A digital twin is a digital representation that closely resembles or replicates a real world object by combining interdisciplinary knowledge and advanced technologies. Digital twins have been applied to various fields, including to the agricultural field. Given big data and systematic data management, digital twins can be used for predicting future outcomes. In this study, we endeavor to create an agricultural digital twin using mandarins as a model crop. We employ an Open API to aggregate data from various sources across Jeju Island, covering an area of approximately 185,000 hectares. The collected data are visualized and analyzed at regional, inter-orchard, and intra-orchard scales. We observe that the intra-orchard analysis explains the variation of fruit quality substantially more than the inter-orchard analysis. Our data visualization and analysis, incorporating statistical models and machine learning algorithms, demonstrate the potential use of agricultural digital twins in the future, particularly in the context of micro-precision and individualized agriculture. This concept extends the current management practices based on data-driven decisions, and it offers a glimpse into the future of individualized agriculture by enabling customized treatment for plants, akin to personalized medicine for humans.

Each strawberry (Fragaria × ananassa) cultivar has its own growth and yield characteristics. However, the characteristics of many cultivars have not been determined at a consistent time and place, making direct comparative analysis difficult. The objective of this study was to identify characteristics and trends of five Korean strawberry cultivars in the same environment during an entire season. Therefore, environmental factors such as daily average air temperature, daily average relative humidity, daily average solar radiation, daily soil temperature, daily soil water content, daily soil electrical conductivity (EC), plant growth characteristics such as the number of leaves, plant height, leaf length, leaf width, and crown diameter, and productivity characteristics such as flowering and fruiting were measured to investigate the possible correlations of the data over one season. The vegetative growth of “Seolhyang” and “Keumsil” was greater than that of “Jukhyang” and “Maehyang”. The yield of “Arihyang” was greater than that of all other cultivars. “Arihyang” also presented the greatest weight per number of fruits. Among environmental factors, higher variability in air temperature and soil water content was correlated to lower total fresh weight in the following week at different degrees for each cultivar. Among the cultivars, the time to the first flowering was delayed by about seven days when the number of leaves increased by one and was reduced by one day per 1 cm increase in plant height. The total fresh weight was enhanced up to 271 g per experimental unit, while the average number of leaves increased by one. The results indicate that the data can be used by those who need information regarding the characteristics of the strawberry cultivars through direct comparative analysis.

Higher mean platelet volume (MPV) is an indicator of larger, reactive platelets, and has been associated with a higher risk of thrombosis and cardiovascular events in the general population. Hemodialysis patients have a higher risk for cardiovascular death and predisposition to platelet dysfunction (thrombosis and bleeding diathesis), but the relationship between MPV and mortality in this population is unknown. Among a 5-year cohort (1/2007-12/2011) of 149,118 incident hemodialysis patients from a large national dialysis organization, we examined the association between MPV and all-cause mortality. In primary analyses, we granularly analyzed MPV across five categories: 7.2-7.5, >7.5-9.5, >9.5-11.5, >11.5-13.5, and >13.5-15.0fL. In secondary analyses, we examined MPV categorized as low, normal, and high based on thresholds in the general population: 7.2-7.5, >7.5-11.5, and >11.5fL, respectively. Associations between baseline and time-dependent MPV with mortality were estimated using traditional and time-dependent Cox models in order to determine long-term and short-term exposure-mortality associations, respectively, using three adjustment levels: unadjusted, case-mix, and case-mix+laboratory models. In primary analyses, higher baseline and time-dependent MPV levels were associated with incrementally higher death risk in case-mix+laboratory analyses (reference: >9.5-11.5fL). In secondary analyses, high baseline and time-dependent MPV levels were associated with higher mortality, whereas low MPV was associated with lower death risk across all multivariable models (reference: normal MPV). Hemodialysis patients with higher MPV have heightened mortality risk. Further studies are needed to determine the pathophysiologic basis for the higher risk, and if modification of MPV ameliorates mortality in this population.

Fog is challenging to predict, and the accuracy of fog prediction may depend on location and time of day. Furthermore, accurate detection of fog is difficult, since, historically, it is often carried out based on visual observations which can be biased and are often not very frequent. Furthermore, visual observations are more challenging to make during the night. To overcome these limitations, we detected fog using FM-120 instruments, which continuously measured liquid water content in the air in the Monterey, California (USA), area. We used and compared the prediction performance of logistic regression (LR) and random forest (RF) models each evening between 5 pm and 9 pm, which is often the time when advection fog is generated in this coastal region. The relative performances of the models depended on the hours between 5 pm and 9 pm, and the two models often generated different predictions. In such cases, a consensus approach was considered by revisiting the past performance of each model and weighting more heavily the more trustworthy model for a given hour. The LR resulted in a higher sensitivity (hit rate) than the RF model early in the evening, but the overall performance of the RF was usually better than that of the LR. The consensus approach provided more robust prediction performance (closer to a better accuracy level between the two methods). It was difficult to conclude which of the LR and RF models was superior consistently, and the consensus approach provided robustness in 3 and 2 h forecasts.

There are two schools of thought in statistical analysis, frequentist, and Bayesian. Though the two approaches produce similar estimations and predictions in large-sample studies, their interpretations are different. Bland Altman analysis is a statistical method that is widely used for comparing two methods of measurement. It was originally proposed under a frequentist framework, and it has not been used under a Bayesian framework despite the growing popularity of Bayesian analysis. It seems that the mathematical and computational complexity narrows access to Bayesian Bland Altman analysis. In this article, we provide a tutorial of Bayesian Bland Altman analysis. One approach we suggest is to address the objective of Bland Altman analysis via the posterior predictive distribution. We can estimate the probability of an acceptable degree of disagreement (fixed a priori) for the difference between two future measurements. To ease mathematical and computational complexity, an interface applet is provided with a guideline.