Complex Decision Trees Research Articles

Personalising dietary advice for disease prevention: concepts and experiences.

Personalised nutrition (PN) as a new endeavour emerged in the background of the human genome project with the ease to analyse genetic heterogeneity. First commercial offers with recommendations for diet and lifestyle changes, usually based on a few polymorphisms, entered markets soon after the presentation of the human genome blueprint. Although PN has seen many attempts, meanwhile, with the inclusion of other biomedical measures such as microbiome and/or continuous glucose monitoring, scientific assessments of such approaches in various settings revealed limited success. Although personalisation improved general compliance over generic advice, particular benefits in referring to biomedical measures and individual risks did, in most cases, not provide any significant advantage. Moreover, scholars criticised such approaches as of limited impact from a public health perspective by attracting mainly technology-open individuals of high social status and proper financial capabilities. Based on these experiences, new avenues for personalising dietary advice are developed, and those are going beyond pure biomedical data by assessing the entire food environment of the individual with its capabilities and constraints in the given life setting. Embedded into digital environments for data collection but also for bidirectional communication, new possibilities emerge. Artificial intelligence methods allow for the multitude of input data and highly complex decision trees to be derived to customize advice. And that can be delivered on the spot and in time in any language whenever decisions are made on what to buy or what to eat. But systems can also be employed to increase physical activity levels and for the adoption of a more healthy lifestyle in general.

Artificial neural network models: implementation of functional near-infrared spectroscopy-based spontaneous lie detection in an interactive scenario.

Deception is an inevitable occurrence in daily life. Various methods have been used to understand the mechanisms underlying brain deception. Moreover, numerous efforts have been undertaken to detect deception and truth-telling. Functional near-infrared spectroscopy (fNIRS) has great potential for neurological applications compared with other state-of-the-art methods. Therefore, an fNIRS-based spontaneous lie detection model was used in the present study. We interviewed 10 healthy subjects to identify deception using the fNIRS system. A card game frequently referred to as a bluff or cheat was introduced. This game was selected because its rules are ideal for testing our hypotheses. The optical probe of the fNIRS was placed on the subject's forehead, and we acquired optical density signals, which were then converted into oxy-hemoglobin and deoxy-hemoglobin signals using the Modified Beer-Lambert law. The oxy-hemoglobin signal was preprocessed to eliminate noise. In this study, we proposed three artificial neural networks inspired by deep learning models, including AlexNet, ResNet, and GoogleNet, to classify deception and truth-telling. The proposed models achieved accuracies of 88.5%, 88.0%, and 90.0%, respectively. These proposed models were compared with other classification models, including k-nearest neighbor, linear support vector machines (SVM), quadratic SVM, cubic SVM, simple decision trees, and complex decision trees. These comparisons showed that the proposed models performed better than the other state-of-the-art methods.

Semi-automation of keratopathy visual acuity grading of corneal events in belantamab mafodotin clinical trials: clinical decision support software.

Belantamab mafodotin (belamaf) has demonstrated clinically meaningful antimyeloma activity in patients with heavily pretreated multiple myeloma. However, it is highly active against dividing cells, contributing to off-target adverse events, particularly ocular toxicity. Changes in best corrected visual acuity (BCVA) and corneal examination findings are routinely monitored to determine Keratopathy Visual Acuity (KVA) grade to inform belamaf dose modification. We aimed to develop a semiautomated mobile app to facilitate the grading of ocular events in clinical trials involving belamaf. The paper process was semiautomated by creating a library of finite-state automaton (FSA) models to represent all permutations of KVA grade changes from baseline BCVA readings. The transition states in the FSA models operated independently of eye measurement units (e.g., Snellen, logMAR, decimal) and provided a uniform approach to determining KVA grade changes. Together with the FSA, the complex decision tree for determining the grade change based on corneal examination findings was converted into logical statements for accurate and efficient overall KVA grade computation. First, a web-based user interface, conforming to clinical practice settings, was developed to simplify the input of key KVA grading criteria. Subsequently, a mobile app was developed that included additional guided steps to assist in clinical decision-making. The app underwent a robust Good Clinical Practice validation process. Outcomes were reviewed by key stakeholders, our belamaf medical lead, and the systems integration team. The time to compute a patient's overall KVA grade using the Belamaf Eye Exam (BEE) app was reduced from a 20- to 30-min process to <1-2 min. The BEE app was well received, with most investigators surveyed selecting "satisfied" or "highly satisfied" for its accuracy and time efficiency. Our semiautomated approach provides for an accurate, simplified method of assessment of patients' corneal status that reduces errors and quickly delivers information critical for potential belamaf dose modifications. The app is currently available on the Apple iOS and Android platforms for use by investigators of the DREAMM clinical trials, and its use could easily be extended to the clinic to support healthcare providers who need to make informed belamaf treatment decisions.

