Practical Decision Support System Research Articles

Firefighting Vehicle Dispatch and Route Planning Model Based on Priority Passage Rights

<div>Efficient fire rescue operations in urban environments are critical for saving lives and reducing property damage. By utilizing connected vehicle systems (CVS) for firefighting vehicles planning, we can reduce the response time to fires while lowering the operational costs of fire stations. This research presents an innovative nonlinear mixed-integer programming model to enhance fire rescue operations in urban settings. The model focuses on expediting the movement of firefighting vehicles within intricate traffic networks, effectively tackling the complexities associated with collaborative dispatch decisions and optimal path planning for multiple response units. This method is validated using a small-scale traffic network, providing foundational insights into parameter impacts. A case study in Sioux Falls shows its superiority over traditional “nearest dispatch” methods, optimizing both cost and response time significantly. Sensitivity analyses involving clearance speed, clearance time, minimum rescue force, and fire loss parameters contribute to the enhancement of urban fire rescue operations and the refinement of practical decision support systems.</div>

Development of mathematical methods for diagnosing kidney diseases using fuzzy set tools

An approach based on fuzzy set theory is presented in the scientific article to enhance the efficiency of diagnosing kidney diseases by decreasing the time required for making medical decisions. The suggested approach employs fuzzy models and algorithms that consider the uncertainty and variability of clinical data to optimize the assessment of the functional state of the kidneys, taking into account various risk factors and individual characteristics of patients. The paper suggests a technique to develop a system of fuzzy decision rules. This technique combines E. Shortliff’s iterative rules with functions from the studied classes of kidney diseases. Mathematical modeling and experimental studies have indicated relatively high effectiveness in classifying different forms of kidney diseases. The results can be used to formulate intelligent decision support systems in clinical practice and improve diagnostic and monitoring processes. Moreover, the findings may aid in shaping more targeted and effective health policies at the national and regional levels, enhancing access to healthcare, and promoting the population’s overall health.

Personalizing renal replacement therapy initiation in the intensive care unit: a reinforcement learning-based strategy with external validation on the AKIKI randomized controlled trials.

The timely initiation of renal replacement therapy (RRT) for acute kidney injury (AKI) requires sequential decision-making tailored to individuals' evolving characteristics. To learn and validate optimal strategies for RRT initiation, we used reinforcement learning on clinical data from routine care and randomized controlled trials. We used the MIMIC-III database for development and AKIKI trials for validation. Participants were adult ICU patients with severe AKI receiving mechanical ventilation or catecholamine infusion. We used a doubly robust estimator to learn when to start RRT after the occurrence of severe AKI for three days in a row. We developed a "crude strategy" maximizing the population-level hospital-free days at day 60 (HFD60) and a "stringent strategy" recommending RRT when there is significant evidence of benefit for an individual. For validation, we evaluated the causal effects of implementing our learned strategies versus following current best practices on HFD60. We included 3748 patients in the development set and 1068 in the validation set. Through external validation, the crude and stringent strategies yielded an average difference of 13.7 [95% CI -5.3 to 35.7] and 14.9 [95% CI -3.2 to 39.2] HFD60, respectively, compared to current best practices. The stringent strategy led to initiating RRT within 3 days in 14% of patients versus 38% under best practices. Implementing our strategies could improve the average number of days that ICU patients spend alive and outside the hospital while sparing RRT for many. We developed and validated a practical and interpretable dynamic decision support system for RRT initiation in the ICU.

Agrarian Synthesis and Precision Cultivation Optimization System

The ever-growing demand for food production calls for innovative solutions in agriculture. This research introduces a machine learning-based approach, specifically utilizing logistic regression, to predict optimal crops based on soil and weather conditions. The dataset encompasses crucial attributes including Nitrogen (N), Phosphorus (P), Potassium (K), temperature, humidity, pH, rainfall, with corresponding crop labels. The proposed methodology employs logistic regression, a powerful classification algorithm, to model the relationships between input features and crop types. Through careful feature engineering, the model is fine-tuned to enhance its predictive accuracy. Rigorous evaluation metrics validate the model's performance, ensuring its reliability in real-world applications. Results showcase the logistic regression model's efficacy in accurately predicting suitable crops for given soil and weather parameters. This predictive tool serves as a practical decision support system for farmers, aiding in crop selection and resource allocation. This research contributes to the synergy of machine learning and agriculture, showcasing logistic regression as a valuable tool for crop prediction and resource optimization. As technology continues to transform traditional farming, the integration of logistic regression in precision agriculture offers a practical and efficient approach to crop selection.

