METHODS: In contrast, ViTs have demonstrated proficiency in capturing global signal patterns. In light of these observations, we propose a novel approach to enhance AD risk assessment. Our proposition involves a hybrid architecture, merging the strengths of CNNs and ViTs to compensate for their respective feature extraction limitations. Our proposed Dual-Branch Feature Fusion Network (DBN) leverages both CNN and ViT components to acquire texture features and global semantic information from EEG signals. These elements are pivotal in capturing dynamic electrical signal changes in the cerebral cortex. Additionally, we introduce Spatial Attention (SA) and Channel Attention (CA) blocks within the network architecture. These attention mechanisms bolster the model's capacity to discern abnormal EEG signal patterns from the amalgamated features. To make well-informed predictions, we employ a two-factor decision-making mechanism. Specifically, we conduct correlation analysis on predicted EEG signals from the same subject to establish consistency.
RESULTS: This is then combined with results from the Clinical Neuropsychological Scale (MMSE) assessment to comprehensively evaluate the subject's susceptibility to AD. Our experimental validation on the publicly available OpenNeuro database underscores the efficacy of our approach. Notably, our proposed method attains an impressive 80.23% classification accuracy in distinguishing between AD, Frontotemporal dementia (FTD), and Normal Control (NC) subjects.
DISCUSSION: This outcome outperforms prevailing state-of-the-art methodologies in EEG-based AD prediction. Furthermore, our methodology enables the visualization of salient regions within pathological images, providing invaluable insights for interpreting and analyzing AD predictions.
METHODS: This is a post hoc analysis of a cluster-randomized clinical trial that assesses the effect of implementing a feeding protocol on mortality in critically ill patients. Patients who stayed in the ICUs for at least 7 days and received exclusive EN were included in this analysis. Multivariable Cox hazard regression models and restricted cubic spline models were used to assess the relationship between the different doses of EN delivery and 28-day mortality. Subgroups with varying lactate levels at enrollment were additionally analyzed to address the potential confounding effect brought in by the presence of shock-related hypoperfusion.
RESULTS: Overall, 1322 patients were included in the analysis. The median (interquartile range) daily energy and protein delivery during the first week of enrollment were 14.6 (10.3-19.6) kcal/kg and 0.6 (0.4-0.8) g/kg, respectively. An increase of 5 kcal/kg energy delivery was associated with a significant reduction (approximately 14%) in 28-day mortality (adjusted hazard ratio [HR] = 0.865, 95% confidence interval [CI]: 0.768-0.974, P = 0.016). For protein intake, a 0.2 g/kg increase was associated with a similar mortality reduction with an adjusted HR of 0.868 (95% CI 0.770-0.979). However, the benefits associated with enhanced nutrition delivery could be observed in patients with lactate concentration ≤ 2 mmol/L (adjusted HR = 0.804 (95% CI 0.674-0.960) for energy delivery and adjusted HR = 0.804 (95% CI 0.672-0.962) for protein delivery, respectively), but not in those > 2 mmol/L.
CONCLUSIONS: During the first week of critical illness, enhanced nutrition delivery is associated with reduced mortality in critically ill patients receiving exclusive EN, only for those with lactate concentration ≤ 2 mmol/L.
TRIAL REGISTRATION: ISRCTN12233792, registered on November 24, 2017.