Model-based Decision Research Articles

Reporting Standards for Bayesian Network Modelling.

Reproducibility is a key measure of the veracity of a modelling result or finding. In other research areas, notably in medicine, reproducibility is supported by mandating the inclusion of an agreed set of details into every research publication, facilitating systematic reviews, transparency and reproducibility. Governments and international organisations are increasingly turning to modelling approaches in the development and decision-making for policy and have begun asking questions about accountability in model-based decision making. The ethical issues of relying on modelling that is biased, poorly constructed, constrained by heroic assumptions and not reproducible are multiplied when such models are used to underpin decisions impacting human and planetary well-being. Bayesian Network modelling is used in policy development and decision support across a wide range of domains. In light of the recent trend for governments and other organisations to demand accountability and transparency, we have compiled and tested a reporting checklist for Bayesian Network modelling which will bring the desirable level of transparency and reproducibility to enable models to support decision making and allow the robust comparison and combination of models. The use of this checklist would support the ethical use of Bayesian network modelling for impactful decision making and research.

A randomized control trial evaluating the advice of a physiological-model/digital twin-based decision support system on mechanical ventilation in patients with acute respiratory distress syndrome.

Acute respiratory distress syndrome (ARDS) is highly heterogeneous, both in its clinical presentation and in the patient's physiological responses to changes in mechanical ventilator settings, such as PEEP. This study investigates the clinical efficacy of a physiological model-based ventilatory decision support system (DSS) to personalize ventilator therapy in ARDS patients. This international, multicenter, randomized, open-label study enrolled patients with ARDS during the COVID-19 pandemic. Patients were randomized to either receive active advice from the DSS (intervention) or standard care without DSS advice (control). The primary outcome was to detect a reduction in average driving pressure between groups. Secondary outcomes included several clinically relevant measures of respiratory physiology, ventilator-free days, time from control mode to support mode, number of changes in ventilator settings per day, percentage of time in control and support mode ventilation, ventilation- and device-related adverse events, and the number of times the advice was followed. A total of 95 patients were randomized in this study. The DSS showed no significant effect on average driving pressure between groups. However, patients in the intervention arm had a statistically improved oxygenation index when in support mode ventilation (-1.41, 95% CI: -2.76, -0.08; p = 0.0370). Additionally, the ventilatory ratio significantly improved in the intervention arm for patients in control mode ventilation (-0.63, 95% CI: -1.08, -0.17, p = 0.0068). The application of the DSS led to a significantly increased number of ventilator changes for pressure settings and respiratory frequency. The use of a physiological model-based decision support system for providing advice on mechanical ventilation in patients with COVID-19 and non-COVID-19 ARDS showed no significant difference in driving pressure as a primary outcome measure. However, the application of approximately 60% of the DSS advice led to improvements in the patient's physiological state. clinicaltrials.gov, NCT04115709.

Carbonate rock physics model using different approaches to estimate rock frame stiffness

Deepwater carbonate reservoirs in the Brazilian pre-salt have emerged as significant hydrocarbon plays in the region and globally. Seismic monitoring of these reservoirs is crucial to provide information for well placement assessments, to reduce uncertainty in reservoir models, and to enhance model-based decision making. To integrate seismic data quantitatively, a petroelastic model (PEM) is required to estimate elastic properties. The modeling of the rock frame stiffness is an essential part in the study of PEM. However, this becomes more complex for carbonate rocks given their diverse pore-types. One way for estimating stiffness is to model each distinct pore-type and include them into the overall rock frame (different inclusion models). In this research, an alternative approach is explored to estimate the rock frame stiffness. A proxy model is employed, and its coefficients are subsequently calibrated with well-log information. Two PEMs were developed using different approaches for rock frame stiffness modeling. The first was inspired by inclusion models and incorporated two pore types (compliant and stiff pores) for the modeling process. The second considered a mathematical regression model as a proxy to relate reservoir porosity to the rock frame stiffness. These two PEMs were tested on the Brazilian pre-salt carbonate rocks of the Barra Velha (BVE) formation using well-log data from three wells and core sample measurements for one well. The accuracy of the PEMs’ performances was assessed by measures (such as, mean absolute error and root mean square error) and visual comparisons of their predictions. The results showed that both PEMs predicted velocities (compressional and shear) within an acceptable match with the actual well-log data. Similarly, when tested on the core samples, the PEMs produced comparable results, which indicates the validity of the proxy, not only at the well-log scale but at the core scale. The visual comparison and the accuracy analyses of the results for all three wells confirmed that the two PEMs had comparable predictions. This is significant given that the proxy simplifies the modeling process and facilitates the representation of various pore-types in the prediction of elastic properties in carbonate rocks. Overall, the proxy for the rock frame stiffness modeling is a computationally less expensive model compared to the inclusion models which involve modeling of various pore-types. It also offers a straightforward model which enables the quantitative integration of seismic data in a multidisciplinary team of geosciences and petroleum engineers (for instance, early engagement of petroleum engineers in 3D/4D seismic history matching).

