Health Decision Support System Research Articles

Swarm Intelligence Based Public Health Informatics and Decision Support Systems

Swarm Intelligence Based Public Health Informatics and Decision Support Systems

Development of a Big Data Analysis and Management Decision Support System for Student Mental Health in Higher Education

Abstract In recent years, the number of psychological problems occurring in college students has been increasing day by day, and timely, convenient, and accurate psychological warning is an important way to prevent college students from psychological crises. Starting from the basic needs of the system and aiming at realizing the embedding of data mining technology in the psychological management system, this paper elaborates on the design and implementation of the data mining technology module suitable for the psychological management system. The clustering algorithm-K mean algorithm of data mining is used to distinguish groups of students with different categories of psychological problems, which provides data support for the next decision-making. Through simulation experiments between the system designed in this paper and four other mental health management systems, the clustering performance and system evaluation performance of this paper’s system are better than those of the other systems. The result of mining the mental health of college students through this system shows that 82.2% of the student’s assessment results show normal, and 8.35% of the students need to focus on, and this result is not much different from that obtained from the statistical survey of a personality questionnaire, which indicates that the use of clustering analysis in the system can be an effective prediction of the mental health of college students.

Behavioral Health Decision Support Systems and User Interface Design in the Emergency Department.

The objective of this qualitative study is to gauge physician sentiment about an emergency department (ED) clinical decision support (CDS) system implemented in multiple adult EDs within a university hospital system. This CDS system focuses on predicting patients' likelihood of ED recidivism and/or adverse opioid-related events. The study was conducted among adult emergency physicians working in three EDs of a single academic health system in Rhode Island. Qualitative, semistructured interviews were conducted with ED physicians. Interviews assessed physicians' prior experience with predictive analytics, thoughts on the alert's placement, design, and content, the alert's overall impact, and potential areas for improvement. Responses were aggregated and common themes identified. Twenty-three interviews were conducted (11 preimplementation and 12 postimplementation). Themes were identified regarding each physician familiarity with predictive analytics, alert rollout, alert appearance and content, and on alert sentiments. Most physicians viewed these alerts as a neutral or positive EHR addition, with responses ranging from neutral to positive. The alert placement was noted to be largely intuitive and nonintrusive. The design of the alert was generally viewed positively. The alert's content was believed to be accurate, although the decision to respond to the alert's call-to-action was physician dependent. Those who tended to ignore the alert did so for a few reasons, including already knowing the information the alert contains, the alert offering information that is not relevant to this particular patient, and the alert not containing enough information to be useful. Ultimately, this alert appears to have a marginally positive effect on ED physician workflow. At its most beneficial, the alert reminded physicians to deeply consider the care provided to high-risk populations and to potentially adjust their care and referrals. At its least beneficial, the alert did not affect physician decision-making but was not intrusive to the point of negatively impacting workflow.

Predicting the evolution of COVID-19 mortality risk: A Recurrent Neural Network approach

Background:In December 2020, the COVID-19 disease was confirmed in 1,665,775 patients and caused 45,784 deaths in Spain. At that time, health decision support systems were identified as crucial against the pandemic. Methods:This study applies Deep Learning techniques for mortality prediction of COVID-19 patients. Two datasets with clinical information were used. They included 2,307 and 3,870 COVID-19 infected patients admitted to two Spanish hospitals. Firstly, we built a sequence of temporal events gathering all the clinical information for each patient, comparing different data representation methods. Next, we used the sequences to train a Recurrent Neural Network (RNN) model with an attention mechanism exploring interpretability. We conducted an extensive hyperparameter search and cross-validation. Finally, we ensembled the resulting RNNs to enhance sensitivity. Results:We assessed the performance of our models by averaging the performance across all the days in the sequences. Additionally, we evaluated day-by-day predictions starting from both the hospital admission day and the outcome day. We compared our models with two strong baselines, Support Vector Classifier and Random Forest, and in all cases our models were superior. Furthermore, we implemented an ensemble model that substantially increased the system’s sensitivity while producing more stable predictions. Conclusions:We have shown the feasibility of our approach to predicting the clinical outcome of patients. The result is an RNN-based model that can support decision-making in healthcare systems aiming at interpretability. The system is robust enough to deal with real-world data and can overcome the problems derived from the sparsity and heterogeneity of data.