Artificial intelligence for predicting mortality in hospitalized COVID-19 patients.

The global demographic situation has been significantly impacted by the COVID-19 pandemic. The objective of this study was to develop a model that predicts the risk of COVID-associated mortality using clinical and laboratory data collected within 72 h of hospital admission. A total of 3024 subjects with PCR-confirmed COVID-19 were admitted to Almazov National Research Medical Center between May 2020 and August 2021. Among them, 6.25% (n = 189) of patients had a fatal outcome. Five machine learning models and the Boruta-SHAP feature selection method were utilized to assess the risk of mortality during COVID-19 hospitalization. All methods demonstrated high efficacy, with ROC AUC (Receiver Operating Characteristic Area Under the Curve) values exceeding 80%. The selected Boruta-SHAP features, when incorporated into the random forest model, achieved an ROC AUC of 93.1% in the validation. Throughout the study, close collaboration with healthcare professionals ensured that the developed tool met their practical needs. The success of our model validates the potential of machine learning techniques as decision support systems in clinical practice.

Effective decision support systems in clinical practice and prevention: literature review

Clinical decision support systems (CDSS) often outperform human capabilities for processing a large amount of information, dramatically simplifying the work of specialists and avoiding medical errors. The implementation of such systems is a complex task that requires high-tech developments. The annual increase in the development of such systems has a geometric progression. However, it is unclear if most of them will be integrated into clinical practice and recommendations. The use of CDSS to address various disease diagnosis, treatment, and prevention issues is demonstrated, and possible linkages between scientific clinical observations and CDSS are examined. Currently, many data gathering and processing systems use machine learning algorithms and convolutional technologies to create CDSS, resulting in data that exceeds the ability of human thinking to determine the logic of recommended decisions. This study presents the most studied modern CDSS, the possibilities of their application, and the implementation issues.

Psychological assessment of AI-based decision support systems: tool development and expected benefits.

This study aimed to develop an evaluation tool that assesses the use of AI-based decision support systems (DSSs) in professional practice from a human-centered perspective. Following the International Organization for Standardization, this perspective aims to ensure that the use of interactive technologies improves users' psychological load experience and behavior, e.g., in the form of reduced stress experience or increased performance. Concomitantly, this perspective attempts to proactively prevent or detect and correct the potential negative effects of these technologies on user load, such as impaired satisfaction and engagement, as early as possible. Based on this perspective, we developed and validated a questionnaire instrument, the Psychological Assessment of AI-based DSSs (PAAI), for the user-centered evaluation of the use of AI-based DSSs in practice. In particular, the instrument considers central design characteristics of AI-based DSSs and the corresponding work situation, which have a significant impact on users' psychological load. The instrument was tested in two independent studies. In Study 1, N = 223 individuals were recruited. Based on the results of item and scale analyses and an exploratory factor analysis, the newly developed instrument was refined, and the final version was tested using a confirmatory factor analysis. Findings showed acceptable-to-good fit indices, confirming the factorial validity of the PAAI. This was confirmed in a second study, which had N = 471 participants. Again, the CFA yielded acceptable-to-good fit indices. The validity was further confirmed using convergent and criterion validity analyses.

