Decision Support Solutions Research Articles

Airflow Modeling for Citrus under Protective Screens

This study explores the development and validation of an airflow model to support climate prediction for Citrus Under Protective Screens (CUPS) in California. CUPS is a permeable screen structure designed to protect a field of citrus trees from large insects including the vector that causes the devastating citrus greening disease. Because screen structures modify the environmental conditions (e.g., temperature, relative humidity, airflow), farm management and treatment strategies (e.g., pesticide spraying events) must be modified to account for these differences. Toward this end, we develop a model for predicting wind speed and direction in a commercial-scale research CUPS, using a computational fluid dynamics (CFD) model. We describe the model and validate it in two ways. In the first, we model a small-scale replica CUPS under controlled conditions and compare modeled and measured airflow in and around the replica structure. In the second, we model the full-scale CUPS and use historical measurements to “back test” the model’s accuracy. In both settings, the modeled airflow values fall within statistical confidence intervals generated from the corresponding measurements of the conditions being modeled. These findings suggest that the model can aid decision support and smart agriculture solutions for farmers as they adapt their farm management practices for CUPS structures.

Deep Learning to Detect Intracranial Hemorrhage in a National Teleradiology Program and the Impact on Interpretation Time.

The diagnostic performance of an artificial intelligence (AI) clinical decision support solution for acute intracranial hemorrhage (ICH) detection was assessed in a large teleradiology practice. The impact on radiologist read times and system efficiency was also quantified. A total of 61 704 consecutive noncontrast head CT examinations were retrospectively evaluated. System performance was calculated along with mean and median read times for CT studies obtained before (baseline, pre-AI period; August 2021 to May 2022) and after (post-AI period; January 2023 to February 2024) AI implementation. The AI solution had a sensitivity of 75.6%, specificity of 92.1%, accuracy of 91.7%, prevalence of 2.70%, and positive predictive value of 21.1%. Of the 56 745 post-AI CT scans with no bleed identified by a radiologist, examinations falsely flagged as suspected ICH by the AI solution (n = 4464) took an average of 9 minutes 40 seconds (median, 8 minutes 7 seconds) to interpret as compared with 8 minutes 25 seconds (median, 6 minutes 48 seconds) for unremarkable CT scans before AI (n = 49 007) (P < .001) and 8 minutes 38 seconds (median, 6 minutes 53 seconds) after AI when ICH was not suspected by the AI solution (n = 52 281) (P < .001). CT scans with no bleed identified by the AI but reported as positive for ICH by the radiologist (n = 384) took an average of 14 minutes 23 seconds (median, 13 minutes 35 seconds) to interpret as compared with 13 minutes 34 seconds (median, 12 minutes 30 seconds) for CT scans correctly reported as a bleed by the AI (n = 1192) (P = .04). With lengthened read times for falsely flagged examinations, system inefficiencies may outweigh the potential benefits of using the tool in a high volume, low prevalence environment. Keywords: Artificial Intelligence, Intracranial Hemorrhage, Read Time, Report Turnaround Time, System Efficiency Supplemental material is available for this article. © RSNA, 2024.

The Impact of a Skill-Driven Model on Scrum Teams in Software Projects: A Catalyst for Digital Transformation

Human skills are a critical factor in the success or failure of a digital project. Limited studies have been conducted to identify the industry demand for skills of scrum roles (product owner, scrum master, web developer) and levels (entry, associate, mid-senior). The evaluation of skills over time benefits both decision-makers and associated team members, which leads to successful project completions. The aim of this research is to improve decision making concerning the level-specific skills of selected scrum roles for digital projects. The study identifies major and minor skills, patterns, and relationships between levels, and formulates the mathematical equations as the most important inputs to the skill-driven model’s implementation and evaluation. Both qualitative and quantitative research methods were used to analyse 900 surveyed job advertisements published on LinkedIn in Europe. Descriptive analysis was used to analyse quantitative data while the deductive approach was followed with thematic analysis. There are required skill sets for each level of roles, level-specific skills, industry-demanded skills, and formulas related to the initial and individual skill ratings that are investigated. A new mechanism for evaluation is introduced based on “the time spent with skills”. As a result, the proposed model is implemented by feeding research findings into the Mendix programming platform. The skill-driven model is a decision-support solution in software project management to evaluate skills which assist in assigning the right person to the right digital project. Further investigation on different job portals can help to improve the accuracy of industry standards and reduce the lack of progression skills by overcoming limitations identified in this paper.

