Expert-based Rules Research Articles

Environmental catastrophes, climate change, and attribution.

In our discussion of environmental and ecological catastrophes or disasters resulting from extreme weather events, we unite disparate literatures, the biological and the physical. Our goal is to tie together biological understandings of extreme environmental events with physical understandings of extreme weather events into joint causal accounts. This requires fine‐grained descriptions, in both space and time, of the ecological, evolutionary, and biological moving parts of a system together with fine‐grained descriptions, also in both space and time, of the extreme weather events. We find that both the “storyline” approach to extreme event attribution and the probabilistic “risk‐based” approach have uses in such descriptions. However, the storyline approach is more readily aligned with the forensic approach to evidence that is prevalent in the ecological literature, which cultivates expert‐based rules of thumb, that is, heuristics, and detailed methods for analyzing causes and mechanisms. We introduce below a number of preliminary examples of such studies as instances of what could be pursued in the future in much more detail.

A novel sleep stage scoring system: Combining expert-based features with the generalized linear model.

In this study, we aim to automate the sleep stage scoring process of overnight polysomnography (PSG) data while adhering to expert-based rules. We developed a sleep stage scoring algorithm utilizing the generalized linear modelling (GLM) framework and extracted features from electroencephalogram (EEG), electromyography (EMG) and electrooculogram (EOG) signals based on predefined rules of the American Academy of Sleep Medicine (AASM) Manual for Scoring Sleep. Specifically, features were computed in 30-s epochs in the time and frequency domains of the signals and were then used to model the probability of an epoch being in each of five sleep stages: N3, N2, N1, REM or Wake. Finally, each epoch was assigned to a sleep stage based on model predictions. The algorithm was trained and tested on PSG data from 38 healthy individuals with no reported sleep disturbances. The overall scoring accuracy reached on the test set was 81.50±1.14% (Cohen's kappa, ). The test set results were highly comparable to the training set, indicating robustness of the algorithm. Furthermore, our algorithm was compared to three well-known commercialized sleep-staging tools and achieved higher accuracies than all of them. Our results suggest that automatic classification is highly consistent with visual scoring. We conclude that our algorithm can reproduce the judgement of a scoring expert and is also highly interpretable. This tool can assist visual scorers to speed up their process (from hours to minutes) and provides a method for a more robust, quantitative, reproducible and cost-effective PSG evaluation, supporting assessment of sleep and sleep disorders.

Excasafezone synthesizing expert based ‘on-the-fly’ safety risk heat maps

Excavation work takes place almost continually in most cities around the Western hemisphere. Many cities are already full of infrastructures, buried networks, and street furniture, so excavation work is not without any thread to the operator and surrounding environment. Small construction sites, for example, are often constrained by operating infrastructure on surface level and underground. Although different agencies and network owners have information about the location of the objects that put excavation work at risk, this information is not centralized. Different organizations manage location information of buried cables, unexploded ordnance, and pollution, for example. This significantly complicates the early-stage planning and last minute risk assessment processes because professionals need to manually collect, assess, and integrate data about subsurface objects into a comprehensive risk assessment. To smoothen this process, ExcaSafeZone project, therefore, develops a system that collects location data, defines expert-based rules for safety risk assessment, and that synthesizes this into an open source prototype that visualized safety risks on a heat map.

Machine learning applied to multi-sensor information to reduce false alarm rate in the ICU.

Studies reveal that the false alarm rate (FAR) demonstrated by intensive care unit (ICU) vital signs monitors ranges from 0.72 to 0.99. We applied machine learning (ML) to ICU multi-sensor information to imitate a medical specialist in diagnosing patient condition. We hypothesized that applying this data-driven approach to medical monitors will help reduce the FAR even when data from sensors are missing. An expert-based rules algorithm identified and tagged in our dataset seven clinical alarm scenarios. We compared a random forest (RF) ML model trained using the tagged data, where parameters (e.g., heart rate or blood pressure) were (deliberately) removed, in detecting ICU signals with the full expert-based rules (FER), our ground truth, and partial expert-based rules (PER), missing these parameters. When all alarm scenarios were examined, RF and FER were almost identical. However, in the absence of one to three parameters, RF maintained its values of the Youden index (0.94-0.97) and positive predictive value (PPV) (0.98-0.99), whereas PER lost its value (0.54-0.8 and 0.76-0.88, respectively). While the FAR for PER with missing parameters was 0.17-0.39, it was only 0.01-0.02 for RF. When scenarios were examined separately, RF showed clear superiority in almost all combinations of scenarios and numbers of missing parameters. When sensor data are missing, specialist performance worsens with the number of missing parameters, whereas the RF model attains high accuracy and low FAR due to its ability to fuse information from available sensors, compensating for missing parameters.