Enhancing community resilience in arid regions: A smart framework for flash flood risk assessment

This paper presents a novel framework for smart integrated risk management in arid regions. The framework combines flash flood modelling, statistical methods, artificial intelligence (AI), geographic evaluations, risk analysis, and decision-making modules to enhance community resilience. Flash flood is simulated by using Watershed Modelling System (WMS). Statistical methods are also used to trim outlier data from physical systems and climatic data. Furthermore, three AI methods, including Support Vector Machine (SVM), Artificial Neural Network (ANN), and Nearest Neighbours Classification (NNC), are used to predict and classify flash flood occurrences. Geographic Information System (GIS) is also utilised to assess potential risks in vulnerable regions, together with Failure Mode and Effects Analysis (FMEA) and Hazard and Operability Study (HAZOP) methods. The decision-making module employs the Classic Delphi technique to classify the appropriate solutions for flood risk control. The methodology is demonstrated by its application to the real case study of the Khosf region in Iran, which suffers from both drought and severe floods simultaneously, exacerbated by recent climate changes. The results show high Coefficient of determination (R2) scores for the three AI methods, with SVM at 0.88, ANN at 0.79, and NNC at 0.89. FMEA results indicate that over 50% of scenarios are at high flood risk, while HAZOP indicates 30% of scenarios with the same risk rate. Additionally, peak flows of over 24 m3/s are considered flood occurrences that can cause financial damage in all scenarios and risk techniques of the case study. Finally, our research findings indicate a practical decision support system that is compatible with sustainable development concepts and can enhance community resilience in arid regions.

Natural Language Processing Markers for Psychosis and Other Psychiatric Disorders: Emerging Themes and Research Agenda From a Cross-Linguistic Workshop.

This workshop summary on natural language processing (NLP) markers for psychosis and other psychiatric disorders presents some of the clinical and research issues that NLP markers might address and some of the activities needed to move in that direction. We propose that the optimal development of NLP markers would occur in the context of research efforts to map out the underlying mechanisms of psychosis and other disorders. In this workshop, we identified some of the challenges to be addressed in developing and implementing NLP markers-based Clinical Decision Support Systems (CDSSs) in psychiatric practice, especially with respect to psychosis. Of note, a CDSS is meant to enhance decision-making by clinicians by providing additional relevant information primarily through software (although CDSSs are not without risks). In psychiatry, a field that relies on subjective clinical ratings that condense rich temporal behavioral information, the inclusion of computational quantitative NLP markers can plausibly lead to operationalized decision models in place of idiosyncratic ones, although ethical issues must always be paramount.

Valetudinarian Situation Categorize Using Drug Reviews

Human healthcare is one of society's most important issues. To ensure that patients receive the care they require as quickly as possible, it searches for the best diagnosis and thorough disease. For the health-related component of searching, other disciplines are required, such as statistics and computer science, because this recognition is frequently complex[3]. These disciplines must go beyond the traditional ones in order to follow innovative methodologies. Because there are so many new strategies, it is possible to give a general summary without focusing on certain details. In order to do this, we propose a thorough examination of machine learning-related disorders in humans. In order to find interesting trends, make unimportant forecasts, and aid in decision-making[1], this research focuses on existing approaches connected to machine learning growth applied to the diagnosis of human disorders in the medical industry. In order to produce suitable decision support, this study examines distinctive machine learning methods employed in healthcare applications[4], [5]. With the goal of developing a practical decision support system for medical applications, the research gap is to be filled up by this work. Key Words: Human disease, Machine learning, Neural Networks.

Decision support software-guided medication reviews in elderly patients with polypharmacy: a prospective analysis of routine data from community pharmacies (OPtiMed study protocol).