COVID-Net Biochem: an explainability-driven framework to building machine learning models for predicting survival and kidney injury of COVID-19 patients from clinical and biochemistry data

Since the World Health Organization declared COVID-19 a pandemic in 2020, the global community has faced ongoing challenges in controlling and mitigating the transmission of the SARS-CoV-2 virus, as well as its evolving subvariants and recombinants. A significant challenge during the pandemic has not only been the accurate detection of positive cases but also the efficient prediction of risks associated with complications and patient survival probabilities. These tasks entail considerable clinical resource allocation and attention. In this study, we introduce COVID-Net Biochem, a versatile and explainable framework for constructing machine learning models. We apply this framework to predict COVID-19 patient survival and the likelihood of developing Acute Kidney Injury during hospitalization, utilizing clinical and biochemical data in a transparent, systematic approach. The proposed approach advances machine learning model design by seamlessly integrating domain expertise with explainability tools, enabling model decisions to be based on key biomarkers. This fosters a more transparent and interpretable decision-making process made by machines specifically for medical applications. More specifically, the framework comprises two phases: In the first phase, referred to as the “clinician-guided design” phase, the dataset is preprocessed using explainable AI and domain expert input. To better demonstrate this phase, we prepared a benchmark dataset of carefully curated clinical and biochemical markers based on clinician assessments for survival and kidney injury prediction in COVID-19 patients. This dataset was selected from a patient cohort of 1366 individuals at Stony Brook University. Moreover, we designed and trained a diverse collection of machine learning models, encompassing gradient-based boosting tree architectures and deep transformer architectures, specifically for survival and kidney injury prediction based on the selected markers. In the second phase, called the “explainability-driven design refinement” phase, the proposed framework employs explainability methods to not only gain a deeper understanding of each model’s decision-making process but also to identify the overall impact of individual clinical and biochemical markers for bias identification. In this context, we used the models constructed in the previous phase for the prediction task and analyzed the explainability outcomes alongside a clinician with over 8 years of experience to gain a deeper understanding of the clinical validity of the decisions made. The explainability-driven insights obtained, in conjunction with the associated clinical feedback, are then utilized to guide and refine the training policies and architectural design iteratively. This process aims to enhance not only the prediction performance but also the clinical validity and trustworthiness of the final machine learning models. Employing the proposed explainability-driven framework, we attained 93.55% accuracy in survival prediction and 88.05% accuracy in predicting kidney injury complications. The models have been made available through an open-source platform. Although not a production-ready solution, this study aims to serve as a catalyst for clinical scientists, machine learning researchers, and citizen scientists to develop innovative and trustworthy clinical decision support solutions, ultimately assisting clinicians worldwide in managing pandemic outcomes.

Intelligent Risk-Based Cybersecurity Protection for Industrial Systems Control—A Feasibility Study

Summary Intelligent automated industrial process control, enabled by IR 4.0 technology, requires increased system integration and connectedness. With such development comes a heightened risk of cyberattack for operational technology (OT) systems, including industrial control systems (ICS) and industrial automation and control systems (IACS), historically shielded from cyberspace. Cyberattacks on OT systems can also impact the physical world. The recent attack on the US Colonial Pipeline interrupted delivery of oil along the US West coast for several days, resulting in a declaration of a state of emergency, while the Ukraine power grid attacks interrupted electrical power supply in Kyiv, impacting thousands of consumers. In 2021, an Equinor-operated platform on the Norwegian Continental shelf was hit by malware. The incident had the potential to affect the drilling control system (DCS). For ICS, the concern is ensuring a safe process, as opposed to the more traditional concern with data protection for information technology (IT) system management. To ensure safe implementation of advanced industrial systems control, while avoiding unnecessary operational downtime, there is a need for cybersecurity solutions that account for risk for the whole cyberphysical system (CPS), including the ICS, the work environment, the product, and the physical surroundings. Diagnostic models and methods must therefore be developed that cover the functionality of the whole CPS. Further, to account for process dynamics and evolving cyberattack threats, adaptability is required with respect to both process state and new types of attacks. For this purpose, knowledge-based awareness monitoring combined with existing cyberattack detection tools for ICS, known as ICS intrusion detection systems (IDS), is proposed, providing an element of artificial intelligence. Results from studies in a DCS environment indicate that existing monitoring applications can be used to detect and discern between different types of cyberattacks on CPSs. This indicates feasibility with respect to monitoring the control and IT components of the ICS system for building risk-based cybersecurity decision support solutions. The key challenge and novelty of the proposed approach is to extend the capability of cybermonitoring in systems control to automated evaluation of process risk, including risk to the physical environment. Such capability will enable appropriate decision support with respect to both process risk and operational downtime.