Therapeutic management of diseases based on fuzzy logic system- hypertriglyceridemia as a case study

The support systems for assisting clinical decision highly improve the quality and efficiency of the therapeutic and diagnostic treatment in medicine. The proper implementation of such systems can emulate the reasoning of health care professionals in such a way that suggest reasonable decisions on patient treatment. The fuzzy logic system can be considered as one of the efficient techniques for converting a complex decision tree that usually facing the physician into artificial intelligent procedure embedded in a computer program. So many properties in fuzzy logic system that can facilitate the process of medical diagnosis and therapeutic management. In this paper, a system for therapeutic management of hypertriglyceridemia was efficiently realized using a fuzzy logic technique. The obtained results had shown that the proposed fuzzy logic contributes a reliable managing procedure for assisting the physicians and pharmacist in treating the hypertriglyceridemia. Many different hypertriglyceridemia treatment cases showed a perfect matching decision between the standard guidelines and that given by the proposed system.

Can a Single Variable Predict Early Dropout From Digital Health Interventions? Comparison of Predictive Models From Two Large Randomized Trials.

A single generalizable metric that accurately predicts early dropout from digital health interventions has the potential to readily inform intervention targets and treatment augmentations that could boost retention and intervention outcomes. We recently identified a type of early dropout from digital health interventions for smoking cessation, specifically, users who logged in during the first week of the intervention and had little to no activity thereafter. These users also had a substantially lower smoking cessation rate with our iCanQuit smoking cessation app compared with users who used the app for longer periods. This study aimed to explore whether log-in count data, using standard statistical methods, can precisely predict whether an individual will become an iCanQuit early dropout while validating the approach using other statistical methods and randomized trial data from 3 other digital interventions for smoking cessation (combined randomized N=4529). Standard logistic regression models were used to predict early dropouts for individuals receiving the iCanQuit smoking cessation intervention app, the National Cancer Institute QuitGuide smoking cessation intervention app, the WebQuit.org smoking cessation intervention website, and the Smokefree.gov smoking cessation intervention website. The main predictors were the number of times a participant logged in per day during the first 7 days following randomization. The area under the curve (AUC) assessed the performance of the logistic regression models, which were compared with decision trees, support vector machine, and neural network models. We also examined whether 13 baseline variables that included a variety of demographics (eg, race and ethnicity, gender, and age) and smoking characteristics (eg, use of e-cigarettes and confidence in being smoke free) might improve this prediction. The AUC for each logistic regression model using only the first 7 days of log-in count variables was 0.94 (95% CI 0.90-0.97) for iCanQuit, 0.88 (95% CI 0.83-0.93) for QuitGuide, 0.85 (95% CI 0.80-0.88) for WebQuit.org, and 0.60 (95% CI 0.54-0.66) for Smokefree.gov. Replacing logistic regression models with more complex decision trees, support vector machines, or neural network models did not significantly increase the AUC, nor did including additional baseline variables as predictors. The sensitivity and specificity were generally good, and they were excellent for iCanQuit (ie, 0.91 and 0.85, respectively, at the 0.5 classification threshold). Logistic regression models using only the first 7 days of log-in count data were generally good at predicting early dropouts. These models performed well when using simple, automated, and readily available log-in count data, whereas including self-reported baseline variables did not improve the prediction. The results will inform the early identification of people at risk of early dropout from digital health interventions with the goal of intervening further by providing them with augmented treatments to increase their retention and, ultimately, their intervention outcomes.

