Guidelines for model adaptation: a study of the transferability of a general seagrass ecosystem DBN model

Ecological models are extensively and increasingly used in support of environmental policy and decision making. Dynamic Bayesian Networks (DBN) as a tool for conservation have been demonstrated to be a valuable tool for providing a systematic and intuitive approach to integrating data and other critical information to help guide the decision-making process. However, data for a new ecosystem are often sparse. In this case, a general DBN developed for similar ecosystems could be applicable, but this may require the adaptation of key elements of the network. The research presented in this paper focused on a case study to identify and implement guidelines for model adaptation. We adapted a general DBN of a seagrass ecosystem to a new location where nodes were similar, but the conditional probability tables varied. We focused on two species of seagrass (Zostera noltei and Z. marina) located in Arcachon Bay, France. Expert knowledge was used to complement peer-reviewed literature to identify which components needed adjustment including parameterisation and quantification of the model and desired outcomes. We adopted both linguistic labels and scenario-based elicitation to elicit from experts the conditional probabilities used to quantify the DBN. Following the proposed guidelines, the model structure of the DBN was retained, but the conditional probability tables were adapted for nodes that characterised the growth dynamics in Zostera spp. population located in Arcachon Bay, as well as the seasonal variation on their reproduction. Particular attention was paid to the light variable as it is a crucial driver of growth and physiology for seagrasses. Our guidelines provide a way to adapt a general DBN to specific ecosystems to maximise model reuse and minimise re-development effort. Especially important from a transferability perspective are guidelines for ecosystems with limited data, and how simulation and prior predictive approaches can be used in these contexts.

A Multimodal Approach for the Risk Prediction of Intensive Care and Mortality in Patients with COVID-19

Background: Although several studies have been launched towards the prediction of risk factors for mortality and admission in the intensive care unit (ICU) in COVID-19, none of them focuses on the development of explainable AI models to define an ICU scoring index using dynamically associated biological markers. Methods: We propose a multimodal approach which combines explainable AI models with dynamic modeling methods to shed light into the clinical features of COVID-19. Dynamic Bayesian networks were used to seek associations among cytokines across four time intervals after hospitalization. Explainable gradient boosting trees were trained to predict the risk for ICU admission and mortality towards the development of an ICU scoring index. Results: Our results highlight LDH, IL-6, IL-8, Cr, number of monocytes, lymphocyte count, TNF as risk predictors for ICU admission and survival along with LDH, age, CRP, Cr, WBC, lymphocyte count for mortality in the ICU, with prediction accuracy 0.79 and 0.81, respectively. These risk factors were combined with dynamically associated biological markers to develop an ICU scoring index with accuracy 0.9. Conclusions: to our knowledge, this is the first multimodal and explainable AI model which quantifies the risk of intensive care with accuracy up to 0.9 across multiple timepoints.

Discovering gene regulatory networks of multiple phenotypic groups using dynamic Bayesian networks

Dynamic Bayesian networks (DBNs) can be used for the discovery of gene regulatory networks from time series gene expression data. Here, we suggest a strategy for learning DBNs from gene expression data by employing a Bayesian approach that is scalable to large networks and is targeted at learning models with high predictive accuracy. Our framework can be used to learn DBNs for multiple groups of samples and highlight differences and similarities in their gene regulatory networks. We learn these DBN models based on different structural and parametric assumptions and select the optimal model based on the cross-validated predictive accuracy. We show in simulation studies that our approach is better equipped to prevent overfitting than techniques used in previous studies. We applied the proposed DBN-based classification approach to two time series transcriptomic datasets from the Gene Expression Omnibus database, each comprising data from distinct phenotypic groups of the same tissue type. In the first case, we used DBNs to characterize responders and non-responders to anti-cancer therapy. In the second case, we compared normal to tumor cells of colorectal tissue. The classification accuracy reached by the DBN-based classifier for both datasets was higher than reported previously. For the colorectal cancer dataset, our analysis suggested that GRNs for cancer and normal tissues have a lot of differences, which are most pronounced in the neighborhoods of oncogenes and known cancer tissue markers. The identified differences in gene networks of cancer and normal cells may be used for the discovery of targeted therapies.

Seepage behavior assessment of earth-rock dams based on Bayesian network

Seepage behavior assessment is an important part of the safety operation assessment of earth-rock dams, because of insufficient intelligent analysis of monitoring information, abnormal phenomena or measured values are often ignored or improperly processed. To improve the intelligent performance of the monitoring system, this article has established an assessment framework covering project quality, maintenance status, monitoring data analysis, and on-site inspection based on the relevant norms of seepage safety assessment of earth-rock dams and the expert survey scoring method, and the Leaky Noisy-OR Gate extended model were used to determine the probability of events, and the dynamic and static Bayesian networks used to assess the possibility of seepage failure of earth-rock dams and diagnose the most likely cause of failure. The function of static and dynamic Bayesian networks to assess the seepage behavior of earth-rock dams, abnormal measured values, and causes of anomalies can make up for the limitations of reservoir management personnel and monitoring system in seepage failure experience and seepage knowledge of earth-rock dams and enable better handling of abnormal phenomena and monitoring information, making the monitoring system more intelligent.