Pharmacist-led medication reviews are considered a valuable measure to address risks of polypharmacy. The software Medinspector® is used in community pharmacies to assist the performance of this complex service by structuring the medication review process and supporting pharmacists in their decision-making with targeted clinical knowledge. Key feature is a computerized risk assessment of both the initial and adjusted medication regimen of a patient in multiple domains, thus aiming to support the identification and solving of drug-related problems. This study will examine the effects of medication reviews performed with the clinical decision support system in daily routine practice on medication-related and patient-reported outcomes in elderly patients with polypharmacy. A prospective, before-after observational study is conducted in German community pharmacies aiming to include 148 patients aged 65 or older, who chronically use five or more active pharmaceutical substances with systemic effects and utilize the software-supported medication review service. The study is based on routine documentation within the software over the course of the medication review, including a patient's baseline medication, the medication proposed by pharmacists, and the final medication regimen. A software-implemented questionnaire comprising self-developed and literature-derived instruments is used to collect patient-reported outcome data at baseline and follow-up. Primary outcome is the appropriateness of medication measured with an adapted version of the Medication Appropriateness Index (MAI). Secondary medication-related outcomes are medication underuse, exposition towards anticholinergic/sedative drugs, number of drugs in long-term use and the implementation of pharmacist-proposed medication adjustments by the physicians. Secondary patient-reported outcomes are symptom burden, medication-related quality of life, adherence, fulfillment of medication review-related goals, and perception of the service. With the recently introduced remuneration of community pharmacist-led MR in Germany, the demand for digital tools supporting the MR process is assumed to rise. The OPtiMed-study is expected to create evidence on the effects of a novel tool on patient care in a vulnerable patient population. Trial registration German Clinical Trials Register, DRKS00027410. Registered 22 December 2021, https://www.drks.de/drks_web/navigate.do?navigationId=trial.HTML&TRIAL_ID=DRKS00027410 . Also available on the WHO meta-registry: https://trialsearch.who.int/?TrialID=DRKS00027410.

Clinical Decision Support Systems in Practice: Status and Challenges

Decision support systems (DSS) are computer programs supported AI styles that contribute to reaching an accurate decision in an frequently-narrow sphere of interest. Clinical decision support systems( CDSS) are similar DSSs which may be used by medical professionals in conventions and hospitals. They ’re used for opinion, treatment protocol recommendations, treatment outgrowth prognostications and other tasks. CDSS are constructed supported emblematic and machine literacy ( including deep literacy) approaches to represent and infer medical knowledge. The end of this work is to supply an overview of history and current styles in designing a successful CDSS. The study considers the systems that were claimed to be enforced in clinical practice. presently, the event of a CDSS is generally pursued in two directions 1) a more traditional approach grounded on rules, ontologies, probabilistic models, and therefore the application of norms; 2) machine literacy grounded approach. Both approaches could indeed be used reciprocal within a healthcare data system. This work seeks to supply an objective view on the advantages and limitations of the approaches as well to bandy unborn exploration avenues that could lead to more accurate and secure CDSS and bettered healthcare Keywords- AI, CLINICAL DECISION NETWORK, DECISION NETWORK, DEEP LITERACY, HEALTHCARE.

Deep learning-based image analysis predicts PD-L1 status from H&E-stained histopathology images in breast cancer

Programmed death ligand-1 (PD-L1) has been recently adopted for breast cancer as a predictive biomarker for immunotherapies. The cost, time, and variability of PD-L1 quantification by immunohistochemistry (IHC) are a challenge. In contrast, hematoxylin and eosin (H&E) is a robust staining used routinely for cancer diagnosis. Here, we show that PD-L1 expression can be predicted from H&E-stained images by employing state-of-the-art deep learning techniques. With the help of two expert pathologists and a designed annotation software, we construct a dataset to assess the feasibility of PD-L1 prediction from H&E in breast cancer. In a cohort of 3,376 patients, our system predicts the PD-L1 status in a high area under the curve (AUC) of 0.91 – 0.93. Our system is validated on two external datasets, including an independent clinical trial cohort, showing consistent prediction performance. Furthermore, the proposed system predicts which cases are prone to pathologists miss-interpretation, showing it can serve as a decision support and quality assurance system in clinical practice.