Teach me to play, gamer! Imitative learning in computer games via linguistic description of complex phenomena and decision trees

In this article, we present a new machine learning model by imitation based on the linguistic description of complex phenomena. The idea consists of, first, capturing the behaviour of human players by creating a computational perception network based on the execution traces of the games and, second, representing it using fuzzy logic (linguistic variables and if-then rules). From this knowledge, a set of data (dataset) is automatically created to generate a learning model based on decision trees. This model will be used later to automatically control the movements of a bot. The result is an artificial agent that mimics the human player. We have implemented, tested and evaluated this technology from two different points of view: performance by using classical metrics (accuracy, ROC area and PRC area) and believability by using a Turing test for trained bots. The results obtained are interesting and promising, showing that this method can be a good alternative to design and implement the behaviour of intelligent agents in video game development.

Integrated structure-based protein interface prediction

BackgroundIdentifying protein interfaces can inform how proteins interact with their binding partners, uncover the regulatory mechanisms that control biological functions and guide the development of novel therapeutic agents. A variety of computational approaches have been developed for predicting a protein’s interfacial residues from its known sequence and structure. Methods using the known three-dimensional structures of proteins can be template-based or template-free. Template-based methods have limited success in predicting interfaces when homologues with known complex structures are not available to use as templates. The prediction performance of template-free methods that only rely only upon proteins’ intrinsic properties is limited by the amount of biologically relevant features that can be included in an interface prediction model.ResultsWe describe the development of an integrated method for protein interface prediction (ISPIP) to explore the hypothesis that the efficacy of a computational prediction method of protein binding sites can be enhanced by using a combination of methods that rely on orthogonal structure-based properties of a query protein, combining and balancing both template-free and template-based features. ISPIP is a method that integrates these approaches through simple linear or logistic regression models and more complex decision tree models. On a diverse test set of 156 query proteins, ISPIP outperforms each of its individual classifiers in identifying protein binding interfaces.ConclusionsThe integrated method captures the best performance of individual classifiers and delivers an improved interface prediction. The method is robust and performs well even when one of the individual classifiers performs poorly on a particular query protein. This work demonstrates that integrating orthogonal methods that depend on different structural properties of proteins performs better at interface prediction than any individual classifier alone.

Augmented Reality for Guideline Presentation in Medicine: Randomized Crossover Simulation Trial for Technically Assisted Decision-making.

BackgroundGuidelines provide instructions for diagnostics and therapy in modern medicine. Various mobile devices are used to represent the potential complex decision trees. An example of time-critical decisions is triage in case of a mass casualty incident.ObjectiveIn this randomized controlled crossover study, the potential of augmented reality for guideline presentation was evaluated and compared with the guideline presentation provided in a tablet PC as a conventional device.MethodsA specific Android app was designed for use with smart glasses and a tablet PC for the presentation of a triage algorithm as an example for a complex guideline. Forty volunteers simulated a triage based on 30 fictional patient descriptions, each with technical support from smart glasses and a tablet PC in a crossover trial design. The time to come to a decision and the accuracy were recorded and compared between both devices.ResultsA total of 2400 assessments were performed by the 40 volunteers. A significantly faster time to triage was achieved in total with the tablet PC (median 12.8 seconds, IQR 9.4-17.7; 95% CI 14.1-14.9) compared to that to triage with smart glasses (median 17.5 seconds, IQR 13.2-22.8, 95% CI 18.4-19.2; P=.001). Considering the difference in the triage time between both devices, the additional time needed with the smart glasses could be reduced significantly in the course of assessments (21.5 seconds, IQR 16.5-27.3, 95% CI 21.6-23.2) in the first run, 17.4 seconds (IQR 13-22.4, 95% CI 17.6-18.9) in the second run, and 14.9 seconds (IQR 11.7-18.6, 95% CI 15.2-16.3) in the third run (P=.001). With regard to the accuracy of the guideline decisions, there was no significant difference between both the devices.ConclusionsThe presentation of a guideline on a tablet PC as well as through augmented reality achieved good results. The implementation with smart glasses took more time owing to their more complex operating concept but could be accelerated in the course of the study after adaptation. Especially in a non–time-critical working area where hands-free interfaces are useful, a guideline presentation with augmented reality can be of great use during clinical management.

