A hierarchical multilabel graph attention network method to predict the deterioration paths of chronic hepatitis B patients.

OBJECTIVE: Estimating the deterioration paths of chronic hepatitis B (CHB) patients is critical for physicians' decisions and patient management. A novel, hierarchical multilabel graph attention-based method aims to predict patient deterioration paths more effectively. Applied to a CHB patient data set, it offers strong predictive utilities and clinical value. MATERIALS AND METHODS: The proposed method incorporates patients' responses to medications, diagnosis event sequences, and outcome dependencies to estimate deterioration paths. From the electronic health records maintained by a major healthcare organization in Taiwan, we collect clinical data about 177 959 patients diagnosed with hepatitis B virus infection. We use this sample to evaluate the proposed method's predictive efficacy relative to 9 existing methods, as measured by precision, recall, F-measure, and area under the curve (AUC). RESULTS: We use 20% of the sample as holdouts to test each method's prediction performance. The results indicate that our method consistently and significantly outperforms all benchmark methods. It attains the highest AUC, with a 4.8% improvement over the best-performing benchmark, as well as 20.9% and 11.4% improvements in precision and F-measures, respectively. The comparative results demonstrate that our method is more effective for predicting CHB patients' deterioration paths than existing predictive methods. DISCUSSION AND CONCLUSION: The proposed method underscores the value of patient-medication interactions, temporal sequential patterns of distinct diagnosis, and patient outcome dependencies for capturing dynamics that underpin patient deterioration over time. Its efficacious estimates grant physicians a more holistic view of patient progressions and can enhance their clinical decision-making and patient management.

Effect of parenteral selenium supplementation in critically ill patients: a systematic review and meta-analysis.

BACKGROUND: It is currently unclear whether parenteral selenium supplementation should be recommended in the management of critically ill patients. Here we conducted a systematic review and meta-analysis to assess the efficacy of parenteral selenium supplementation on clinical outcomes. METHODS/PRINCIPAL FINDINGS: Randomized trials investigating parenteral selenium supplementation administered in addition to standard of care to critically ill patients were included. CENTRAL, Medline, EMBASE, the Science Citation Index, and CINAHL were searched with complementary manual searches. The primary outcome was all-cause mortality. Trials published in any language were included. Two authors independently extracted data and assessed trial quality. A third author was consulted to resolve disagreements and for quality assurance. Twelve trials were included and meta-analysis was performed on nine trials that recruited critically ill septic patients. These comprised 965 participants in total. Of these, 148 patients (30.7%) in the treatment groups, and 180 patients (37.3%) in control groups died. Parenteral selenium treatment significantly reduced all-cause mortality in critically ill patients with sepsis (relative risk [RR] 0.83, 95% CI 0.70-0.99, p = 0.04, I(2) = 0%). Subgroup analyses demonstrated that the administration schedule employing longer duration (RR 0.77, 95% CI 0.63-0.94, p = 0.01, I(2) = 0%), loading boluses (RR 0.73, 95% CI 0.58-0.94, p = 0.01, I(2) = 0%) or high-dose selenium treatment (RR 0.77, 95% CI 0.61-0.99, p = 0.04, I(2) = 0%) might be associated with a lower mortality risk. There was no evidence of adverse events. CONCLUSIONS/SIGNIFICANCE: Parenteral selenium supplementation reduces risk of mortality among critically ill patients with sepsis. Owing to the varied methodological quality of the studies, future high-quality randomized trials that directly focus on the effect of adequate-duration of parenteral selenium supplementation for severe septic patients are needed to confirm our results. Clinicians should consider these findings when treating this high-risk population. SYSTEMATIC REVIEW REGISTRATION: PROSPERO 2011; CRD42011001768.