An analytics-based decision support system for evaluating the fiscal health of academic programs

Academia’s business model is significantly different than most other businesses. The underlying fiscal unit is the academic program. The fiscal model is complex when balancing full-time students attending with flat-rate tuition or per credit and faculty members who may teach in multiple programs and may have other administrative or scholarship duties. This study presents an analytics-based Decision Support System (DSS) for academic programming and fiscal health. The proposed model computes, without averaging, the operating/direct contribution margin of academic programs. The model uses descriptive analytics to provide information for predictive analytics—such as developing business cases for the new program(s), diagnostic analytics for sensitivity analysis—such as a “what if” analysis of program data and/or a faculty hiring rubric, and prescriptive analytics—such as interactive forecasting for choosing the best alternative courses of action. The model takes the guesswork out and allows academic officers to convert data into knowledge and insight for effective decision-making and academic programming.

Human-centered XAI: Developing design patterns for explanations of clinical decision support systems

Much of the research on eXplainable Artificial Intelligence (XAI) has centered on providing transparency of machine learning models. More recently, the focus on human-centered approaches to XAI has increased. Yet, there is a lack of practical methods and examples on the integration of human factors into the development processes of AI-generated explanations that humans prove to uptake for better performance. This paper presents a case study of an application of a human-centered design approach for AI-generated explanations. The approach consists of three components: Domain analysis to define the concept & context of explanations, Requirements elicitation & assessment to derive the use cases & explanation requirements, and the consequential Multi-modal interaction design & evaluation to create a library of design patterns for explanations. In a case study, we adopt the DoReMi-approach to design explanations for a Clinical Decision Support System (CDSS) for child health. In the requirements elicitation & assessment, a user study with experienced paediatricians uncovered what explanations the CDSS should provide. In the interaction design & evaluation, a second user study tested the consequential interaction design patterns. This case study provided a first set of user requirements and design patterns for an explainable decision support system in medical diagnosis, showing how to involve expert end users in the development process and how to develop, more or less, generic solutions for general design problems in XAI.

Will high-speed railway influence the healthcare seeking behaviour of patients? Quasi-experimental evidence from China.

This study examines the impacts of high-speed railway (HSR) transportation on the healthcare-seeking behaviour of patients along newly integrated areas of Sichuan province, China. The opening of the Cheng-Mian-Le intercity HSR is considered as quasi-experimental evidence from China, and we make a propensity score matching the difference in differences research design, using data from the monthly report database of the Sichuan Province health statistical data collection and Decision Support System from 2014 to 2015. We find that, first, the opening of the HSR resulted in significant healthcare-seeking behaviour with great heterogeneity. Second, patients are more likely to go to areas with high-density healthcare resources, in which case HSR may mitigate the diagnostic inaccuracies that patients face locally. Third, the 'distance enhancement effect' was present, and its marginal effect is more significant for long-distance patients. Fourth, the tiered-network healthcare policy has no significant restrictive impact on patients seeking high-level medical services. Our results show that HSR establishment has a substantial impact on the behaviour of people seeking medical treatment and medicine. Furthermore, we discuss the results' policy implications for the allocation and integration of China's healthcare market, and the accessibility of medical and health services.

Predictive modeling based on small data in clinical medicine: RBF-based additive input-doubling method.

The paper considers the problem of handling short sets of medical data. Effectively solving this problem will provide the ability to solve numerous classification and regression tasks in case of limited data in health decision support systems. Many similar tasks arise in various fields of medicine. The authors improved the regression method of data analysis based on artificial neural networks by introducing additional elements into the formula for calculating the output signal of the existing RBF-based input-doubling method. This improvement provides averaging of the result, which is typical for ensemble methods, and allows compensating for the errors of different signs of the predicted values. These two advantages make it possible to significantly increase the accuracy of the methods of this class. It should be noted that the duration of the training algorithm of the advanced method remains the same as for existing method. Experimental modeling was performed using a real short medical data. The regression task in rheumatology was solved based on only 77 observations. The optimal parameters of the method, which provide the highest prediction accuracy based on MAE and RMSE, were selected experimentally. A comparison of its efficiency with other methods of this class has been performed. The highest accuracy of the proposed RBF-based additive input-doubling method among the considered ones is established. The method can be modified by using other nonlinear artificial intelligence tools to implement its training and application algorithms and such methods can be applied in various fields of medicine.