A conceptual digital twin framework for city logistics

Urban logistics is one of the key elements of urban mobility planning. The use of real-time information systems in logistics operations generates an enormous amount of data, nowadays used mainly for the purpose of monitoring and control of large flows of goods. At the same time, urban planners, business stakeholders, and city administrators are in need of adaptive, data-driven decision support solutions to address today's urban logistics problems. Recently, digital twins have received a lot of attention to support advanced experimentation, simulation and decision-making for on-demand logistics operations. Questions still remain on how to realize these for urban logistics management in a mixed public-private stakeholder context. We argue that this lack of a specific framework for city logistics with a model library for data mergers, linking physical and virtual data exchange, can compromise the timely adoption of digital twin technology. We contribute to filling this gap by presenting a systematic review of the literature, proposing a conceptual framework for digital twin applications in urban logistics, and providing use case scenarios for their demonstration. Together, these should advance the technical implementation of digital twins in a sustainable city logistics context.

Formalizing the information requirements for decision-making of field managers during indoor construction activities

PurposeInformation technology provides important support for on-site decision-making of field personnel. Most literature focuses on the technological aspects of decision-support systems, without fully understanding the information required for effective decision-making. This study aimed to formalize decision-makers’ requirements in terms of the major goals, decisions and information.Design/methodology/approachThe situation awareness (SA) approach was applied through the goal-directed task analysis (GDTA) method, narrowing the scope to field managers’ decision-making during indoor construction activities. This method was based on a series of interviews to define, revise and validate the decision-making requirements for the given scope.FindingsThe study yielded 1,056 highly interrelated elements. The results indicate that the field manager’s overall goal is to execute and handoff work within the established deadlines, with the required quality, maximizing profits, within a safe work environment. The overall goal construes into five main goals regarding work progress, quality, costs, safety and communication. These goals include subgoals, decisions, and the information necessary to attain them, depicted in diagrams.Practical implicationsThe findings allow enhancing the design of decision-support solutions by identifying information required for future developments and showing the interrelations between goals and information requirements that need to be addressed to present interfaces for effectively assisting on-site decision-making. Moreover, the results allow for the assessment of solutions regarding the sufficiency of information.Originality/valueThis is the first effort to fully understand the information required by field managers for on-site decision-making during indoor construction activities.

GEO: A Computational Design Framework for Automotive Exterior Facelift

Exterior facelift has become an effective method for automakers to boost the consumers’ interest in an existing car model before it is redesigned. To support the automotive facelift design process, this study develops a novel computational framework – Generator, Evaluator, Optimiser (GEO) , which comprises three components: a StyleGAN2-based design generator that creates different facelift designs; a convolutional neural network (CNN) -based evaluator that assesses designs from the aesthetics perspective; and a recurrent neural network (RNN) -based decision optimiser that selects designs to maximise the predicted profit of the targeted car model over time. We validate the GEO framework in experiments with real-world datasets and describe some resulting managerial implications for automotive facelift. Our study makes both methodological and application contributions. First, the generator’s mapping network and projection methods are carefully tailored to facelift where only minor changes are performed without affecting the family signature of the automobile brands. Second, two evaluation metrics are proposed to assess the generated designs. Third, profit maximisation is taken into account in the design selection. From a high-level perspective, our study contributes to the recent use of machine learning and data mining in marketing and design studies. To the best of our knowledge, this is the first study that uses deep generative models for automotive regional design upgrading and that provides an end-to-end decision-support solution for automakers and designers.

