Search

Objective: The healthcare challenge driven by an aging population and rising demand is one of the most pressing issues leading to emergency department (ED) overcrowding. An emerging solution lies in machine learning's potential to predict ED dispositions, thus leading to promising substantial benefits. This study's objective is to create a predictive model for ED patient dispositions by employing ensemble learning. It harnesses diverse data types, including structured and unstructured information gathered during ED visits to address the evolving needs of localized healthcare systems. Methods: In this cross-sectional study, 80,073 ED patient records were amassed from a major southern Taiwan hospital in 2018–2019. An ensemble model incorporated structured (demographics, vital signs) and pre-processed unstructured data (chief complaints, preliminary diagnoses) using bag-of-words (BOW) and term frequency-inverse document frequency (TF-IDF). Two random forest base-learners for structured and unstructured data were employed and then complemented by a multi-layer perceptron meta-learner. Results: The ensemble model demonstrates strong predictive performance for ED dispositions, achieving an area under the receiver operating characteristic curve of 0.94. The models based on unstructured data encoded with BOW and TF-IDF yield similar performance results. Among the structured features, the top five most crucial factors are age, pulse rate, systolic blood pressure, temperature, and acuity level. In contrast, the top five most important unstructured features are pneumonia, fracture, failure, suspect, and sepsis. Conclusions: Findings indicate that utilizing ensemble learning with a blend of structured and unstructured data proves to be a predictive method for determining ED dispositions. [ABSTRACT FROM AUTHOR]

Drying‐induced soil curling is commonly observed in nature. In order to investigate the change of water distribution in soil caused by drying and curling, a series of drying tests is performed on thin soil layers in the laboratory. A discrete element model (DEM) is then proposed to simulate soil desiccation curling. The outermost particles are selected as the surface of the model to address the importance of surface evaporation. Moisture distribution change during drying is governed by both surface evaporation and moisture exchange between DEM particles. The results show that the model can capture the main features of the soil curling process from concave curling (∪ $\cup $) to convex curling (∩ $\cap $). The model enables variation in moisture gradient during the drying process. The trend of the curling development obtained from the model agreed well with the laboratory observation. A newly exposed surfaces in the model can provide a more realistic moisture distribution and influence the curling behavior. Generally, the drying‐induced soil curling process and extent are associated with the change in moisture gradient along the depth. The movement of tensile stress concentration change from the upper layer in the early drying stage to the lower part of the sample in the late drying stage is the intrinsic mechanical factor responsible for the initial concave curling deformation and the subsequent fallback and convex curling phenomenon. Curling deformation in the present model is compared with other existing explanations on desiccation curling. The influences of evaporation rate and hydraulic conductivity on the model performance were also discussed. It is found that a higher evaporation rate and a lower hydraulic conductivity would result in a higher extent of curling during drying. Plain Language Summary: Drying‐induced soil curling is a common phenomenon in earth surface. This study presents a discrete element model (DEM) to simulate soil curling. Understanding soil curling phenomenon has been of particular interest in several environment, engineering and geosciences fields. An essential step toward this topic is to study the role of evaporation and moisture distribution, which dominates curling process. All of the existing models for soil curling simulation only consider the top surface as evaporation boundary. However, evaporation could happen from the side when perpendicular cracking opens up new surfaces. The onset of curling also creates new interface from the bottom of soil. The moisture distribution and the drying differences would then be significantly altered as the presence of these new evaporating boundaries. However, currently, no model in the literature accounts for the effect of the additional evaporation boundaries, and the relationship between evaporation‐induced moisture gradient and curling process. The current study aims to fill this gap by presenting new model to simulate soil curling phenomenon. Key Points: A discrete element model (DEM) is proposed to simulate soil desiccation curling and validated by experimental dataDrying‐induced soil curling process and extent is controlled by the evolution of moisture gradient along the depthThe movement of tensile stress concentration is the intrinsic mechanical factor responsible for the curling deformation process [ABSTRACT FROM AUTHOR]