Point-of-care detection and differentiation of anticoagulant therapy - development of thromboelastometry-guided decision-making support algorithms

BackgroundDOAC detection is challenging in emergency situations. Here, we demonstrated recently, that modified thromboelastometric tests can reliably detect and differentiate dabigatran and rivaroxaban. However, whether all DOACs can be detected and differentiated to other coagulopathies is unclear. Therefore, we now tested the hypothesis that a decision tree-based thromboelastometry algorithm enables detection and differentiation of all direct Xa-inhibitors (DXaIs), the direct thrombin inhibitor (DTI) dabigatran, as well as vitamin K antagonists (VKA) and dilutional coagulopathy (DIL) with high accuracy.MethodsFollowing ethics committee approval (No 17–525-4), and registration by the German clinical trials database we conducted a prospective observational trial including 50 anticoagulated patients (n = 10 of either DOAC/VKA) and 20 healthy volunteers. Blood was drawn independent of last intake of coagulation inhibitor. Healthy volunteers served as controls and their blood was diluted to simulate a 50% dilution in vitro. Standard (extrinsic coagulation assay, fibrinogen assay, etc.) and modified thromboelastometric tests (ecarin assay and extrinsic coagulation assay with low tissue factor) were performed. Statistical analyzes included a decision tree analyzes, with depiction of accuracy, sensitivity and specificity, as well as receiver-operating-characteristics (ROC) curve analysis including optimal cut-off values (Youden-Index).ResultsFirst, standard thromboelastometric tests allow a good differentiation between DOACs and VKA, DIL and controls, however they fail to differentiate DXaIs, DTIs and VKAs reliably resulting in an overall accuracy of 78%. Second, adding modified thromboelastometric tests, 9/10 DTI and 28/30 DXaI patients were detected, resulting in an overall accuracy of 94%. Complex decision trees even increased overall accuracy to 98%. ROC curve analyses confirm the decision-tree-based results showing high sensitivity and specificity for detection and differentiation of DTI, DXaIs, VKA, DIL, and controls.ConclusionsDecision tree-based machine-learning algorithms using standard and modified thromboelastometric tests allow reliable detection of DTI and DXaIs, and differentiation to VKA, DIL and controls.Trial registrationClinical trial number: German clinical trials database ID: DRKS00015704.

Health-Related Symptoms and Working Conditions on Vulnerability to Presenteeism Among Nurses in South Korea.

Presenteeism among nurses is reported to be higher than that of other professional workers, and affects one's health and the safety of their patients. Therefore, study on correlation between complex working condition and presenteeism is needed among nurses. This study aimed to predict characteristics of high-risk groups for presenteeism among nurses. The analysis used data which generated 478 nurses who participated in the fifth Korean Working Conditions Survey (2017). This study built a complex samples logistic regression model and decision tree analysis. Presenteeism was significantly higher among those who experienced musculoskeletal pain, high emotional demands, discrimination, and psychological adverse social behavior at the workplace. Combined presence of psychological adverse social behavior and musculoskeletal pain was predicted presenteeism, and should be treated as groups with a high risk of presenteeism. Nurses should be aware of high-risk group for presenteeism and treat them as a priority group to manage.

Diagnosis and management of von Willebrand disease: A community-wide effort to deliver evidence-based clinical practice guidelines.

Diagnosis and management of von Willebrand disease: A community-wide effort to deliver evidence-based clinical practice guidelines.

Infrared Thermography-Based Fault Diagnosis of Induction Motor Bearings Using Machine Learning

Bearing is one of the most crucial parts in induction motor (IM) as a result there is a constant call for effective diagnosis of bearing faults for reliable operation. Infrared thermography (IRT) is appreciably used as a non-destructive and non-contact method to detect the bearing defects in a rotary machine. However, its performance is limited by insignificant information and string noise present in the infrared thermal image. To address this issue, an emergent two dimensional discrete wavelet transform (2D-DWT) based IRT method has been proposed in this article for diagnosing the different bearing faults in IM, namely, inner and outer race defects, and lack of lubrication. The dimensionality of the extracted features was reduced using principal component analysis (PCA) and thereafter the selected features were ranked in the order of most relevant features using the Mahalanobis distance (MD) method to achieve the optimal feature set. Finally these selected features have been passed to the complex decision tree (CDT), linear discriminant analysis (LDA) and support vector machine (SVM) for fault classification and performance evaluation. The classification results reveal that the SVM outperformed CDT and LDA. The proposed strategy can be used for self-adaptive recognition of bearing faults in IM which helps to avoid the unplanned and unwanted system shutdowns due to the bearing failure.