PAnno: A pharmacogenomics annotation tool for clinical genomic testing.

Introduction: Next-generation sequencing (NGS) technologies have been widely used in clinical genomic testing for drug response phenotypes. However, the inherent limitations of short reads make accurate inference of diplotypes still challenging, which may reduce the effectiveness of genotype-guided drug therapy. Methods: An automated Pharmacogenomics Annotation tool (PAnno) was implemented, which reports prescribing recommendations and phenotypes by parsing the germline variant call format (VCF) file from NGS and the population to which the individual belongs. Results: A ranking model dedicated to inferring diplotypes, developed based on the allele (haplotype) definition and population allele frequency, was introduced in PAnno. The predictive performance was validated in comparison with four similar tools using the consensus diplotype data of the Genetic Testing Reference Materials Coordination Program (GeT-RM) as ground truth. An annotation method was proposed to summarize prescribing recommendations and classify drugs into avoid use, use with caution, and routine use, following the recommendations of the Clinical Pharmacogenetics Implementation Consortium (CPIC), etc. It further predicts phenotypes of specific drugs in terms of toxicity, dosage, efficacy, and metabolism by integrating the high-confidence clinical annotations in the Pharmacogenomics Knowledgebase (PharmGKB). PAnno is available at https://github.com/PreMedKB/PAnno. Discussion: PAnno provides an end-to-end clinical pharmacogenomics decision support solution by resolving, annotating, and reporting germline variants.

Towards Optimal Freight Selection Strategies in Transport Planning

The problem of inland freight planning from the perspective of a medium size transport company operating in the EU market is studied. The authors embark on a path to put Information Technologies to work in order to reduce manual overhead and facilitate decision-making within this domain by algorithmically approximating the potential profitability of freight offers based on multiple criteria and shortlisting the most promising offers. Following the Design Science Research approach, the Proof of Concept of IT-enabled decision support solution for logistics companies is delivered and validated. Several strategies for choosing an offer among many for a truck with a given spatio-temporal location are investigated and compared.

Decision Support System for Indonesia Smart Card (KIP) Scholarship Selection Using The AHP And VIKOR Method Integrated with EKTP

. The Indonesia Smart Card (KIP) for college is a government-sponsored educational assistance program that builds upon the foundation of the Bidikmisi scholarship, providing continued support for eligible students. In the 2018/2019 academic year, Sebelas Maret University Surakarta (UNS) received approximately 2,000 applicants, while the recipient quota was limited to less than 1,000. In the current selection process for KIP UNS scholarships, the criteria weighting method is utilized. However, there is a notable absence of consistency checking during this crucial process. Additionally, incorporating consistency checks is imperative to attain an optimal solution and mitigate the risks associated with subjective decision-making. Relying solely on Microsoft Excel software can also restrict data accessibility and availability. Therefore, this study aims to address the issue of inconsistency in KIP scholarship process by proposing Decision Support System that incorporates consistency checking and improves decision outcomes. The AHP and VIKOR method were used in providing recommendations for KIP scholarship recipients. By employing the AHP method, the calculated weighting criteria exhibited consistency, as indicated by a ratio value of 0.0746. The result showed that VIKOR produced a ranking of KIP scholarship candidates, validated by the confusion matrix method, indicating an accuracy and error rate value of 87% and 13% respectively. These findings confirmed that the AHP and VIKOR method serve as an effective decision-support solution for selecting KIP scholarship recipient

Farklı Sınıflandırıcılar ve Yeniden Örnekleme Teknikleri Kullanılarak Kalp Hastalığı Teşhisine Yönelik Karşılaştırmalı Bir Çalışma