BigO: A public health decision support system for measuring obesogenic behaviors of children in relation to their local environment

BigO: A public health decision support system for measuring obesogenic behaviors of children in relation to their local environment

BigO: A public health decision support system for measuring obesogenic behaviors of children in relation to their local environment.

Obesity is a complex disease and its prevalence depends on multiple factors related to the local socioeconomic, cultural and urban context of individuals. Many obesity prevention strategies and policies, however, are horizontal measures that do not depend on context-specific evidence. In this paper we present an overview of BigO (http://bigoprogram.eu), a system designed to collect objective behavioral data from children and adolescent populations as well as their environment in order to support public health authorities in formulating effective, context-specific policies and interventions addressing childhood obesity. We present an overview of the data acquisition, indicator extraction, data exploration and analysis components of the BigO system, as well as an account of its preliminary pilot application in 33 schools and 2 clinics in four European countries, involving over 4,200 participants.

Big data analytics in health: an overview and bibliometric study of research activity.

The study presents an overview of the research activity in Big Data Analytics (BDA) in the field of health and demonstrates the existing knowledge through related examples. The objective is to inform health librarians about the nature and magnitude of the technological innovations in health information analysis tools, its influence, and where and how further material could be searched. We performed a bibliometric and co-citation analysis within a total of 804 papers published between 2000 and 2016 and retrieved from the Web of Science and Scopus databases. Using the NVivo text analysis software, we identified the stakeholders of BDA in health and innovative decision support systems in the field. Our findings show a tremendous increase in published papers after 2014. Most of them are relevant to neurology and medical oncology. The stakeholders are clinicians, researchers, patients, administrators, IT specialists, vendors and policymakers. New BDA tools in medicine are mostly developed for disease monitoring purposes while they utilise visualisation to identify disease patterns and statistical analysis of past data for making predictions. Health analytics provide a unique opportunity for advancing health information research and medical decision making. It provides health information professionals with new tools in problem-solving offering new perspectives in prognosis and diagnosis of diseases.

A Health Decision Support System for Disease Diagnosis based on Machine Learning via Big Data

A Health Decision Support System for Disease Diagnosis based on Machine Learning via Big Data

Functionality and Interfaces of a Herd Health Decision Support System for Practising Dairy Cattle Veterinarians in New Zealand

Decision-making processes to assess and improve the health of dairy herds are often unstructured due to the complexity of interactions that exist between the health and productivity of the herd, for which there are no ready to hand solutions. Decisions made in the face of these complex herd health problems are often based on the experience and perceptions of what might be a quick or the easiest solution. To shift from this unstructured process to semistructured decision-making requires a more holistic understanding of potential health problems and access to herd productivity information and to analytical methods suitable for examining and evaluating such data. Technological advances in agriculture have made the development of such information technology systems both possible and relatively accessible to decision makers working with dairy herds (e.g., veterinarians). The timely access and appropriate analysis of herd productivity data provides the herd health advisor with the opportunity to track and benchmark the performance of dairy herds. Thus, a decision support system (DSS) will use best available evidence to guide the allocation of resources to specific, most promising herd health interventions. This article presents an example of a DSS-based on collection of data and algorithm of analysis.

A Health Decision Support System for Disease Diagnosis Based on Wearable Medical Sensors and Machine Learning Ensembles