Retrospective analysis of fractionated intensity-modulated radiotherapy (IMRT) in the interdisciplinary management of primary optic nerve sheath meningiomas

BackgroundAs optic nerve sheath meningiomas (ONSM) are rare, there are no prospective studies. Our retrospective analysis focusses on a cohort of patients with uniform disease characteristics all treated with the same radiotherapy regimen. We describe treatment decision making, radiotherapy planning and detailed neuro-ophthalmological outcome of the patients.Methods26 patients with unilateral ONSM extending only to the orbit and the optic canal were evaluated for neuro-ophthalmological outcome. Radiation treatment was planned in a simultaneous integrated boost approach to gross tumor volume (GTV) + 2 mm / 5 mm to 54 Gy / 51 Gy in 1.8 Gy / 1.7 Gy fractions. Follow-up was done by specialized neuro-ophthalmologists. Visual acuity and visual field defects were evaluated after therapy as well as during follow-up.ResultsInterdisciplinary treatment decision for patients with ONSM follows a rather complex decision tree. Radiation treatment planning (equivalent uniform dose (EUD), maximum dose to the optic nerve) improved with experience over time. With this patient selection visual acuity as well as visual field improved significantly at first follow-up after treatment. For visual acuity this also applied to patients with severe defects before treatment. Long term evaluation showed 16 patients with improved visual function, 6 were stable, in 4 patients visual function declined. Interdisciplinary case discussion rated the visual decline as radiation-associated in two patients.ConclusionsWith stringent patient selection radiotherapy for unilateral primary ONSM to 51 Gy / 54 Gy is safe and leads to significantly improved visual function. Interdisciplinary treatment decision and experience of the radiation oncology team play a major role.

Clinical Process Modeling: An Approach for Enhancing the Assessment of Physicians' Clinical Reasoning.

To ensure the validity of test interpretation and use, educators must be able to connect the dots between the tasks examinees perform on a test and the decisions educators make based on the results of that test. Toward that end, in this article, the authors introduce an approach called clinical process modeling, which combines clinical reasoning and principled assessment design. The goal of this approach is to better align test item development with the assessment of physicians' application of knowledge. Clinical process modeling involves creating complex decision trees that can mirror clinical reasoning illness scripts and include the steps and pathways a physician could take to address a specific patient presentation from initial presentation to correct diagnosis and therapy. Once created, these decision trees can be used to inform the assessment development process.To illustrate this approach, the authors describe how they developed one such clinical process model for a common presentation of low back pain from the field of internal medicine. They explain the steps they took to develop their model and the corresponding test item. They conclude by discussing potential future directions and additional implications for this work, emphasizing how clinical process models can be used to inform other educational processes and clinical practice.