Heart diseases are common worldwide and cause one-third of global deaths. The difficulty in distinguishing the symptoms of heart disease and the fact that most heart patients are not aware of the symptoms until the moment of crisis make the diagnosis of the disease difficult. Machine learning, an artificial intelligence discipline, provides experts with successful decision support solutions in diagnosing new cases based on known data. In this study, classifications were made using various machine learning techniques for the early diagnosis of heart diseases. The study was carried out on the UCI heart disease dataset, which is widely used in the literature. In order to increase the classification success, resampling techniques were used to ensure the class balance of the dataset. For each of 8 different machine learning techniques, namely Naive Bayes, Decision Trees, Support Vector Machine, K Nearest Neighbor, Logistic Regression, Random Forest, AdaBoost, and CatBoost, in addition to no-sampling classification, 8 different methods from oversampling and undersampling techniques were used to make a total of 72 classification processes were carried out. The result of each classification process is reported with 5 different parameters: accuracy, precision, recall, F1 score, and AUC. The highest accuracy value was obtained as 98.46% in the classification using Random Forest and InstanceHardnessThreshold undersampling technique. It was observed that the measurements obtained were higher than the results obtained in similar studies conducted in the literature in recent years.

Pharmacogenomics decision support in the U-PGx project: Results and advice from clinical implementation across seven European countries

The clinical implementation of pharmacogenomics (PGx) could be one of the first milestones towards realizing personalized medicine in routine care. However, its widespread adoption requires the availability of suitable clinical decision support (CDS) systems, which is often impeded by the fragmentation or absence of adequate health IT infrastructures. We report results of CDS implementation in the large-scale European research project Ubiquitous Pharmacogenomics (U-PGx), in which PGx CDS was rolled out and evaluated across more than 15 clinical sites in the Netherlands, Spain, Slovenia, Italy, Greece, United Kingdom and Austria, covering a wide variety of healthcare settings. We evaluated the CDS implementation process through qualitative and quantitative process indicators. Quantitative indicators included statistics on generated PGx reports, median time from sampled upload until report delivery and statistics on report retrievals via the mobile-based CDS tool. Adoption of different CDS tools, uptake and usability were further investigated through a user survey among healthcare providers. Results of a risk assessment conducted prior to the implementation process were retrospectively analyzed and compared to actual encountered difficulties and their impact. As of March 2021, personalized PGx reports were produced from 6884 genotyped samples with a median delivery time of twenty minutes. Out of 131 invited healthcare providers, 65 completed the questionnaire (response rate: 49.6%). Overall satisfaction rates with the different CDS tools varied between 63.6% and 85.2% per tool. Delays in implementation were caused by challenges including institutional factors and complexities in the development of required tools and reference data resources, such as genotype-phenotype mappings. We demonstrated the feasibility of implementing a standardized PGx decision support solution in a multinational, multi-language and multi-center setting. Remaining challenges for future wide-scale roll-out include the harmonization of existing PGx information in guidelines and drug labels, the need for strategies to lower the barrier of PGx CDS adoption for healthcare institutions and providers, and easier compliance with regulatory and legal frameworks.

Recent innovations in the LaRiMiT risk mitigation tool: implementing a novel methodology for expert scoring and extending the database to include nature-based solutions

AbstractThis paper presents recent innovations implemented in the LaRiMiT (Landslide Risk Mitigation Toolbox) webtool. These include an innovative methodology for utilising experts’ scoring of landslide risk mitigation measures directly within the decision support tool, and updating of the tool’s database over structural measures to include many nature-based solutions (NBS) for mitigating landslide risk. Landslides are a common and treacherous natural hazard, and due to the effect of climate change on the frequency and intensity of extreme weather events, climate-driven landslides are expected to become more frequent. This negative trend is a driving factor for developing decision-support solutions for risk mitigation, and the LaRiMiT toolbox and the expert scoring methodology presented in this article contribute to meeting these needs. Selecting suitable structural measures is complicated due to factors such as site-specific conditions, local knowledge and resources, socio-economic constraints, and environmental considerations. LaRiMiT uses user-input attributes of site-specific slope movements combined with expert scoring of the suitability of various mitigation solutions to identify and select appropriate mitigation measures from an extensive database of structural solutions. While the LaRiMiT database initially included only conventional (grey) solutions relying on traditional methods, it has been recently expanded to include nature-based solutions (NBS), which are sustainable techniques for managing erosion and mitigating shallow landslides using vegetation and the use of natural materials. NBS and conventional solutions can also be combined to provide hybrid solutions. The proposed methodology for implementing expert scoring is dynamic and iterative, consisting of statistical pooling of experts’ scores collected via online surveys, and consolidating these into an expert scoring utility embedded in the LaRiMiT web portal engine. An initial application of this methodology has been tested by surveying a set of landslides experts, mostly in Europe, and applying it to the structural measures database in the LaRiMiT tool. Although the number of experts in this initial application is limited, dynamic updating of the scoring allows the tool to continuously improve as additional experts contribute.