Deep Learning Model-based Decision Support System for Kidney Cancer on Renal Images

Deep Learning Model-based Decision Support System for Kidney Cancer on Renal Images

CARETestLung: A mechanical test lung with Configurable airway Resistance, lung Elastance, and breathing efforts

A mechanical test lung is a crucial tool in accurately simulating patient-specific physiological responses of patients undergoing mechanical ventilation (MV), which, in turn, offer clinicians insight into lung mechanics during MV. In particular, it can be used to facilitate better methods to identify optimal ventilator settings, modes for individual patients by providing a platform to experiment with different MV settings. This addresses the challenge of optimising MV settings caused by variability in pathological conditions and the progression of respiratory disease over time within patients. However, the accessibility and cost of versatile test lungs limit widespread adoption in clinical settings, underscoring the need for affordable alternatives. This paper presents detailed instructions for the design and construction of a replicable, cost-effective mechanical test lung. The design features 3 subsystems: 1) the lung compartment; 2) the airway; and 3) a spontaneous breathing system. A detailed tests series shows its ability to replicate clinically realistic lung elastance values ranging from 25 to 85 cmH2O/L and airway resistance values from 10 to 45 cmH2O·s/L. It can also simulate a range of clinically realistic spontaneous breathing patterns. These capabilities yield pressure and flow ventilation data comparable to certified clinical test lungs across diverse scenarios, as well as matching clinically observed behaviours and dynamics. This accessible and versatile test lung offers valuable opportunities for optimising MV settings and advancing patient care, as well as its use in developing a range of physiological models for model-based decision support.

Innovative regression model-based decision support tool for optimizing radiotherapy techniques in thoracic esophageal cancer.

Modern radiotherapy exemplified by intensity-modulated radiation therapy (IMRT) and volumetric modulated arc therapy (VMAT), has transformed esophageal cancer treatment. Facing challenges in treating thoracic esophageal cancer near vital organs, this study introduces a regression model-based decision support tool for the optimal selection of radiotherapy techniques. We enrolled 106 patients diagnosed with locally advanced thoracic esophageal cancer in this study and designed individualized IMRT and VMAT radiotherapy plans for each patient. Detailed dosimetric analysis was performed to evaluate the differences in dose distribution between the two radiotherapy techniques across various thoracic regions. Single-factor and multifactorial logistic regression analyses were employed to establish predictive models (P1 and P2) and factors such as TLV/PTV ratio. These models were used to predict the compliance and potential advantages of IMRT and VMAT plans. External validation was performed in a validation group of 30 patients. Using predictive models, we developed a data-driven decision support tool. For upper thoracic cases, VMAT plans were recommended; for middle/lower thoracic cases, the tool guided VMAT/IMRT choices based on TLV/PTV ratio. Models P1 and P2 assessed IMRT and VMAT compliance. In validation, the tool showed high specificity (90.91%) and sensitivity (78.95%), differentiating IMRT and VMAT plans. Balanced performance in compliance assessment demonstrated tool reliability. In summary, our regression model-based decision support tool provides practical guidance for selecting optimal radiotherapy techniques for thoracic esophageal cancer patients. Despite a limited sample size, the tool demonstrates potential clinical benefits, alleviating manual planning burdens and ensuring precise, individualized treatment decisions for patients.