Implementing and Improving Pancreas SBRT: One Challenge at a Time

The role of stereotactic body radiotherapy (SBRT) is being investigated as an alternative to long-course chemoradiation in select patients with locally advanced pancreatic cancer. Implementation of this technique is complex and fraught with challenges including the proximity of dose sensitive organs, target visibility for image guidance and respiratory motion. Pancreas SBRT was introduced at our institution in 2015 and the technique has undergone multiple iterations since with the goal of improving outcomes. The aim of our current work is to establish optimal respiratory motion management for pancreas SBRT. Initially, all pancreas SBRT cases were treated with a 3 fraction regimen (21-36 Gy) using abdominal compression to minimize respiratory motion. Patients were required to have a biliary stent in situ or fiducial markers implanted prior to simulation for image guidance. Based on our initial local control and toxicity results, institutional policy changed to a 5 fraction regimen (30-50 Gy). Abdominal compression is beneficial for minimizing motion and immobilizing the patient; however, there is a concern that it also causes anatomic deformation to organs at risk (OARs) such as the stomach and duodenum, moving them closer to the target. Thus, an upfront motion assessment was introduced to determine if motion management was necessary. At the time of planning, a free breathing low dose 4D computed tomography (4DCT) scan is acquired to quantify the pancreatic motion using the fiducials. If motion amplitude is <5mm, planning proceeds without abdominal compression, potentially eliminating the problem of moving the OARs into the high dose region. If on free breathing, the motion amplitude is >5mm, either abdominal compression or Active Breathing Coordinator (ABC) is required. For patients where ABC treatment is an option, breath hold and stability assessments will follow to confirm they are suitable candidates. Performing motion assessment prior to CT simulation allows for a more personalized approach to treating patients with pancreas SBRT. Treating patients without the abdominal plate provides greater comfort for patients and has potential dosimetric benefits for OARs. As access to MRI for simulation/planning increases, the free breathing option can readily be built upon as it does not rely on any treatment accessories. Unfortunately, motion evaluation is currently limited to patients with fiducials because stents are not reliable surrogates for motion as they are quite mobile themselves. Interpreting a complex decision tree like this can be confusing for those performing the simulation, making communication and education integral to the success of the program. To date, 18 patients have undergone motion assessment. Five patients were treated free-breathing, 12 patients with abdominal compression, and 1 with ABC. The 12 patients simulated with abdominal compression had an average motion reduction of 9.2 mm to 5.2 mm, confirming the effectiveness of abdominal compression. To quantify intrafraction motion for the free-breathing option, post treatment Cone Beam Computed Tomography (CBCT) images were acquired after every fraction. Upon analysis of the CBCT data, an average overall shift of only 1.5 mm was seen, suggesting there are no issues with stability. The development of 3 different methods for treating SBRT pancreas has resulted in an approach tailored to individual patient’s motion management needs and prepared us for future technological innovations.

Using machine learning models to predict oxygen saturation following ventilator support adjustment in critically ill children: A single center pilot study.

BackgroundIn an intensive care units, experts in mechanical ventilation are not continuously at patient’s bedside to adjust ventilation settings and to analyze the impact of these adjustments on gas exchange. The development of clinical decision support systems analyzing patients’ data in real time offers an opportunity to fill this gap.ObjectiveThe objective of this study was to determine whether a machine learning predictive model could be trained on a set of clinical data and used to predict transcutaneous hemoglobin oxygen saturation 5 min (5min SpO2) after a ventilator setting change.Data sourcesData of mechanically ventilated children admitted between May 2015 and April 2017 were included and extracted from a high-resolution research database. More than 776,727 data rows were obtained from 610 patients, discretized into 3 class labels (< 84%, 85% to 91% and c92% to 100%).Performance metrics of predictive modelsDue to data imbalance, four different data balancing processes were applied. Then, two machine learning models (artificial neural network and Bootstrap aggregation of complex decision trees) were trained and tested on these four different balanced datasets. The best model predicted SpO2 with area under the curves < 0.75.ConclusionThis single center pilot study using machine learning predictive model resulted in an algorithm with poor accuracy. The comparison of machine learning models showed that bagged complex trees was a promising approach. However, there is a need to improve these models before incorporating them into a clinical decision support systems. One potentially solution for improving predictive model, would be to increase the amount of data available to limit over-fitting that is potentially one of the cause for poor classification performances for 2 of the three class labels.

Simple Inclusion of Complex Diagnostic Algorithms in Infectious Disease Models for Economic Evaluation.

Cost-effectiveness models for infectious disease interventions often require transmission models that capture the indirect benefits from averted subsequent infections. Compartmental models based on ordinary differential equations are commonly used in this context. Decision trees are frequently used in cost-effectiveness modeling and are well suited to describing diagnostic algorithms. However, complex decision trees are laborious to specify as compartmental models and cumbersome to adapt, limiting the detail of algorithms typically included in transmission models. We consider an approximation replacing a decision tree with a single holding state for systems where the time scale of the diagnostic algorithm is shorter than time scales associated with disease progression or transmission. We describe recursive algorithms for calculating the outcomes and mean costs and delays associated with decision trees, as well as design strategies for computational implementation. We assess the performance of the approximation in a simple model of transmission/diagnosis and its role in simplifying a model of tuberculosis diagnostics. When diagnostic delays were short relative to recovery rates, our approximation provided a good account of infection dynamics and the cumulative costs of diagnosis and treatment. Proportional errors were below 5% so long as the longest delay in our 2-step algorithm was under 20% of the recovery time scale. Specifying new diagnostic algorithms in our tuberculosis model was reduced from several tens to just a few lines of code. For conditions characterized by a diagnostic process that is neither instantaneous nor protracted (relative to transmission dynamics), this novel approach retains the advantages of decision trees while embedding them in more complex models of disease transmission. Concise specification and code reuse increase transparency and reduce potential for error.