An interpretable machine learning approach for engineering change management decision support in automotive industry

As an essential part of the Product Life Cycle (PLC), the time-to-market of products is influenced by Engineering Change Management (ECM) processes. An Engineering Change (EC) is part of a formal process in industry to describe, rationalize, determine, release components for final production or make changes to already released design. It includes information about shape, functionality, production location, cost, and other relevant data entries. The duration from creation to the approval of a change request can take weeks or even months, without apparent reasons for the bottleneck. In addition, changes to one component can lead to unexpected chain reactions to other components. Therefore, identifying impacts of changes is challenging for all Original Equipment Manufacturers (OEMs). To address the above challenges, the authors have developed and built a machine learning-based decision support solution in this article. Community detection and stacking algorithms were applied to build more robust models. Impacts and lead time of Engineering Change Requests (ECRs) are predicted and explained by Local Interpretable Model-agnostic Explanations (LIME). A case study was conducted on real-world data from an automotive company. After evaluation with industry experts, the solution approach was proved to have positive contributions to increasing the quality, efficiency, and transparency of the existing ECM processes.

Decision support system for adaptive sourcing and inventory management in small- and medium-sized enterprises

• Proposes a data-driven approach in sourcing and inventory management for SMEs. • Exploits Observational and empirical data to generate decision support solutions. • Promote the adoption and initiation of artificial intelligence in SMEs. • Promising implementation results in a real-world case study. Elevated business uncertainties and competition over recent years have caused changes to the data-driven supply chain management of sourcing and inventories across industries. However, only large-sized enterprises have the resources to harness data for aiding their decision-making and planning. By contrast, small- and medium-sized enterprises (SMEs) commonly have limited resources and knowledge, which affects their ability to collect and utilize data. Thus, it is a challenge for them to implement advanced decision support tools to mitigate the effects of market uncertainties. This paper proposes a decision support system (DSS) for sourcing and inventory management, with the aims of helping SMEs compile and exploit data, and supporting their decisions under business ambiguities. The DSS was developed using a simulation-optimization approach by incorporating an artificial neural network and a genetic algorithm for problem representation and optimizing decision support solutions. The exploitation of observational and empirical data reduces the burden of data compilation obtained from unorganized data sources across SME operations. Further, uncertainty factors such as raw material demand, price, and supply lead time were considered. When implemented in a medium-sized food industry company, the DSS can provide decision support solutions that integrate the selection of recommended suppliers and optimal order quantities. It can also help decision-makers to shape their inventory management policies under uncertain raw material demands, while also considering service levels, sales promotions, lead times, and material availability from multiple suppliers. Consequently, implementation of the DSS helped to reduce the total purchased raw material costs by an average of 51.62% and reduced the order interval and on-hand inventory costs by an average of 54.24%.

A Human-Centric Model for Sustainable Asset Management in Railway: A Case Study

The ongoing digital transformation is changing asset management in the railway industry. Emerging digital technologies and Artificial Intelligence is expected to facilitate decision-making in management, operation, and maintenance of railway by providing an integrated data-driven and model-driven solution. An important aspect when developing decision-support solutions based on AI and digital technology is the users’ experience. User experience design process aims to create relevance, context-awareness, and meaningfulness for the end-user. In railway contexts, it is believed that applying a human-centric design model in the development of AI-based artefacts, will enhance the usability of the solution, which will have a positive impact on the decision-making processes. In this research, the applicability of such advanced technologies i.e., Virtual Reality, Mixed Reality, and AI have been reviewed for the railway asset management. To carry out this research work, literature review has been conducted related to available Virtual Reality/Augmented Reality/Mixed Reality technologies and their applications within railway industry. It has been found that these technologies are available, but not applied in railway asset management. Thus, the aim of this paper is to propose a human-centric design model for the enhancement of railway asset management using Artificial Intelligence, Virtual Reality, and Mixed Reality technologies. The practical implication of the findings from this work will benefit in increased efficiency and effectiveness of the operation and maintenance processes in railway.