Purpose: Serum cystatin C is a promising new biomarker of glomerular filtration rate and cardiovascular events, but few studies focused on serum cystatin C levels in obstructive sleep apnea (OSA) patients. The aim of our study was to evaluate the association between serum cystatin C and OSA in younger men (≤40 years old of age) without complications. Methods: We prospectively recruited consecutive participants without comorbidities who underwent polysomnography. Fasting blood samples were obtained from all subjects for biological profile measurements. Statistical analysis was used to evaluate the relationship between serum cystatin C and other parameters. Results: The population consisted of 98 subjects (mean age = 32.5 years, mean body mass index = 27.93 kg/m) that were divided according to polysomnographic finding into control group ( n = 23), mild ( n = 15), moderate ( n = 24), and severe ( n = 36) OSA group. Compared with the control group, patients with severe OSA were significantly heavier (body mass index, 29.69 ± 3.81 vs. 26.42 ± 3.10) and presented significantly higher levels of high sensitive C-reactive protein (hsCRP) (1.10 ± 0.28 vs. 0.88 ± 0.20 mg/l) and serum cystatin C (0.87 ± 0.12 vs. 0.74 ± 0.10 mg/l) ( p < 0.05 for all comparisons). Cystatin C was correlated with Apnea Hypopnea Index (AHI), Oxygen Desaturation Index, hsCRP, creatinine, and estimated glomerular filtration rate ( r = 0.319, 0.279, 0.321, 0.233, −0.241, p = 0.001, 0.005, 0.001, 0.021, 0.017, respectively). After adjustment for confounding factors, AHI was significantly and positively associated with serum cystatin C levels ( β = 0.284, p = 0.007). Conclusions: Our finding indicated that serum cystatin C was associated with the severity of OSA in younger men. Further study is needed to find out whether OSA patients with increased serum cystatin C levels are prone to subclinical cardiovascular and renal diseases. [ABSTRACT FROM AUTHOR]

Currently available application frameworks that target at the automatic design of real-time embedded software are poor in integrating functional and non-functional requirements for mobile and ubiquitous systems. In this work, we present the internal architecture and design flow of a newly proposed framework called Verifiable Embedded Real-Time Application Framework (VERTAF), which integrates three techniques namely software component-based reuse, formal synthesis, and formal verification. The proposed architecture for VERTAF is component-based which allows plug-and-play for the scheduler and the verifier. The architecture is also easily extensible because reusable hardware and software design components can be added. Application examples developed using VERTAF demonstrate significantly reduced relative design effort, which shows how high-level reuse of software components combined with automatic synthesis and verification increases design productivity. [ABSTRACT FROM AUTHOR]

Increased oxidative stress has been observed to contribute the development of insulin resistance. Oxidative stress is known to increase the conversion of deoxyguanosine (dG) to 8-hydroxy-2′-deoxyguanosine (8-OHdG). Human 8-oxoguanine glycosylase (hOGG1) is the key component responsible for the removal of 8-OHdG from oxidatively damaged DNA. The repair activity of the hOGG1 Ser326Cys gene variant has been demonstrated to be lower than that of the hOGG1 Ser/Ser genotype. Therefore, the possible association of the hOGG1 Ser326Cys gene variant with insulin sensitivity was investigated in 279 normal glucose-tolerant subjects without history of cancer. Allele frequency was 21.5% for the Ser/Ser genotype ( n=60), 45.9% for the Ser/Cys genotype ( n=128), and 32.6% for the Cys/Cys genotype ( n=91). Subjects carrying the Cys/Cys genotype had significantly lower insulin sensitivity levels, assessed by homeostasis model assessment-insulin resistance (HOMA-IR), compared with the Ser/Ser and Ser/Cys genotypes ( P<0.001 and P<0.001, respectively). In a multiple linear regression analysis, the Cys/Cys genotype was a significant determinant of HOMA-IR, independent of age, sex, body mass index, fasting plasma cholesterol, triglyceride, HDL cholesterol, LDL cholesterol, or hypertension. The present study indicates that the hOGG1 gene Cys/Cys variant is associated with a significant decrease in insulin sensitivity in subjects with normal glucose tolerance. [ABSTRACT FROM AUTHOR]

SUMMARY: Apolipoprotein E (apo E) regulates the metabolism of lipoproteins. Three different apo E alleles (ℇ2, ℇ3 and ℇ4) produce the apo E isoproteins, apo E2, apo E3 and apo E4. Previous studies had revealed that lipid abnormalities might contribute to the development and progression of kidney diseases. Thus, the apo E gene polymorphism may be associated with end-stage renal disease. To investigate the distribution of apo E genotype in patients under haemodialysis in Taiwan, we examined 205 dialysis patients and 103 age-matched normal controls. The apo E gene polymorphism was analysed by the use of polymerase chain reaction. Our study revealed that the frequency of apo E2 allele was significantly higher in dialysis patients (8.0%) than in normal controls (1.9%; P < 0.01). The E3 and E4 allele frequencies were not significantly different between the groups. Our findings indicate that apo E polymorphism was apparently associated with the development of end-stage renal disease in Taiwan. [ABSTRACT FROM AUTHOR]