A model-based approach for self-adaptive security in CPS: Application to smart grids

Security risk assessment is an important challenge in the design of Cyber Physical Systems (CPS). Even more importantly, the intrinsically dynamical nature of these systems, due to changes in their environment, as well as evolutions in their infrastructures, makes them self-adaptive systems, where security aspects have to be considered in terms of management of detections and reactions for self-protection. In this work, we propose an approach to autonomously mitigate the threats in each reconfiguration at application or infrastructure levels of CPS. We propose and implement a framework for self-adaptive security: software architecture, design method, and integration with model-based decision. We use Attack-Defense Trees for modeling threats, and our approach involves security risk assessment, taking into account its balancing and coordination with quality-of-service aspects. We formulate and formalize the on-line decision problem to be solved at each cycle of the self-adaptation control loop in terms of Constraint Programming (CP) modeling and resolution. The CP model implements a set of constraints that allow to specify secure configurations, evaluated regarding their impact on system performance to pinpoint the most relevant one portraying a good balance between the security and quality of service. We perform validation of our approach with its application to Smart Grids, more particularly to an industrial case study from RTE (the French Energy Transmission company).

Model-based Decision Support System Using a System Dynamics Approach to Increase Corn Productivity

Background: As the population increases, the need for corn products also increases. Corn is needed for various purposes, such as food consumption, industry, and animal feed. Therefore, increasing corn production is crucial to support food availability and the food industry. Objective: The objective of this project is to create a model to increase corn farming productivity using scenarios from drip irrigation systems and farmer field school programs. Methods: A system dynamics approach is utilized to model the complexity and nonlinear behaviour of the corn farming system. In addition, several scenarios are formulated to achieve the objective of increasing corn productivity. Results: Simulation results showed that adopting a drip irrigation system and operating a farmer field school program would increase corn productivity. Conclusion: The corn farming system model was successfully developed in this research. The scenario of implementing a drip irrigation system and the farmer field school program allowed farmers to increase corn productivity. Through the scenario of implementing a drip irrigation system, farmers can save water use, thereby reducing the impact of drought. Meanwhile, the scenario of the farmer field school program enables farmers to manage agriculture effectively. This study suggests that further research could consider the byproducts of corn production to increase the profits of corn farmers. Keywords: Corn Farming, Decision Support System, Modeling, Simulation, System Dynamics

Development and prognostic validation of a three-level NHG-like deep learning-based model for histological grading of breast cancer

BackgroundHistological grade is a well-known prognostic factor that is routinely assessed in breast tumours. However, manual assessment of Nottingham Histological Grade (NHG) has high inter-assessor and inter-laboratory variability, causing uncertainty in grade assignments. To address this challenge, we developed and validated a three-level NHG-like deep learning-based histological grade model (predGrade). The primary performance evaluation focuses on prognostic performance.MethodsThis observational study is based on two patient cohorts (SöS-BC-4, N = 2421 (training and internal test); SCAN-B-Lund, N = 1262 (test)) that include routine histological whole-slide images (WSIs) together with patient outcomes. A deep convolutional neural network (CNN) model with an attention mechanism was optimised for the classification of the three-level histological grading (NHG) from haematoxylin and eosin-stained WSIs. The prognostic performance was evaluated by time-to-event analysis of recurrence-free survival and compared to clinical NHG grade assignments in the internal test set as well as in the fully independent external test cohort.ResultsWe observed effect sizes (hazard ratio) for grade 3 versus 1, for the conventional NHG method (HR = 2.60 (1.18–5.70 95%CI, p-value = 0.017)) and the deep learning model (HR = 2.27, 95%CI 1.07–4.82, p-value = 0.033) on the internal test set after adjusting for established clinicopathological risk factors. In the external test set, the unadjusted HR for clinical NHG 2 versus 1 was estimated to be 2.59 (p-value = 0.004) and clinical NHG 3 versus 1 was estimated to be 3.58 (p-value < 0.001). For predGrade, the unadjusted HR for predGrade 2 versus 1 HR = 2.52 (p-value = 0.030), and 4.07 (p-value = 0.001) for preGrade 3 versus 1 was observed in the independent external test set. In multivariable analysis, HR estimates for neither clinical NHG nor predGrade were found to be significant (p-value > 0.05). We tested for differences in HR estimates between NHG and predGrade in the independent test set and found no significant difference between the two classification models (p-value > 0.05), confirming similar prognostic performance between conventional NHG and predGrade.ConclusionRoutine histopathology assessment of NHG has a high degree of inter-assessor variability, motivating the development of model-based decision support to improve reproducibility in histological grading. We found that the proposed model (predGrade) provides a similar prognostic performance as clinical NHG. The results indicate that deep CNN-based models can be applied for breast cancer histological grading.

