Search

Aiming at the problems of premature convergence of cuckoo search algorithm and difficulty in effectively identifying the fault modes of wind turbine gearboxes, an intelligent diagnosis technology based on BP neural network trained by the self-adaptive cuckoo search (SaCS-BP) algorithm is proposed. By constructing SaCS algorithm, the step size and discovery probability are adjusted adaptively, and a set of benchmark functions are employed to test the effectiveness of the proposed algorithm. Then, the fault diagnosis model of wind turbine gearbox is constructed by combining SaCS with BP neural network. Experimental results show that SaCS algorithm has better optimization accuracy and universality. Moreover, compared with BP neural network and BP network trained by cuckoo search algorithm (CS-BP), SaCS-BP method has the highest diagnostic accuracy, so as to realize the effective identification of the fault modes of wind turbine gearboxes.

Background The concentrations of linezolid, its optimal regimen and the associated side effects in elderly patients remain unclear. Methods In this multicentre, prospective study, elderly patients receiving linezolid at four tertiary hospitals in Beijing between May 2021 and December 2022 were included. Linezolid concentrations and haematological toxicity were monitored dynamically. Risk factors for linezolid overexposure and moderate-to-severe linezolid-induced thrombocytopenia (M/S LIT) were analysed, and a predictive model of M/S LIT was developed. Results A total of 860 linezolid concentrations were measured in 313 patients. The median trough concentrations of linezolid were 24.4 (15.3, 35.8) mg/L at 36–72 h and 26.1 (17.0, 38.1) mg/L at 5–10 days (P  = 0.132). Severe linezolid exposure was independently associated with age, estimated glomerular filtration rate (eGFR) and the worst SOFA score (SOFA1), and we further recommended dose regimens for elderly patients based on these findings. The incidences of linezolid-induced thrombocytopenia(LIT) and M/S LIT were 73.5% and 47.6%, respectively. M/S LIT was independently correlated with treatment duration, average trough concentration (TDMa), baseline platelet count, eGFR and baseline SOFA score (SOFA0). The developed nomogram predicted M/S LIT with an area under the curve of 0.767 (95% CI 0.715–0.820), a sensitivity of 71.1% and a specificity of 73.2%. Conclusions Linezolid trough concentrations increased dramatically in the elderly, by about 10 mg/L in patients aged 65–80 years, followed by a further increase of 10 mg/L for every 10 years of age. Therapeutic drug monitoring is recommended in elderly patients receiving linezolid. The developed nomogram may predict M/S LIT and guide dosage adjustments of linezolid. Clinical trial registration number : ChiCTR2100045707 [ABSTRACT FROM AUTHOR]

To explore a machine learning model for the early prediction of acute kidney injury (AKI) and to screen the related factors affecting new-onset AKI in the ICU. A retrospective analysis was performed used the MIMIC-III data source. New onset of AKI defined based on the serum creatinine changed. We included 19 variables for AKI assessment using four machine learning models: support vector machines, logistic regression, and random forest. and XGBoost, using accuracy, specificity, precision, recall, F1 score, and AUROC (area under the ROC curve) to evaluate model performance. The four models predicted new-onset AKI 3–6–9–12 h ahead. The SHapley Additive exPlanation (SHAP) value is used to evaluate the feature importance of the model. We finally respectively extracted 1130 AKI patients and non-AKI patients from the MIMIC-III database. With the extension of the early warning time, the prediction performance of each model showed a downward trend, but the relative performance was consistent. The prediction performance comparison of the four models showed that the XGBoost model performed the best in all evaluation indicators in all the time point at new-onset AKI 3–6–9–12 h ahead (accuracy 0.809 vs 0.78 vs 0.744 vs 0.741, specificity 0.856 vs 0.826 vs 0.797 vs 0.787, precision 0.842 vs 0.81 vs 0.775 vs 0.766, recall 0.759 vs 0.734 vs 0.692 vs 0.694, Fl score 0.799 vs 0.769 vs 0.731 vs 0.729, AUROC 0.892 vs 0.857 vs 0.827 vs 0.818). In the prediction of AKI 6, 9 and 12 h ahead, the importance of creatinine, platelets, and height was the most important based on SHapley. The machine learning model described in this study can predict AKI 3–6–9–12 h before the new-onset of AKI in ICU. In particular, platelet plays an important role. The new-onset of AKI in ICU is a common and important problem, which early be identified the risk of AKI can improve patients' outcomes. We explored MIMIC-III and determined the exact time point of occurrence of AKI as the basis for the new-onset of AKI in ICU. XGBoost model performed the best prediction in all the time point at new-onset AKI 3–6–9–12 h ahead. For patients with the new-onset of AKI in ICU, platelets become an important factor associated with AKI. [ABSTRACT FROM AUTHOR]