COVID-Net CXR-S: Deep Convolutional Neural Network for Severity Assessment of COVID-19 Cases from Chest X-ray Images.

The world is still struggling in controlling and containing the spread of the COVID-19 pandemic caused by the SARS-CoV-2 virus. The medical conditions associated with SARS-CoV-2 infections have resulted in a surge in the number of patients at clinics and hospitals, leading to a significantly increased strain on healthcare resources. As such, an important part of managing and handling patients with SARS-CoV-2 infections within the clinical workflow is severity assessment, which is often conducted with the use of chest X-ray (CXR) images. In this work, we introduce COVID-Net CXR-S, a convolutional neural network for predicting the airspace severity of a SARS-CoV-2 positive patient based on a CXR image of the patient’s chest. More specifically, we leveraged transfer learning to transfer representational knowledge gained from over 16,000 CXR images from a multinational cohort of over 15,000 SARS-CoV-2 positive and negative patient cases into a custom network architecture for severity assessment. Experimental results using the RSNA RICORD dataset showed that the proposed COVID-Net CXR-S has potential to be a powerful tool for computer-aided severity assessment of CXR images of COVID-19 positive patients. Furthermore, radiologist validation on select cases by two board-certified radiologists with over 10 and 19 years of experience, respectively, showed consistency between radiologist interpretation and critical factors leveraged by COVID-Net CXR-S for severity assessment. While not a production-ready solution, the ultimate goal for the open source release of COVID-Net CXR-S is to act as a catalyst for clinical scientists, machine learning researchers, as well as citizen scientists to develop innovative new clinical decision support solutions for helping clinicians around the world manage the continuing pandemic.

Pharmacogenomics testing in patients with liver transplant and potential impact on prospective management.

Aim: Pharmacogenomics (PGx) tests are performed on whole-blood or saliva specimens. In patients with a transplanted liver, PGx results may be discordant with hepatic drug metabolizing enzyme activity. We evaluate the incidence and impact of PGx testing in liver transplant recipients, detail potential errors and describe clinical decision support (CDS) solution implemented. Materials & methods: A retrospective cohort study of liver transplant recipients at Mayo Clinic who underwent PGx testing between 1January 1996 and 7October 2019 were characterized. Impact of a CDS solution was evaluated. Results: There were 129 PGx tests in 117 patients. PGx testing incidence increased before (per year incidence rate ratio=1.45, 95% CI: 1.20-1.74, p<0.001) and after transplant (incidence rate ratio=1.48, 95% CI: 1.27-1.72, p<0.001). Three erroneous PGx tests were avoided 6 months following CDS implementation. Conclusion: Incidence of PGx testing in liver transplant recipients is increasing, leading to erroneous therapeutic decisions. CDS interventions and education are needed to prevent errors.

Contextual learning for energy forecasting in buildings

Energy consumers are becoming active players in the power and energy system. However, their informed and real-time responsiveness to the variations of renewable-based generation and, consequently, energy prices, is not possible without decision support solutions. This paper proposes a novel contextual learning approach for energy forecasting, which supports the decisions of Building Energy Management Systems (BEMS). The proposed forecasting approach includes a contextual dimension that identifies different observed contexts and clusters them according to their similarity. The identification of such contexts is used by the learning process of state-of-the-art artificial intelligence-based forecasting methods to select and adapt the most relevant data that is used in the training phase in each context. Forecasts for energy consumption, generation, temperature, brightness and occupancy are used by the BEMS to provide recommendations to the consumers and to support automated control of building devices. Real consumption, generation and contextual data gathered from several sensors in a building are used to validate the results, which show that the proposed contextual learning model improves forecasts of energy consumption, generation and other relevant factors for energy management in buildings.