Cost-Effectiveness of Community-Based Active Case Finding Strategy for Tuberculosis: Evidence From Shenzhen, China.

Active case finding (ACF) is a potentially promising approach for the early identification and treatment of tuberculosis patients. However, evidence on its cost-effectiveness, particularly in low- and middle-income countries, remains limited. This study evaluates the cost-effectiveness of a community-based ACF practice in Shenzhen, China. We employed a Markov model-based decision analytic method to assess the costs and effectiveness of 3 tuberculosis detection strategies: passive case finding (PCF), basic ACF, and advanced ACF. The analysis was conducted from a societal perspective on a dynamic cohort over a 20-year horizon, focusing on active tuberculosis (ATB) prevalence and the incremental cost-effectiveness ratio (ICER). Compared to the PCF strategy, the basic and advanced ACF strategies effectively reduced ATB cases by 6.8 and 10.2 per 100 000 population, respectively, by the final year of this 20-year period. The ICER for the basic and advanced ACF strategies were ¥14 757 and ¥8217 per quality-adjusted life-year, respectively. Both values fell below the cost-effectiveness threshold. Our findings indicate that the community-based ACF screening strategy, which targets individuals exhibiting tuberculosis symptoms, is cost-effective. This underscores the potential benefits of adopting similar community-based ACF strategies for symptomatic populations in tuberculosis-endemic areas.

Correction of Thermal Errors in Machine Tools by a Hybrid Model Approach

Thermally induced position errors are one of the main error sources on the workpiece caused by the behavior of the machine tool. In today’s industrial environment, the correction of thermal errors is usually based on simple regression approaches, where the characteristic diagrams for correction are generated experimentally. The performance of these approaches is only valid for the corresponding load regimes, which often results in insufficient correction quality in practical applications. Consequently, there is only a limited benefit or even a deterioration in machine behavior if the correction characteristic is based on an inapplicable load case compared to the initial experiment. Simulation-generated characteristic diagrams using finite element models solve this disadvantage, but do not answer the question about the choice of the right characteristic matching the current load situation, and, in addition, calculate very slowly. Structural model-based correction using reduced models, on the other hand, calculates quickly, but requires a high modeling effort for accurate correction. The approach, presented in this contribution, combines simulation-generated characteristic diagrams and a structural model-based decision algorithm for a new hybrid model in order to select the appropriate characteristic diagram for the present load situation in the control system. This paper presents the simulative characteristic diagram generation by a finite element model validated by experiments in a climate chamber and a validated structural model including the concept for the decision algorithm.

Dosimetric comparison of IMPT vs VMAT for multiple lung lesions: an NTCP model-based decision-making strategy