A Novel Sleep Stage Scoring System: Combining Expert-Based Rules with a Decision Tree Classifier.

Overnight polysomnography (PSG) is the gold standard tool used to characterize sleep and for diagnosing sleep disorders. PSG is a non-invasive procedure that collects various physiological data which is then scored by sleep specialists who assign a sleep stage to every 30-second window of the data according to predefined scoring rules. In this study, we aimed to automate the process of sleep stage scoring of overnight PSG data while adhering to expert-based rules. We developed an algorithm utilizing a likelihood ratio decision tree classifier and extracted features from EEG, EMG and EOG signals based on predefined rules of the American Academy of Sleep Medicine Manual. Specifically, features were computed in 30-second epochs in the time and the frequency domains of the signals and used as inputs to the classifier which assigned each epoch to one of five possible stages: N3, N2, N1, REM or Wake. The algorithm was trained and tested on PSG data from 38 healthy individuals with no reported sleep disturbances. The overall scoring accuracy was 80.70% on the test set, which was comparable to the training set. Our results imply that the automatic classification is highly robust, fast, consistent with visual scoring and is highly interpretable.

SENSOR- AND SCENE-GUIDED INTEGRATION OF TLS AND PHOTOGRAMMETRIC POINT CLOUDS FOR LANDSLIDE MONITORING

Abstract. Terrestrial and airborne 3D imaging sensors are well-suited data acquisition systems for the area-wide monitoring of landslide activity. State-of-the-art surveying techniques, such as terrestrial laser scanning (TLS) and photogrammetry based on unmanned aerial vehicle (UAV) imagery or terrestrial acquisitions have advantages and limitations associated with their individual measurement principles. In this study we present an integration approach for 3D point clouds derived from these techniques, aiming at improving the topographic representation of landslide features while enabling a more accurate assessment of landslide-induced changes. Four expert-based rules involving local morphometric features computed from eigenvectors, elevation and the agreement of the individual point clouds, are used to choose within voxels of selectable size which sensor’s data to keep. Based on the integrated point clouds, digital surface models and shaded reliefs are computed. Using an image correlation technique, displacement vectors are finally derived from the multi-temporal shaded reliefs. All results show comparable patterns of landslide movement rates and directions. However, depending on the applied integration rule, differences in spatial coverage and correlation strength emerge.

Remote monitoring of vigilance behavior in large herbivores using acceleration data

BackgroundBiotelemetry offers an increasing set of tools to monitor animals. Acceleration sensors in particular can provide remote observations of animal behavior at high temporal resolution. While recent studies have demonstrated the capability of this technique for a wide range of species and behaviors, a coherent methodology is still missing (1) for behavior monitoring of large herbivores that are usually tagged with neck collars and frequently switch between diverse behaviors and (2) for monitoring of vigilance behavior. Here, we present an approach that aims at remotely monitoring different types of large herbivore behavior including vigilance with acceleration data.MethodsWe pioneered this approach with field observations of eight collared roe deer (Capreolus capreolus). First, we trained a classification model for distinguishing seven structural behavior categories: lying, standing, browsing, walking, trotting, galloping and ‘others’. Second, we developed a model that predicted the internal states, active and resting, based on the predicted sequence of structural behaviors and expert-based rules. Further, we applied both models to automatically monitor vigilance behavior and compared model predictions with expert judgment of vigilance behavior. To exemplify the practical application of this approach, we predicted behavior, internal state and vigilance continuously for a collared roe deer.ResultsThe structural behaviors were predicted with high accuracy (overall cross-validated accuracy 71%). Only behaviors that are similar in terms of posture and dynamic body movements were prone to misclassification. Active and resting states showed clear distinction and could be utilized as behavioral context for the detection of vigilance behavior. Here, model predictions were characterized by excellent consistency with expert judgment of vigilance behavior (mean accuracy 96%).ConclusionIn this study, we demonstrated the strong potential and practical applicability of acceleration data for continuous, high-resolution behavior monitoring of large herbivores and showed that vigilance behavior is well detectable. In particular, when combined with spatial data, automated behavior recognition will enrich many fields in behavioral ecology by providing extensive access to behaviors of animals in the wild.