Even with an annual expenditure of more than $3 trillion, the U.S. healthcare system is far from optimal. For example, the third leading cause of death in the U.S. is preventable medical error, immediately after heart disease and cancer. Computer-based clinical decision support systems (CDSSs) have been proposed to address such deficiencies and have significantly improved clinical practice over the past decade. However, they remain limited to clinics and hospitals, and do not take advantage of patient data that are obtained on a daily basis using wearable medical sensors (WMSs) that have the ability to bridge this information gap. WMSs can collect physiological signals from anyone anywhere anytime. Thus, they have the potential to usher in an era of pervasive healthcare. However, most prior work on WMSs only focuses on hardware and protocol design, and not on an information system that can fully utilize the collected signals for efficient disease diagnosis. In this paper, for the first time, we introduce a hierarchical health decision support system for disease diagnosis that integrates health data from WMSs into CDSSs. The proposed system has a multi-tier structure, starting with a WMS tier, backed by robust machine learning, that enables diseases to be tracked individually by a disease diagnosis module. We demonstrate the feasibility of such a system through six disease diagnosis modules aimed at four ICD-10-CM disease categories. We show that the system is scalable using five more disease categories. Just the WMS tier offers impressive diagnostic accuracies for various diseases: arrhythmia (86 percent), type-2 diabetes (78 percent), urinary bladder disorder (99 percent), renal pelvis nephritis (94 percent), and hypothyroid (95 percent). We estimate that the disease diagnosis modules of all known 69,000 human diseases would require just 62 GB of storage space in the WMS tier. This is practical even in today’s cloud or base station oriented WMS systems.

Blogs as Channels for Disseminating Health Technology Innovations.

ObjectivesThe objective of this study was to describe the features of health informatics blogs on the Internet.MethodsA search was conducted in August, 2016 using the search engine, Google, and key words: ‘mobile health blog,’ ‘telehealth/telemedicine blog,’ ‘Electronic Health Record blog,’ ‘personalized health record blog,’ ‘population health decision support system blog,’ and ‘public/population health dashboard blog.’ The first 24 blogs resulting from each key word search were recorded, generating 144 blogs. A total of 109 unique blogs resulted after removing duplicates and non-functional sites.ResultsBlogs with ‘.com’ extensions were most prevalent (72%, n = 79). More than half of the blogs (79%, n = 86) were created by industries. Mobile health (88%, n = 96), telehealth (82%, n = 89), and health IT (78%, n = 85) were the predominant topics covered. Health providers (44%, n = 48), industries (33%, n = 36), patients/consumers (25%, n = 27) and payers/insurance providers (19%, n = 21) constituted the most common target audience. Blogs catering to payers commonly used ‘.org’ extension (n = 10 out of 21), compared to ‘.com’ (n = 7) or ‘.gov’ (n = 2) (p < 0.0001). Significant differences were also observed by topics covered health IT (p = 0.007), subscription (p = 0.048) and LinkedIn social media (p = 0.019) across the website extensions.ConclusionsFurther research is needed to examine the use of blogs as channels of communication of best evidence in health informatics research among diverse stakeholders. The role of blogs as policy informatics tools need to be evaluated in order for stakeholders to collaborate, coordinate and share opportunities and challenges of various public health programs and policies.

Incorporating Pharmacogenomics into Health Information Technology, Electronic Health Record and Decision Support System: An Overview

As the adoption of information technology in healthcare is rising, the potentiality of moving Pharmacogenomics from benchside to bedside is aggravated. This paper reviews the current status of Pharmacogenomics (PGx) information and the attempts for incorporating them into the Electronic Health Record (EHR) system through Decision Support Systems (DSSs). Rigorous review strategies of PGx information and providing context-relevant recommendations in form of action plan- dose adjustment, lab tests rather than just information- would be ideal for making clinical recommendations out of PGx information. Lastly, realistic projections of what pharmacogenomics can provide is another important aspect in incorporating Pharmacogenomics into health information technology.

Ontology based Framework for Hereditary Disease Identification

Health plays an important role in the life of every individual world and it is a major concern all over the world. Healthcare systems have gained significant attention and growth in the recent years. Healthcare system needs mechanism to store health details in electronic form. Effective healthcare systems can be built by integrating heterogeneous information like patient details, family history, disease, and symptoms. The decision system for identification of hereditary disease using family members’ medical record is not available in existing environment. Due to the heterogonous formats available in health records and lack of common vocabulary, the accessibility and utility of the big data of healthcare is very minimal for health data analytics and decision support systems. This work proposes a framework that identifies hereditary disease based on family members’ health record. To identify close blood relatives of a patient a complete semantic knowledge base, rule base for relationship type identification and health record ontologies are used. Automatic decision making capability of the proposed framework helps to provide patient specific assessments and recommendations based on family medical history. The performance of this work is measured in terms of precision, sensitivity and specificity. The obtained precision and sensitivity is between seventy to eighty percent.