To compare the dosimetric differences in volumetric modulated arc therapy (VMAT) and intensity modulated proton therapy (IMPT) in stereotactic body radiation therapy (SBRT) of multiple lung lesions and determine a normal tissue complication probability (NTCP) model-based decision strategy that determines which treatment modality the patient will use. A total of 41 patients were retrospectively selected for this study. The number of patients with 1-6 lesions was 5, 16, 7, 6, 3, and 4, respectively. A prescription dose of 70 GyRBE in 10 fractions was given to each lesion. SBRT plans were generated using VMAT and IMPT. All the IMPT plans used robustness optimization with ± 3.5% range uncertainties and 5 mm setup uncertainties. Dosimetric metrics and the predicted NTCP value of radiation pneumonitis (RP), esophagitis, and pericarditis were analyzed to evaluate the potential clinical benefits between different planning groups. In addition, a threshold for the ratio of PTV to lungs (%) to determine whether a patient would benefit highly from IMPT was determined using receiver operating characteristic curves. All plans reached target coverage (V70GyRBE ≥ 95%). Compared with VMAT, IMPT resulted in a significantly lower dose of most thoracic normal tissues. For the 1-2, 3-4 and 5-6 lesion groups, the lung V5 was 29.90 ± 9.44%, 58.33 ± 13.35%, and 81.02 ± 5.91% for VMAT and 11.34 ± 3.11% (p < 0.001), 21.45 ± 3.80% (p < 0.001), and 32.48 ± 4.90% (p < 0.001) for IMPT, respectively. The lung V20 was 12.07 ± 4.94%, 25.57 ± 6.54%, and 43.99 ± 11.83% for VMAT and 6.76 ± 1.80% (p < 0.001), 13.14 ± 2.27% (p < 0.01), and 19.62 ± 3.48% (p < 0.01) for IMPT. The Dmean of the total lung was 7.65 ± 2.47 GyRBE, 14.78 ± 2.75 GyRBE, and 21.64 ± 4.07 GyRBE for VMAT and 3.69 ± 1.04 GyRBE (p < 0.001), 7.13 ± 1.41 GyRBE (p < 0.001), and 10.69 ± 1.81 GyRBE (p < 0.001) for IMPT. Additionally, in the VMAT group, the maximum NTCP value of radiation pneumonitis was 73.91%, whereas it was significantly lower in the IMPT group at 10.73%. The accuracy of our NTCP model-based decision model, which combines the number of lesions and PTV/Lungs (%), was 97.6%. The study demonstrated that the IMPT SBRT for multiple lung lesions had satisfactory dosimetry results, even when the number of lesions reached 6. The NTCP model-based decision strategy presented in our study could serve as an effective tool in clinical practice, aiding in the selection of the optimal treatment modality between VMAT and IMPT.

Cold Climate Factors in Nitrogen Management for Maize

Among essential crop nutrients, nitrogen is the greatest management challenge in maize (Zea mays L.) production due to high requisite rates as well as dynamic transformations and losses. Climate plays a role in N management through changes in crop calendars, soil properties, agronomic practices, and yield effects. This study focuses on climate influences on maize N management and the objectives are to (i) review cold climate factors impacting economic optimum N rates (EONR), (ii) discuss approaches and climate considerations in estimating optimum N rates, and (iii) illustrate unexplored climate aspects related to optimum N rate assessment. Cold climate effects are expressed through inherent soil properties, agronomic management, and N fertilizer management. Most current N rate calculators do not explicitly account for climate factors, but implicitly integrate them through regional calibrations. Yield and EONR data from the US Corn Belt region indicate a positive correlation where lower means are associated with colder climates. High variability within climate regions is explained by differences in annual production environments, notably seasonal weather. Soil health models show that colder climates in the US are associated with higher stocks of soil organic matter, especially labile fractions. Adapt-N model simulations of a colder (North Central Wisconsin; 45.50, −89.70) and warmer (South Central Illinois; 38.50, −89.70) Corn Belt location show that higher soil organic N stocks do not increase crop N availability, presumably due to temperature-constrained N mineralization rates. The EONR for the colder site is 58 kg N ha−1 lower than the warmer site, which is well explained by differences in yield potential. Overall, abductive inferences suggest that colder climates are generally associated with higher levels of organic N stocks, but lower yields and crop N demands lessen EONRs. Seasonal weather and interactions with soil and agronomic factors also critically impact EONR, which can be assessed with model-based decision tools.

A Guide to an Iterative Approach to Model-Based Decision Making in Health and Medicine: An Iterative Decision-Making Framework.

Decision makers frequently face decisions about optimal resource allocation. A model-based economic evaluation can be used to guide decision makers in their choices by systematically evaluating the magnitude of expected health effects and costs of decision options and by making trade-offs explicit. We provide a guide to an iterative approach to the medical decision-making process by following a coherent framework, and outline the overarching iterative steps of model-based decision making. We systematized the framework by performing three steps. First, we compiled the existing guidelines provided by the ISPOR-SMDM Modeling Good Research Practices Task Force, and the ISPOR ValueofInformation Task Force. Second, we identified other previous work related to frameworks and guidelines for model-based decision analyses through a literature search in PubMed. Third, we assessed the role of the evidence and iterative process in decision making and formalized key steps in a model-based decision-making framework. We provide guidance on an iterative approach to medical decision making by applying the compiled iterative model-based decision-making framework. The framework formally combines the decision problem conceptualization (Part I), the model conceptualization and development (Part II), and the process of model-based decision analysis (Part III). Following the overarching steps of the framework ensures compliance to the principles of evidence-based medicine and regular updates of the evidence, given that valueofinformation analysis represents an essential component of model-based decision analysis in the framework. Following the provided guide and the steps outlined in the framework can help inform various health care decisions, and therefore it has the potential to improve decision making.

