A two-tier feature selection method for predicting mortality risk in ICU patients with acute kidney injury.

Acute kidney injury (AKI) is one of the most important lethal factors for patients admitted to intensive care units (ICUs), and timely high-risk prognostic assessment and intervention are essential to improving patient prognosis. In this study, a stacking model using the MIMIC-III dataset with a two-tier feature selection approach was developed to predict the risk of in-hospital mortality in ICU patients admitted for AKI. External validation was performed using separate MIMIC-IV and eICU-CRD. The area under the curve (AUC) was calculated using the stacking model, and features were selected using the Boruta and XGBoost feature selection methods. This study compares the performance of a stacking model using two-tier feature selection with a model using single-tier feature selection (XGBoost: 85; Boruta: 83; two-tier: 0.91). The predictive effectiveness of the stacking model was further validated by using different datasets (Validation 1: 0.83; Validation 2: 0.85) and comparing it with a simpler model and traditional clinical scores (SOFA: 0.65; APACH IV: 0.61). In addition, this study combined interpretable techniques and causal inference to analyze the causal relationship between features and predicted outcomes.

Experimental Study on the Effects of Matric Suction on Shear Properties of Polypropylene Fiber Reinforced Unsaturated Clay.

Matric suction has an important effect on the behavior of unsaturated soils, and polypropylene fibers are often used to improve soil. In order to probe into the mechanism of matric suction and fiber reinforcement, triaxial shear tests are carried out with changing matric suction and net confining pressure. Unsaturated clay in the Shaoxing section of East Zhejiang Grand Canal is selected with polypropylene fiber as a reinforcement material in the tests. The results show that the total cohesion intercept and effective internal friction angle of soil increase with the increase in matric suction, while the adsorption internal friction angle decreases gradually. Similarly, the contribution of matric suction to shear strength decreases. The total cohesion intercept is more sensitive to matric suction. As the length of fiber is 12 mm, the shear strength parameters of soil will be improved accordingly, which makes the fiber reinforcement achieve the best. The stress-strain relationship is approximately hyperbolic and strain hardening. The characteristics of strain hardening are more obvious with the increase in matric suction, and the soil specimens present plastic failure. The volumetric strain of specimen is more sensitive to the changing net confining pressure. It increases with the increase in net confining pressure, and increases linearly as the matric suction is zero. The failure modes of triaxial tests are divided into tensile failure and friction failure.

An evaluation of adjuvant chemotherapy following neoadjuvant chemotherapy and resection for borderline resectable and locally advanced pancreatic cancer.

BACKGROUND: In borderline resectable and locally advanced (BRLA) pancreatic cancer patients, the role of adjuvant therapy (AT) after neoadjuvant therapy (NAT) and curative-intent resection is poorly understood. METHODS: Using the National Cancer Database (NCDB) between 2011 and 2017, we identified BRLA patients who received NAT and resection. Kaplan-Meier analysis and multivariable Cox proportional hazards (PH) regression were performed to examine the association between AT and overall survival (OS). RESULTS: Of 17,905 BRLA patients identified, 764 received NAT and resection, of which 203 received AT. Median age was 63 years, and 53.1% were female. Kaplan Meier analysis revealed no differences in median OS between AT vs non-AT groups (28.9 vs 30.1months, p = 0.498). In the multivariable Cox PH model, after adjusting for other factors, when margin was positive, AT was associated with an improved survival (HR 0.54, 95%CI 0.32-0.90, p = 0.031). CONCLUSION: AT was not associated with survival in BRLA patients who received NAT and resection except in patients with positive margins. Further research is necessary to better understand the role of AT following NAT in patients with BRLA.