Machine Learning for In-hospital Mortality Prediction in Critically Ill Patients With Acute Heart Failure

Machine Learning for In-hospital Mortality Prediction in Critically Ill Patients With Acute Heart Failure: A Retrospective Analysis Based on the MIMIC-IV Database

This study developed machine learning (ML) models to predict in-hospital mortality among ICU patients with acute heart failure (AHF) using data from the MIMIC-IV database. Among five tested algorithms, XGBoost showed the highest predictive accuracy (AUC: 0.82) and outperformed traditional clinical scoring systems. The model incorporated 18 clinical variables from the first 24 hours of ICU admission to aid early intervention strategies.

Read More »
CytoSorb Filter During Kidney Normothermic

Protocol for a Single-Centre Randomised Pilot Study to Assess the Safety and Feasibility of Adding a CytoSorb Filter During Kidney Normothermic Machine Perfusion to Remove Inflammatory and Immune Mediators Prior to Kidney Transplantation

This pilot study assesses the safety and feasibility of using a CytoSorb filter during kidney normothermic machine perfusion (NMP) before transplantation. The goal is to reduce inflammation-related damage by filtering immune mediators. Twenty patients will be randomized to receive kidneys preserved with or without the filter. Outcomes include immune gene expression, delayed graft function, and kidney performance at intervals post-transplant.

Read More »
Comparison of Del Nido and Histidine-Tryptophan-Ketoglutarate

Comparison of Del Nido and Histidine-Tryptophan-Ketoglutarate Cardioplegia Solutions: An Animal Study With Prolonged Ischaemia

This study compares Del Nido (DN) and Bretschneider histidine-tryptophan-ketoglutarate (HTK) cardioplegia solutions in a porcine model of prolonged ischaemia. DN demonstrated superior left ventricular function, spontaneous return to sinus rhythm, and less haemodilution and endothelial dysfunction compared to HTK. Histological and biochemical markers of injury, stress, and apoptosis showed no significant differences, suggesting DN offers enhanced myocardial protection during cardiac surgery.

Read More »
Quantum Perfusion

Real-Time Blood Gas Management: Evaluating Quantum Perfusion System’s Accuracy Against a Standard Blood Gas Analysis in CPB

This non-inferiority study evaluates the Quantum Perfusion System (QPS) for continuous blood gas monitoring during cardiopulmonary bypass (CPB) against the standard blood gas analyzer (BGA). Using data from 40 patients undergoing cardiac surgeries, QPS demonstrated high accuracy and agreement with BGA across all key parameters, adhering to Clinical Laboratory Improvement Amendments (CLIA) standards, and validating its utility as a trending device.

Read More »