Predicting long-term mortality of patients with postoperative acute kidney injury following noncardiac general anesthesia surgery using machine learning.

Background: This study addresses the gap in knowledge regarding the long-term mortality implications of postoperative acute kidney injury (PO-AKI) utilizing advanced machine learning techniques to predict outcomes more accurately than traditional statistical models. Methods: A retrospective cohort study was conducted using data from seven institutions between March 2009 and December 2019. Machine learning models were developed to predict all-cause mortality of PO-AKI patients using 23 preoperative variables and one postoperative variable. Model performance was compared to a traditional statistical approach with Cox regression analysis. The concordance index was used as a predictive performance metric to compare prediction capabilities among different models. Results: Among 199,403 patients, 2,105 developed PO-AKI. During a median follow-up of 144 months (interquartile range, 99.61-170.71 months), 472 in-hospital deaths occurred. Subjects with PO-AKI had a significantly lower survival rate than those without PO-AKI (p < 0.001). For predicting mortality, the XGBoost with an accelerated failure time model had the highest concordance index (0.7521), followed by random survival forest (0.7371), multivariable Cox regression model (0.7318), survival support vector machine (0.7304), and gradient boosting (0.7277). Conclusion: XGBoost with an accelerated failure time model was developed in this study to predict long-term mortality associated with PO-AKI. Its performance was superior to conventional models. The application of machine learning techniques may offer a promising approach to predict mortality following PO-AKI more accurately, providing a basis for developing targeted interventions and clinical guidelines to improve patient outcomes.

Characterization of a Small Plaque Variant Derived from Genotype V Japanese Encephalitis Virus Clinical Isolate K15P38.

Genotype V (GV) Japanese encephalitis virus (JEV) has been predominantly reported in the Republic of Korea (ROK) since 2010. GV JEV exhibits higher virulence and distinct antigenicity compared to other genotypes, which results in reduced efficacy of existing vaccines. Research on GV JEV is essential to minimize its clinical impact, but the only available clinical strain in the ROK is K15P38, isolated from the cerebrospinal fluid of a patient in 2015. We obtained this virus from National Culture Collection for Pathogens (NCCP) and isolated a variant forming small plaques during our research. We identified that this variant has one amino acid substitution each in the PrM and NS5 proteins compared to the reported K15P38. Additionally, we confirmed that this virus exhibits delayed propagation in vitro and an attenuated phenotype in mice. The isolation of this variant is a critical reference for researchers intending to study K15P38 obtained from NCCP, and the mutations in the small plaque-forming virus are expected to be useful for studying the pathology of GV JEV.

Characterization of genotype V Japanese encephalitis virus isolates from Republic of Korea.

Japanese encephalitis (JE), caused by the Japanese encephalitis virus (JEV) infection, continues to pose significant public health challenges worldwide despite efficient vaccines. The virus is classified into five genotypes, among which genotype V (GV) was not detected for a long period after its initial isolation in 1952, until reports emerged from China and the Republic of Korea (ROK) since 2009. The characteristics of the virus are crucial in estimating its potential epidemiological impact. However, characterization of GV JEVs has so far been limited to two strains: Muar, the original isolate, and XZ0934, isolated in China. Two additional ROK GV JEV isolates, NCCP 43279 and NCCP 43413, are currently available, but their characteristics have not been explored. Our phylogenetic analysis revealed that GV virus sequences from the ROK segregate into two clades. NCCP 43279 and NCCP 43413 belong to different clades and exhibit distinct in vitro phenotypes. NCCP 43279 forms larger plaques but demonstrates inefficient propagation in cell culture compared to NCCP 43413. In vivo, NCCP 43279 induces higher morbidity and mortality in mice than NCCP 43413. Notably, NCCP 43279 shows more severe blood-brain barrier damage, suggesting superior brain invasion capabilities. Consistent with its higher virulence, NCCP 43279 displays more pronounced histopathological and immunopathological outcomes. In conclusion, our study confirms that the two ROK isolates are not only classified into different clades but also exhibit distinct in vitro and in vivo characteristics.

Efficacy of genotype-matched vaccine against re-emerging genotype V Japanese encephalitis virus.

Japanese encephalitis (JE), caused by the Japanese encephalitis virus (JEV), is a highly threatening disease with no specific treatment. Fortunately, the development of vaccines has enabled effective defense against JE. However, re-emerging genotype V (GV) JEV poses a challenge as current vaccines are genotype III (GIII)-based and provide suboptimal protection. Given the isolation of GV JEVs from Malaysia, China, and the Republic of Korea, there is a concern about the potential for a broader outbreak. Under the hypothesis that a GV-based vaccine is necessary for effective defense against GV JEV, we developed a pentameric recombinant antigen using cholera toxin B as a scaffold and mucosal adjuvant, which was conjugated with the E protein domain III of GV by genetic fusion. This GV-based vaccine antigen induced a more effective immune response in mice against GV JEV isolates compared to GIII-based antigen and efficiently protected animals from lethal challenges. Furthermore, a bivalent vaccine approach, inoculating simultaneously with GIII- and GV-based antigens, showed protective efficacy against both GIII and GV JEVs. This strategy presents a promising avenue for comprehensive protection in regions facing the threat of diverse JEV genotypes, including both prevalent GIII and GI as well as emerging GV strains.

The comparative efficacy and safety of basiliximab and antithymocyte globulin in deceased donor kidney transplantation: a multicenter cohort study.

