHapley Additive exPlanations. To facilitate accessibility, we also created a visual online calculation page for the model. The XGBoost exhibited high accuracy in both internal validation (AUC = 0.907) and external validation (AUC = 0.787) sets. Calibration curve and decision curve analyses showed that the model had no significant bias as well as being useful for supporting clinical decisions. XGBoost is an effective algorithm for predicting in-hospital mortality in patients with sICH, indicating its potential significance in the development of early warning systems.

Background: Fractional flow reserve is widely used for the functional evaluation of coronary artery stenosis. Some studies have similarly used the translesional pressure ratio measurements for the functional evaluation of intracranial atherosclerotic stenosis. In this paper, we aimed to investigate the relationship between pressure ratio and cerebral tissue perfusion by MR perfusion imaging and provided a non-invasive method for evaluating the functional significance of intracranial atherosclerotic stenosis. Methods: A total of 18 consecutive patients with intracranial atherosclerotic stenosis patients including 19 stenotic vessels were recruited. The pressure was measured using a pressure guidewire, the pressure ratio before and after the endovascular intervention was calculated and compared with the severity of diameter stenosis and perfusion-derived MR (the time to maximum tissure residue function (Tmax)). Moreover, the DSA-derived pressure ratio was computed using a novel computational fluid dynamics-based model, termed CFD-PR, and was compared with the actual pressure ratio to assess its diagnostic accuracy. Results: The pressure ratio increased after percutaneous transluminal angioplasty or stenting, while the correlation between pressure ratio and diameter stenosis was not significant. The pressure ratio was negatively correlated with Tmax (r = −0.73, P

BackgroundWe aimed to evaluate the predictive power of systemic inflammation response index (SIRI), a novel biomarker, to predict all-cause mortality in patients with traumatic brain injury (TBI) in the intensive care unit (ICU).MethodsClinical data were retrieved from the Medical Information Mart for Intensive Care-IV (MIMIC-IV) database. Kaplan-Meier (KM) methods and cox proportional hazard models were performed to examine the association between SIRI and all-cause mortality. The predictive power of SIRI was evaluated compared to other leukocyte-related indexes including neutrophils, lymphocytes, monocytes and white blood cells (WBC) by the Receiver Operating Characteristic (ROC)curve for 30-day mortality. In addition, propensity score matching (PSM) was conducted to reduce confounding.ResultsA total of 350 TBI patients were enrolled overall in our study. The optimal cutoff point of SIRI was determined at 11.24 × 109/L. After 1:1 PSM, 66 matched pairs (132 patients) were generated. During the 30-day, in-hospital and 365-day follow-up periods, patients with low SIRI level were associated with improved survival (p < 0.05) compared with patients with high SIRI level. Cox regression analysis identified that higher SIRI values was an independent risk factor for all-cause mortality and results were stable on multiple subgroup analyses. Furthermore, ROC analysis indicated that the area under the curve of SIRI [0.6658 (95% Confidence Interval, 0.5630–0.7687)] was greater than that of neutrophils, monocytes, lymphocytes and WBC. The above results were also observed in the matched cohort.ConclusionIt was suggested that TBI patients with high SIRI level would suffer from a high risk of 30-day, in-hospital and 365-day mortality. SIRI is a promising inflammatory biomarker for predicting TBI patients' prognosis with relatively better predictive power than other single indicators related to peripheral differential leukocyte counts.