Identifying Key Risk Factors in Product Development Projects

In a rapidly changing world, most operational processes of enterprises are conducted in the form of development projects. The development of new products is an important organizational strategy to meet consumer needs. During this process, enterprises often encounter many bottlenecks and risks that can cause delays in, and even the failure of, development projects. In this study, we developed a research framework based on relevant literature and expert interviews and then used the decision making trial and evaluation laboratory (DEMATEL) and the analytic network process (ANP) to determine the relationships among and the importance of risks in the development of new products. The results of a case study show that the six key risks of product development projects include project completion time, mastery of key technical capabilities, controlling the progress of the project, uniqueness and complexity of the project, ability to control the market, and functional integrity of the product. According to the results of importance performance analysis, six key factors were classified in the concentrate quadrant. The optoelectronic manufacturing industry should focus on reducing risks to the project. A cause-and-effect diagram shows that if an enterprise wants to improve performance in terms of these key factors, it should first improve the project completion time or the mastery of key technical capabilities. Therefore, it is appropriate to start by improving the project completion time. In this study, we developed a practical and simple decision support system that allows managers of research and development to examine the risk of projects and assess the relevant risks. A case study was also conducted to test the accuracy of the proposed risk-management method.

Concordance rate of radiologists and a commercialized deep-learning solution for chest X-ray: Real-world experience with a multicenter health screening cohort.

Lunit INSIGHT CXR (Lunit) is a commercially available deep-learning algorithm-based decision support system for chest radiography (CXR). This retrospective study aimed to evaluate the concordance rate of radiologists and Lunit for thoracic abnormalities in a multicenter health screening cohort. We retrospectively evaluated the radiology reports and Lunit results for CXR at several health screening centers in August 2020. Lunit was adopted as a clinical decision support system (CDSS) in routine clinical practice. Subsequently, radiologists completed their reports after reviewing the Lunit results. The DLA result was provided as a color map with an abnormality score (%) for thoracic lesions when the score was greater than the predefined cutoff value of 15%. Concordance was achieved when (a) the radiology reports were consistent with the DLA results ("accept"), (b) the radiology reports were partially consistent with the DLA results ("edit") or had additional lesions compared with the DLA results ("add"). There was discordance when the DLA results were rejected in the radiology report. In addition, we compared the reading times before and after Lunit was introduced. Finally, we evaluated systemic usability scale questionnaire for radiologists and physicians who had experienced Lunit. Among 3,113 participants (1,157 men; mean age, 49 years), thoracic abnormalities were found in 343 (11.0%) based on the CXR radiology reports and 621 (20.1%) based on the Lunit results. The concordance rate was 86.8% (accept: 85.3%, edit: 0.9%, and add: 0.6%), and the discordance rate was 13.2%. Except for 479 cases (7.5%) for whom reading time data were unavailable (n = 5) or unreliable (n = 474), the median reading time increased after the clinical integration of Lunit (median, 19s vs. 14s, P < 0.001). The real-world multicenter health screening cohort showed a high concordance of the chest X-ray report and the Lunit result under the clinical integration of the deep-learning solution. The reading time slight increased with the Lunit assistance.

Utilizing AI for Health Promotion and Disease Prevention

The emergence of advanced AI technologies such as expert systems, fuzzy logic, neural networks, and genetic algorithms can directly address people's health needs by providing personalized advice to users while at the same time collecting valuable data related to aspects of diseases and treatments. These technologies have been in use for quite some time and have evolved to the point where they have matured into practical decision support systems that can be used to assist people in health management and promotion efforts. This chapter presents a survey of the current progress and ongoing research in the area of health-related AI technology and end-user products. With the ever-growing trajectory of technology and the increasing demand to reduce healthcare costs while providing better quality healthcare, employing AI as part of or adjunct to medical processes will become increasingly important from both economic and social perspectives. The classical problems of the healthcare industry include providing people with new and better treatments, more effective ways of preventing, detecting, and combating multifaceted chronic diseases, and ways of caring for the ever-increasing numbers of senior citizens.