Computational Prediction of HIV-1 Resistance to Protease Inhibitors.

The development of mutations in HIV-1 protease (PR) hinders the activity of antiretroviral drugs, forcing changes in drug prescription. Most resistance assessments used to date rely on expert-based rules on predefined sets of stereotypical mutations; such an information-driven approach cannot capture new polymorphisms or be applied for new drugs. Computational modeling could provide a more general assessment of drug resistance and could be made available to clinicians through the Internet. We have created a protocol involving sequence comparison and all-atom protein-ligand induced fit simulations to predict resistance at the molecular level. We first compared our predictions with the experimentally determined IC50 values of darunavir, amprenavir, ritonavir, and indinavir from reference PR mutants displaying different resistance levels. We then performed analyses on a large set of variants harboring more than 10 mutations. Finally, several sequences from real patients were analyzed for amprenavir and darunavir. Our computational approach detected all of the genotype changes triggering high-level resistance, even those involving a large number of mutations.

A new in silico classification model for ready biodegradability, based on molecular fragments

Regulations such as the European REACH (Registration, Evaluation, Authorization and restriction of Chemicals) often require chemicals to be evaluated for ready biodegradability, to assess the potential risk for environmental and human health. Because not all chemicals can be tested, there is an increasing demand for tools for quick and inexpensive biodegradability screening, such as computer-based (in silico) theoretical models. We developed an in silico model starting from a dataset of 728 chemicals with ready biodegradability data (MITI-test Ministry of International Trade and Industry). We used the novel software SARpy to automatically extract, through a structural fragmentation process, a set of substructures statistically related to ready biodegradability. Then, we analysed these substructures in order to build some general rules. The model consists of a rule-set made up of the combination of the statistically relevant fragments and of the expert-based rules. The model gives good statistical performance with 92%, 82% and 76% accuracy on the training, test and external set respectively. These results are comparable with other in silico models like BIOWIN developed by the United States Environmental Protection Agency (EPA); moreover this new model includes an easily understandable explanation.

Probabilistic grammatical model for helix‐helix contact site classification

BackgroundHidden Markov Models power many state‐of‐the‐art tools in the field of protein bioinformatics. While excelling in their tasks, these methods of protein analysis do not convey directly information on medium‐ and long‐range residue‐residue interactions. This requires an expressive power of at least context‐free grammars. However, application of more powerful grammar formalisms to protein analysis has been surprisingly limited.ResultsIn this work, we present a probabilistic grammatical framework for problem‐specific protein languages and apply it to classification of transmembrane helix‐helix pairs configurations. The core of the model consists of a probabilistic context‐free grammar, automatically inferred by a genetic algorithm from only a generic set of expert‐based rules and positive training samples. The model was applied to produce sequence based descriptors of four classes of transmembrane helix‐helix contact site configurations. The highest performance of the classifiers reached AUCROC of 0.70. The analysis of grammar parse trees revealed the ability of representing structural features of helix‐helix contact sites.ConclusionsWe demonstrated that our probabilistic context‐free framework for analysis of protein sequences outperforms the state of the art in the task of helix‐helix contact site classification. However, this is achieved without necessarily requiring modeling long range dependencies between interacting residues. A significant feature of our approach is that grammar rules and parse trees are human‐readable. Thus they could provide biologically meaningful information for molecular biologists.

Using a cropping system model at regional scale: Low-data approaches for crop management information and model calibration

Cropping system models are powerful tools for regional impact assessment, but their input data requirements for large heterogeneous areas are difficult to fulfil. Hence, the objectives of this paper are to present low-data approaches for specifying detailed management data required by cropping system models, and for calibrating default crop parameters applied to 12 regions in the European Union (EU). Various downscaling and upscaling procedures for different data types are applied to address both objectives. The Agricultural Production and Externalities Simulator (APES) model is used for illustrative purposes. Combining easy-to-collect regional crop management information and expert knowledge enables to develop generic, expert-based rules for specifying crop management. Effects of these expert-based management rules on simulated yields and nitrogen leaching are illustrated using APES. Simulated yields of grain maize, soft wheat and durum wheat using default crop parameters for phenology are compared with crop yields observed in 12 EU regions. The accuracy of the simulated yields was variable, but generally poor. A regional calibration factor Kpheno is developed based on the temperature sum of the average sowing and harvest dates of the three crops in each region. Applying this calibration factor improved the simulated yields in all cases. Results suggest that it is possible to develop expert-based management rules and to capture yield variation across the EU by using the presented low-data approaches.