Housing Health and Safety Decision Support System with Augmented Reality

Europe's residents spend more than 90% of their time in an enclosed environment. Hence generating conditions for a healthy lifestyle and bettering the quality of life in enclosed facilities take on special meaning because these can increase the work productivity of the residents, improve their quality of life, lower the rates of illnesses and save on expenses designated for medical treatments. The above-mentioned issues are related to a built environment's air pollution, the premise's microclimate, health and safety effects, climate change, real estate market value and the like. However, the Housing Health and Safety Decision Support System with Augmented Reality (HUSSAR) can analyse the above factors in an integrated way. The following models of HUSSAR are aimed to perform this function: Pollution Assessment and Recommender Model, Housing Health and Safety Analysis Model, Advisory Model, Augmented Reality Model and Healthy Homes Living Spaces Certification Model.

Analysis of significant factors for dengue infection prognosis using the Random Forest Classifier

Random forests have emerged as a versatile and highly accurate classification and regression methodology, requiring little tuning and providing interpretable outputs. Here, we briefly explore the possibility of applying this ensemble supervised machine learning technique to predict the vulnerability for complex disease - Dengue which is often baffled with chikungunya viral fever. This study presents a new-fangled approach to determine the significant prognosis factors in dengue patients. Random forests is used to visualize and determine the significant factors that can differentiate between the dengue patients and the healthy subjects and for constructing a dengue disease survivability prediction model during the boosting process to improve accuracy and stability and to reduce over fitting problems. The presented methodology may be incorporated in a variety of applications such as risk management, tailored health communication and decision support systems in healthcare

ETools for the standard classification of health and social services: eDESDE-LTC

Background: The DESDE-LTC is a standardized international classification of services that allows mapping of social and health care based on the European Service Mapping Schedule (ESMS) developed for mental health services. This system provides a standard description of the whole array of services available for a target population both at horizontal and at vertical level allowing for the basic information required to assess integrated care. It is based in coding the “minimum units of production of care” within services called Basic Inputs of Care (BSICs). These units are labeled using the most meaningful activity or “Main Type of Care” (MTC). DESDE-LTC has been extensively used in different sectors such as mental health, disability, elderly and long term care, performed in over 17 countries both in Europe and America. However this system requires extensive training and periodical update of the information gathered. Innovation: In order to overcome this problem several eTools have been developed in the last 4 years. eDESDE-LTC (www.edesdeproject.eu/) is an electronic knowledge base that contains all the information required to code and use the classification system in long term care. It includes the reference documents (conceptual framework, usability, dissemination and reports), a toolkit (coding system, glossary and assessment instrument), and an on-line training package. The use of this knowledge base proved useful as complementary to but not as a replacement of face-to-face World Congress on Integrated Care 2014, Sydney, November 23-26, 2014. International Journal of Integrated Care – Volume 14, 8 December – URN:NBN:NL:UI:10-1-116661– http://www.ijic.org/ training is the DESDE-LTC project (Salvador-Carulla et al, 2011) and in the subsequent REFINEMENT project (Amaddeo et al, 2013). The Refinement Toolkit (http://www.refinementproject.eu/) is a decision support system for Mental Health that includes four major sets of tools: FINCENTO (analysis of financing), REQUALIT (quality of mental health care), REPATO (Pathways of care) and REMAST (service provision). REMAST includes the Verona Socio-demographic Scale (SES) and DESDE-LTC apart from other service information and Atlases of care in several European regions. The REFINEMENT toolkit provides relevant information for system analysis in mental health and for a better understanding of service provision but the need of training has not been modified from the previous knowledge base. In 2014 the Regional Government of Andalucia funded the project DESDE-AND to develop a computer algorithm for automatic collection of data and coding of social services, following an integrated care approach. This research has allowed to test the usability of the coding system outside the health sector and to improve Conclusions: The DESDE-LTC has proven its feasibility for the standard assessment of services in different fields, including mental health, disabilities, ageing, long-term care and social services. This indicates the generalisability of this taxonomy for mapping integrated care. The development of different electronic knowledge-bases and computer algorithms for data completion will be a key factor for the dissemination and spreading of this system as well as its incorporation to Geographical Information Systems.