Propagation of the Malware Used in APTs Based on Dynamic Bayesian Networks

Malware is becoming more and more sophisticated these days. Currently, the aim of some special specimens of malware is not to infect the largest number of devices as possible, but to reach a set of concrete devices (target devices). This type of malware is usually employed in association with advanced persistent threat (APT) campaigns. Although the great majority of scientific studies are devoted to the design of efficient algorithms to detect this kind of threat, the knowledge about its propagation is also interesting. In this article, a new stochastic computational model to simulate its propagation is proposed based on Bayesian networks. This model considers two characteristics of the devices: having efficient countermeasures, and the number of infectious devices in the neighborhood. Moreover, it takes into account four states: susceptible devices, damaged devices, infectious devices and recovered devices. In this way, the dynamic of the model is SIDR (susceptible–infectious–damaged–recovered). Contrary to what happens with global models, the proposed model takes into account both the individual characteristics of devices and the contact topology. Furthermore, the dynamics is governed by means of a (practically) unexplored technique in this field: Bayesian networks.

Assessment of Dynamic Bayesian Models for Gas Turbine Diagnostics, Part 2: Discrimination of Gradual Degradation and Rapid Faults

There are many challenges that an effective diagnostic system must overcome for successful fault diagnosis in gas turbines. Among others, it has to be robust to engine-to-engine variations in the fleet, it has to discriminate between gradual deterioration and abrupt faults, and it has to identify sensor faults correctly and be robust in case of such faults. To combine their benefits and overcome their limitations, two diagnostic methods were integrated in this work to form a multi-layer system. An adaptive performance model was used to track gradual deterioration and detect rapid or abrupt anomalies, while a series of static and dynamic Bayesian networks were integrated to identify component degradation, component abrupt faults, and sensor faults. The proposed approach was tested on synthetic data and field data from a single-shaft gas turbine of 50 MW class. The results showed that the approach could give acceptable accuracy in the isolation and identification of multiple faults, with 99% detection and isolation accuracy and 1% maximum error in the identified fault magnitude. The approach was also proven robust to sensor faults, by replacing the faulty signal with an estimated value that had only 3% error compared to the real measurement.

Assessment of Dynamic Bayesian Models for Gas Turbine Diagnostics, Part 1: Prior Probability Analysis

The reliability and cost-effectiveness of energy conversion in gas turbine systems are strongly dependent on an accurate diagnosis of possible process and sensor anomalies. Because data collected from a gas turbine system for diagnosis are inherently uncertain due to measurement noise and errors, probabilistic methods offer a promising tool for this problem. In particular, dynamic Bayesian networks present numerous advantages. In this work, two Bayesian networks were developed for compressor fouling and turbine erosion diagnostics. Different prior probability distributions were compared to determine the benefits of a dynamic, first-order hierarchical Markov model over a static prior probability and one dependent only on time. The influence of data uncertainty and scatter was analyzed by testing the diagnostics models on simulated fleet data. It was shown that the condition-based hierarchical model resulted in the best accuracy, and the benefit was more significant for data with higher overlap between states (i.e., for compressor fouling). The improvement with the proposed dynamic Bayesian network was 8 percentage points (in classification accuracy) for compressor fouling and 5 points for turbine erosion compared with the static network.

New Approach for Analyzing Marine Ecosystem Structure Using Bayesian Networks

Aquatic ecosystems of the Black Sea are complex multiparametric systems with a hierarchical structure. Thus, the main goal of our research was to investigate possibilities of using Bayesian networks to study the structure fo the natural systems in the Black Sea. We used CMEMS Black Sea environmental dataset, which consists of 7 different variables, that, in our opinion, can describe structural relations in the Black Sea ecosystem -- sea surface temperature and salinity, concentrations of nitrates and phosphates, amount of chlorophyll-a and net primary production and also dissolved oxygen concentration. We think, that these variables can generally define interactions in water environment of the Black Sea, organisms, that live there and human activity. As a modelling result, we receive a structure of environmental variables interactions. At the top of this structure is a dissolved oxygen, as a final result of the ecosystem functioning. Further, we think it's more appropriate to use Dynamic Bayesian networks for investigation of spatio-temporal changes to distinguish main drivers of changes and provide more balanced management of natural territories.