Smart Airport: Evaluation of Performance Standards and Technologies for a Smart Logistics Zone

Owing to the current context of Industry 4.0, the importance of smart technologies in airport systems have increased substantially. State-of-art applications for transportation planning incorporating baggage services, routing, security, and safety are an evolving domain for both practitioners and researchers dealing with aviation applications. In this context, this paper seeks to answer these questions: Which standards are aimed at a smart airport to make the transportation planning sound? Which propositions are made based on the obtained prioritized standards? This study deals with standards including Environmental Effects, Docking &amp; Navigation, Object Detection &amp; Protection, Communications &amp; Integration, and Terminology using a practical decision support system based on an analytical hierarchy process (AHP) and fuzzy inference system (FIS). Computational results reveal that the Object Detection &amp; Protection standard has an effect on a safe and smart system. To give an overview of this standard for a smart logistics zone (SLZ), an architecture of autonomous robot units and a baggage handling system is proposed in this study. The suggested approach analyzes the obstacle photos obtained by cameras and allows the end-user to control the calculations visually. This research could provide advice to airport planners about smart policies and improving operations.

Application of risk-based fuzzy decision support systems in new product development: An R-VIKOR approach

Innovative manufacturing firms strive to sustain and enhance their competitive advantages by running a range of new product development (NPD) projects in a consistent manner. The capital and time required to execute the NPD projects have substantially increased over the past years. This magnified the risk-aversion behavior of R&D managers and has increased their sensitivity towards the underlying risk of NPD projects. In particular, the R&D departments have recently started to proactively assess the accuracy of ambiguous information that is extensively used in preliminary market study and customer requirements analysis. Thanks to its high performance in dynamic environments, the R-numbers method can be employed to capture and analyze the risk of fuzzy numbers in a variety of decision making models. To tackle the complexity of such analysis, this paper proposes a novel risk-based fuzzy VIKOR (R-VIKOR) methodology. Using the interpretive structural modeling, the risk factors are first classified to identify and rank the existing critical risk factors of NPD projects. The ultimate goal of this study is to develop a practical yet simple decision support system tool that enables the R&D managers to effectively examine the riskiness of fuzzy information and assess the relevant risk factors. A real-world case study is presented to test and examine the accuracy and effectiveness of the proposed risk management method.

Segmentation of Brain Tumors from MRI Images Using Convolutional Autoencoder

The use of machine learning algorithms and modern technologies for automatic segmentation of brain tissue increases in everyday clinical diagnostics. One of the most commonly used machine learning algorithms for image processing is convolutional neural networks. We present a new convolutional neural autoencoder for brain tumor segmentation based on semantic segmentation. The developed architecture is small, and it is tested on the largest online image database. The dataset consists of 3064 T1-weighted contrast-enhanced magnetic resonance images. The proposed architecture’s performance is tested using a combination of two different data division methods, and two different evaluation methods, and by training the network with the original and augmented dataset. Using one of these data division methods, the network’s generalization ability in medical diagnostics was also tested. The best results were obtained for record-wise data division, training the network with the augmented dataset. The average accuracy classification of pixels is 99.23% and 99.28% for 5-fold cross-validation and one test, respectively, and the average dice coefficient is 71.68% and 72.87%. Considering the achieved performance results, execution speed, and subject generalization ability, the developed network has great potential for being a decision support system in everyday clinical practice.

Neighbor-aware review helpfulness prediction

Helpfulness prediction techniques have been widely incorporated into online decision support systems to identify high-quality reviews. Most current studies on helpfulness prediction assume that a review's helpfulness only relies on information from itself. In practice, however, consumers hardly process reviews independently because reviews are displayed in sequence; a review is more likely to be affected by its adjacent neighbors in the sequence, which is largely understudied. In this paper, we proposed the first end-to-end neural architecture to capture the missing interaction between reviews and their neighbors. Our model allows for a total of 12 (three selection × four aggregation) schemes that contextualize a review into the context clues learned from its neighbors. We evaluated our model on six domains of real-world online reviews against a series of state-of-the-art baselines. Experimental results confirm the influence of sequential neighbors on reviews and show that our model significantly outperforms the baselines by 1% to 5%. We further revealed how reviews are influenced by their neighbors during helpfulness perception via extensive analysis. The results and findings of our work provide theoretical contributions to the field of review helpfulness prediction and offer insights into practical decision support system design.