Exploratory modelling and analysis to support decision-making under deep uncertainty: A case study from defence resource planning and asset management

Emerging approaches and tools in the literature for model-based decision support under deep uncertainty are rapidly growing. Exploratory modelling and analysis is one of the unifying ideas behind these approaches. This research leverages exploratory modelling and analysis in the context of a case study within defense resource planning for the Royal Australian Navy. We use a multi-method approach that combines exploratory modelling and analysis and a simulation model. Our research analyses a strategy from a deterministic optimization problem to assess the impact of uncertainties and to identify different model behaviours exhibited under uncertainties and the uncertainty space in which the strategy maintains reasonable performance. To achieve these objectives, we draw upon concepts from Robust Decision-Making, Behaviour-Based Scenario Discovery, and Feasible Scenario Space. The results are then analysed to derive decision-making insights and highlight the potential consequences of neglecting uncertainties. For the first time in defense resource planning, we compare the results from these three approaches and tools. These approaches and tools complement each other, leading to similar conclusions with unique insights. The results also reveal the misleading nature of considering the best approximate conditions in the predictive use of simulation models under deep uncertainty.

Impact of epistemic uncertainty on tradeoff in model-based decision support for methane hydrate development system design

Global decarbonization trends encouraged the consideration of a methane hydrate (MH) development system design combined with carbon capture and storage (CCS) in the Japanese national MH research program. Previous studies have also proposed decarbonized MH development systems, but the conclusions vary due to epistemic uncertainty. Epistemic uncertainty arises from the knowledge and understanding of the subject, and each expert has different assumptions. Therefore, we assume that developing a model encompassing epistemic uncertainty as a platform allows for discussion toward system design decisions. This study builds the evaluation model, considering epistemic uncertainty to support the decision-making on the system design and reveal the tradeoff. It identifies the epistemic uncertainty affecting the tradeoff by changing the conditions based on previous studies. The evaluation model consists of technological options as modules, and all system concepts are evaluated by combining them. Epistemic uncertainty was subdivided and implemented within the evaluation model. As a result, the evaluation model output the tradeoff with bands of uncertainty. While the development field significantly impacts the results, hydrogen-related systems and offshore power plant systems score well owing to synergies between options. The epistemic uncertainty analysis regarding the tradeoff suggests that system design uncertainty impacts the conclusions. This finding indicates the importance of the model development as a platform for the evaluation of various system designs. This study contributed to an approach to model-based decision-making with epistemic uncertainty.

High-resolution grids of daily air temperature for Peru - the new PISCOt v1.2 dataset

Gridded high-resolution climate datasets are increasingly important for a wide range of modelling applications. Here we present PISCOt (v1.2), a novel high spatial resolution (0.01°) dataset of daily air temperature for entire Peru (1981–2020). The dataset development involves four main steps: (i) quality control; (ii) gap-filling; (iii) homogenisation of weather stations, and (iv) spatial interpolation using additional data, a revised calculation sequence and an enhanced version control. This improved methodological framework enables capturing complex spatial variability of maximum and minimum air temperature at a more accurate scale compared to other existing datasets (e.g. PISCOt v1.1, ERA5-Land, TerraClimate, CHIRTS). PISCOt performs well with mean absolute errors of 1.4 °C and 1.2 °C for maximum and minimum air temperature, respectively. For the first time, PISCOt v1.2 adequately captures complex climatology at high spatiotemporal resolution and therefore provides a substantial improvement for numerous applications at local-regional level. This is particularly useful in view of data scarcity and urgently needed model-based decision making for climate change, water balance and ecosystem assessment studies in Peru.

A Deep Learning-Based Decision Support Tool for Plant-Parasitic Nematode Management.

