A machine learning-based approach for predicting renal function recovery in general ward patients with acute kidney injury.

BACKGROUND: Acute kidney injury (AKI) is a significant challenge in healthcare. While there are considerable researches dedicated to AKI patients, a crucial factor in their renal function recovery, is often overlooked. Thus, our study aims to address this issue through the development of a machine learning model to predict restoration of kidney function in patients with AKI. METHODS: Our study encompassed data from 350,345 cases, derived from three hospitals. AKI was classified in accordance with the Kidney Disease: Improving Global Outcomes. Criteria for recovery were established as either a 33% decrease in serum creatinine levels at AKI onset, which was initially employed for the diagnosis of AKI. We employed various machine learning models, selecting 43 pertinent features for analysis. RESULTS: Our analysis contained 7,041 and 2,929 patients' data from internal cohort and external cohort respectively. The Categorical Boosting Model demonstrated significant predictive accuracy, as evidenced by an internal area under the receiver operating characteristic (AUROC) of 0.7860, and an external AUROC score of 0.7316, thereby confirming its robustness in predictive performance. SHapley Additive exPlanations (SHAP) values were employed to explain key factors impacting recovery of renal function in AKI patients. CONCLUSION: This study presented a machine learning approach for predicting renal function recovery in patients with AKI. The model performance was assessed across distinct hospital settings, which revealed its efficacy. Although the model exhibited favorable outcomes, the necessity for further enhancements and the incorporation of more diverse datasets is imperative for its application in real- world.

Changes in Flight Altitude of Black-Tailed Gulls According to Temporal and Environmental Differences.

In this study, GPS trackers were attached to black-tailed gulls (Larus crassirostris) breeding on five islands in Republic of Korea during April and May 2021, and their flight frequency, behavioral range, and flight altitude were compared during and after the breeding season. During the breeding season, the flight frequency was lowest on Dongman Island (28.7%), where mudflats were distributed nearby, and the range of activity was narrow. In contrast, it tended to be high on Gungsi Island (52%), where the breeding colony was far from land, resulting in a wider range of activity. Although the flight frequency on Dongman Island increased post-breeding season (42.7%), it decreased on other islands. The mean flight altitude during the breeding season was lowest on Dongman Island and highest on Napdaegi Island. In most breeding areas, the mean flight altitude during the post-breeding season was higher than that during the breeding season. However, the lead flight altitude was lower during the non-breeding season compared to that in the breeding season. The home range expanded after the breeding season, with no significant difference in lead time between the breeding and non-breeding seasons. Our findings reveal that black-tailed gulls exhibit varying home ranges and flight altitudes depending on season and geographical location. As generalists, gulls display flexible responses to environmental changes, indicating that flight behavior adapts to the evolving environment over time and across regions.

First report on the detection of microplastics from the feathers of black-tailed gulls in South Korea.

In this study, microplastics, including polyethylene (PE), polyethylene terephthalate (PET), polypropylene (PP), and polystyrene (PS), adhering to the feathers of all tracked black-tailed gull individuals were studied. PE was detected in the highest number of feathers (n = 26, 35.6 %), followed by PP (n = 21, 28.8 %), PET and other microplastics (n = 16, 21.9 %), and PS (n = 10, 13.7 %). Furthermore, plastic particles of size 50-100 µm were the most common (n = 33, 45.1 %), followed by ≤50 (n = 21, 28.8 %), 100-150 (n = 11, 15.1 %), ≥200 (n = 7, 9.6 %), and 150-200 µm (n = 1, 1.4 %). Microplastic levels did not differ considerably between the Dokdo and Ulleungdo populations. As black-tailed gulls spend >95 % of their time in coastal areas, coastal pollution caused by oil spills and increasing microplastic levels could lead to physical problems, such as the adherence of oil and microplastics onto feathers.

Design of a Smart Conducting Nanocomposite with an Extended Strain Sensing Range by Conjugating Hybrid Structures.

In recent years, flexible and wearable strain sensors, consisting of a polymer matrix and a conducting filler, have received extensive attention owing to their physical advantages, such as being lightweight, stretchable, and having the potential for application to complex forms. However, achieving a low hysteresis of the relative change in resistance, wide sensing range, and reduced plastic deformation is still challenging. To address these issues, in this study, we developed hybrid conducting composites with a wide range of sensing abilities and low hysteresis. The bi-layer composites, comprising a carbon nanotube (CNT) composite layer with reinforced/conducting properties, and a natural rubber-based layer with extreme strain properties, could effectively circumvent their limitations. Compared to single-layer CNT composites, the bi-layer structure could increase the tensile strain with reduced plastic deformation, resulting in the prevention of surface cracks on the CNT composite. In addition, it has the benefit of measuring a wider sensing range, which cannot be measured in a single-CNT composite system. A cyclic stretching/releasing test was performed to demonstrate that the strain sensor exhibited excellent reproducibility. Our results can function as a useful design guide for stretchable sensor applications.