BACKGROUND: Generally, an induction agent is chosen based on the conditions of the deceased donor and the recipient. Antithymocyte globulin (ATG) is preferred in relatively high-risk conditions. No clear evidence indicates which induction agent is safer or more efficient for deceased donor kidney transplantation (DDKT). This study compares the efficacy and safety of basiliximab (BSX) and ATG according to donor characteristics in DDKT. METHODS: A total of 724 kidney transplant recipients from three transplant centers were enrolled, and propensity score matching was performed. Based on a donor age of 60 years, donor kidney with acute kidney injury (AKI), and Kidney Donor Profile Index (KDPI) score of 65%, we investigated how the choice of induction therapy agent affected the posttransplant clinical outcomes of delayed graft function (DGF), acute rejection (AR), infectious complications, and allograft and patient survival. RESULTS: AR and DGF did not differ significantly according to induction agent in elderly/young donor, AKI/non-AKI, and high-KDPI/ low-KDPI subgroups. The infection rate did not show meaningful differences. The differences in death-censored allograft survival and patient survival rates between induction agents were not statistically significant. CONCLUSION: Our study suggests that BSX can produce clinical outcomes similarly favorable to those of ATG even in DDKT cases with relatively poor donor conditions. Nonetheless, the donor and recipient conditions, immunological risk, and infection risk must be all taken into consideration when choosing an induction agent. Therefore, clinicians should carefully select the induction therapy agent for DDKT based on the risks and benefits in each DDKT case.

Hazardous characteristics of dust waste from metal manufacturing industries in South Korea.

The Ministry of Environment, South Korea, is involved in efforts to reduce safety-related accidents that could occur during hazardous waste disposal and minimise the environmental impacts of waste disposal. To reach such goals, new factors have been added that contribute to the hazardous characteristics of hazardous waste. The Ministry is also expanding regulations on inorganic and organic components present in hazardous waste for these factors and continues to correct the classification system to establish standards for waste components. Metallic dust, for example, is mostly generated during the operation of melting furnaces or precipitators and dangerous materials contained in this dust may cause accidents, such as explosions or fires. South Korean accident cases have confirmed that waste containing hazardous materials, such as Mg, Al and Mg-Al alloys, can cause such events. Therefore, this study analysed 28 dust samples collected from the dust accumulated in precipitators in 28 metal manufacturing facilities in South Korea. Nine samples were flammable and four samples were reactive to water. Two samples also exhibited leaching toxicities. The results applied to relevant laws in South Korea showed that the proportion of waste containing hazardous characteristics increased from 7.14% in the previous management category to 39.29% in the extended category, an overall increase of 32.15 percentage points.

Effects of Cognitive Task Training on Dynamic Balance and Gait of Patients with Stroke: A Preliminary Randomized Controlled Study.

BACKGROUND The decreased postural control ability of stroke patients affects their ability to balance in various postures such as sitting and standing. This study aimed to determine whether cognitive task training for stroke patients is effective in improving walking and balancing abilities. MATERIAL AND METHODS Seventeen stroke patients (10 males, 7 females) were randomized by ballot to be assigned to the cognitive task group (CBT) or the general task group (GBT). For the cognitive task training, a dual task of balance and cognition using traffic signals, a familiar form to the subjects, was applied as a program. In both groups the interventions were performed for 30 min a day, 3 times a week, for 4 weeks. The timed up and go test (TUG), the Berg balance scale (BBS), and gait ability evaluation were performed to compare the therapeutic effects. RESULTS After the intervention, the BBS showed significant differences in both groups (p<0.05). The cognitive task training group had significant improvement in all outcome scores after the intervention (p<0.05). The TUG score of the CBT group significantly decreased to 6.17 s (p<0.05), but that of the GBT showed no statistically significant change. CONCLUSIONS Cognitive task training could be used in clinical rehabilitation as a more effective intervention method to improve balance and gait ability of stroke patients.

Composition and abundance of marine debris stranded on the beaches of Sri Lanka: Results from the first island-wide survey.

This study provides the first assessment of marine debris washed ashore on 22 beaches along the coast of Sri Lanka. There was an average of 4.1 large (>25 mm) and 158 small (5-25 mm) pieces of debris per square meter of beach. Classified by use, packaging material (55%) dominated the debris, followed by consumer products (25%) and fishing gear (20%). In terms of materials, plastic was the greatest contributor (93%) to marine debris. Beaches near a river mouth or city and those with a barrier had greater debris accumulations. The east coast had significantly greater small debris density, possibly due to strong northeastern monsoon currents. Large spatial heterogeneity was observed in the amount of debris. Long-term monitoring is imperative to better understand the temporal changes in, and the pathways and possible management of, marine debris.

Distribution and Size Relationships of Plastic Marine Debris on Beaches in South Korea.

The characteristics of the distribution of plastic marine debris were determined on 12 beaches in South Korea in 2013 and 2014. The abundances of large micro- (1-5 mm), meso- (5-25 mm), and macroplastics (>25 mm) were 880.4, 37.7, and 1.0 particles/m(2), respectively. Styrofoam was the most abundant debris type for large microplastics and mesoplastics (99.1 and 90.9 %, respectively). Fiber (including fabric) was the most abundant of the macroplastics (54.7 %). There were no statistical differences in the mean numbers and weights of plastic debris among three beach groups from west, south, and east coasts. No significant differences were detected between the abundances of beached plastics in high strandline and backshore for all three size groups. Spearman's rank correlation was used to determine the relationships between the three debris size classes. The abundance of large microplastics was strongly correlated with that of mesoplastics for most material types, which suggests that the contamination level of large microplastics can be estimated from that of mesoplastics. As surveying of smaller particles is more labor intensive, the surveying of mesoplastics with a 5-mm sieve is an efficient and useful way to determine "hot-spots" on beaches contaminated with large microplastics.