Plant-parasitic nematodes (PPN), especially sedentary endoparasitic nematodes like root-knot nematodes (RKN), pose a significant threat to major crops and vegetables. They are responsible for causing substantial yield losses, leading to economic consequences, and impacting the global food supply. The identification of PPNs and the assessment of their population is a tedious and time-consuming task. This study developed a state-of-the-art deep learning model-based decision support tool to detect and estimate the nematode population. The decision support tool is integrated with the fast inferencing YOLOv5 model and used pretrained nematode weight to detect plant-parasitic nematodes (juveniles) and eggs. The performance of the YOLOv5-640 model at detecting RKN eggs was as follows: precision = 0.992; recall = 0.959; F1-score = 0.975; and mAP = 0.979. YOLOv5-640 was able to detect RKN eggs with an inference time of 3.9 milliseconds, which is faster compared to other detection methods. The deep learning framework was integrated into a user-friendly web application system to build a fast and reliable prototype nematode decision support tool (NemDST). The NemDST facilitates farmers/growers to input image data, assess the nematode population, track the population growths, and recommend immediate actions necessary to control nematode infestation. This tool has the potential for rapid assessment of the nematode population to minimise crop yield losses and enhance financial outcomes.

Development and performance evaluation of an APP for vegetable fertilization and irrigation management originated from EU-Rotate_N

Overapplication of nitrogen fertilizer and water in vegetable production is popular and serious in China and other developing countries, leading to severe environmental pollution. To date, most model-based decision support systems, which are believed to be powerful tools for improving fertilizer and water management in vegetable production, have been established by developed countries, and the weather, soil, hydrology, facility type and vegetable species of developing countries have been less considered. In this study, a water level component and a new vegetable species, pakchoi (Brassica campestris ssp. Chinese) were incorporated into EU-Rotate_N, a simulation model developed by the European Union for open field vegetable cultivation. The experiment was carried out in typical vegetable plastic tunnels in Jiangsu Province, China, an area with high groundwater level. The results indicated that the water level algorithm which differentiate the soil into saturated and unsaturated layers enhanced the model simulation accuracy of the soil nitrogen and water content in the 30 cm soil layer, indicating the significant influence of groundwater on the soil water and nutrient movement. The pakchoi experiment suggest that new vegetables could be added into the crop list of EU-Rotate_N by adjusting parameters with ‘trial and error’ methods after considering its biological characteristics. Furthermore, a novel module was added for recommendations of fertilization and irrigation on daily basis in accordance with reasonable upper and lower thresholds of soil water content and soil available nitrogen after a series of adjustment. Finally, a user-friendly application (APP) was developed based on the improved model. The evaluation experiments in 2021 and 2022 showed that APP could significantly improve nitrogen and water use efficiency without obvious yield loss in pakchoi production. We conclude that simulation models for vegetable fertilizer and water management established by developed countries could be modified by several operable steps to expand its application area.

The importance of expert selection when identifying threatened ecosystems.

Identifying threatened ecosystem types is fundamental to conservation and management decision-making. When identification relies on expert judgment, decisions are vulnerable to inconsistent outcomes and can lack transparency. We elicited judgements of the occurrence of a widespread, critically endangered Australian ecosystem from a diverse pool of 83 experts. We asked 4 questions. First, how many experts are required to reliably conclude that the ecosystem is present? Second, how many experts are required to build a reliable model for predicting ecosystem presence? Third, given expert selection can narrow the range opinions, if enough experts are selected, do selection strategies affect model predictions? Finally, does a diverse selection of experts provide better model predictions? We used power and sample size calculations with a finite population of 200 experts to calculate the number of experts required to reliably assess ecosystem presence in a theoretical scenario. We then used boosted regression trees to model expert elicitation of 122 plots based on real-world data. For a reliable consensus (90% probability of correctly identifying presence and absence) in a relatively certain scenario (85% probability of occurrence), at least 17 experts were required. More experts were required when occurrence was less certain, and fewer were needed if permissible error rates were relaxed. In comparison, only ∼20 experts were required for a reliable model that could predict for a range of scenarios. Expert selection strategies changed modeled outcomes, often overpredicting presence and underestimating uncertainty. However, smaller but diverse pools of experts produced outcomes similar to a model built from all contributing experts. Combining elicited judgements from a diverse pool of experts in a model-based decision support tool provided an efficient aggregation of a broad range of expertise. Such models can improve the transparency and consistency of conservation and management decision-making, especially when ecosystems are defined based on complex criteria.