Classification of Many Abnormal Events in Radial Distribution Feeders Using the Complex Morlet Wavelet and Decision Trees

Monitoring of abnormal events in a distribution feeder by using a single technique is a challenging task. A number of abnormal events can cause unsafe operation, including a high impedance fault (HIF), a partial breakdown to a cable insulation, and a circuit breaker (CB) malfunction due to capacitor bank de-energization. These abnormal events are not detectable by conventional protection schemes. In this paper, a new technique to identify distribution feeder events is proposed based on the complex Morlet wavelet (CMW) and on a decision tree (DT) classifier. First, the event is detected using CMW. Subsequently, a DT using event signatures classifies the event as normal operation, continuous and non-continuous arcing events (C.A.E. and N.C.A.E.). Additional information from the supervisory control and data acquisition (SCADA) can be used to precisely identify the event. The proposed method is meticulously tested on the IEEE 13- and IEEE 34-bus systems and has shown to correctly classify those events. Furthermore, the proposed method is capable of detecting very high impedance incipient faults (IFs) and CB restrikes at the substation level with relatively short detection time. The proposed method uses only current measurements at a low sampling rate of 1440 Hz yielding an improvement of existing methods that require much higher sampling rates.

Repurposing a deep learning network to filter and classify volunteered photographs for land cover and land use characterization

This paper extends recent research into the usefulness of volunteered photos for land cover extraction, and investigates whether this usefulness can be automatically assessed by an easily accessible, off-the-shelf neural network pre-trained on a variety of scene characteristics. Geo-tagged photographs are sometimes presented to volunteers as part of a game which requires them to extract relevant facts about land use. The challenge is to select the most relevant photographs in order to most efficiently extract the useful information while maintaining the engagement and interests of volunteers. By repurposing an existing network which had been trained on an extensive library of potentially relevant features, we can quickly carry out initial assessments of the general value of this approach, pick out especially salient features, and identify focus areas for future neural network training and development. We compare two approaches to extract land cover information from the network: a simple post hoc weighting approach accessible to non-technical audiences and a more complex decision tree approach that involves training on domain-specific features of interest. Both approaches had reasonable success in characterizing human influence within a scene when identifying the land use types (as classified by Urban Atlas) present within a buffer around the photograph’s location. This work identifies important limitations and opportunities for using volunteered photographs as follows: (1) the false precision of a photograph’s location is less useful for identifying on-the-spot land cover than the information it can give on neighbouring combinations of land cover; (2) ground-acquired photographs, interpreted by a neural network, can supplement plan view imagery by identifying features which will never be discernible from above; (3) when dealing with contexts where there are very few exemplars of particular classes, an independent a posteriori weighting of existing scene attributes and categories can buffer against over-specificity.

Can RDA Content, Media, and Carrier Coding Improve Discovery Facet Mapping?

Online public catalogs have provided users with the option to conduct faceted searches for more than a decade. Although faceting is undoubtedly useful to the discovery process, the authors found that their system’s default facet mapping was inadequate for their researchers’ needs, particularly for the faceting of bibliographic formats, and librarians at their institution have worked extensively to revise this mapping. These revisions have relied on creating complex decision trees, which require the system to consult multiple fields and subfields in bibliographic records to assign more precise format facets. When their authority control vendor offered to add Resource Description and Access (RDA) coding to their bibliographic records, including the new Content, Media, and Carrier fields that describe formats with greater granularity than the General Material Designation, they questioned whether the new RDA coding might improve their public catalog’s format faceting. They found that the limitations of the MARC format as a data encoding standard meant that the RDA coding was not appreciably more useful to the format faceting process.