Medical Decision Support Research Articles

Improved Metaheuristics with Machine Learning Enabled Medical Decision Support System

Smart healthcare has become a hot research topic due to the contemporary developments of Internet of Things (IoT), sensor technologies, cloud computing, and others. Besides, the latest advances of Artificial Intelligence (AI) tools find helpful for decision-making in innovative healthcare to diagnose several diseases. Ovarian Cancer (OC) is a kind of cancer that affects women’s ovaries, and it is tedious to identify OC at the primary stages with a high mortality rate. The OC data produced by the Internet of Medical Things (IoMT) devices can be utilized to differentiate OC. In this aspect, this paper introduces a new quantum black widow optimization with a machine learning-enabled decision support system (QBWO-MLDSS) for smart healthcare. The primary intention of the QBWO-MLDSS technique is to detect and categorize the OC rapidly and accurately. Besides, the QBWO-MLDSS model involves a Z-score normalization approach to pre-process the data. In addition, the QBWO-MLDSS technique derives a QBWO algorithm as a feature selection to derive optimum feature subsets. Moreover, symbiotic organisms search (SOS) with extreme learning machine (ELM) model is applied as a classifier for the detection and classification of ELM model, thereby improving the overall classification performance. The design of QBWO and SOS for OC detection and classification in the smart healthcare environment shows the study’s novelty. The experimental result analysis of the QBWO-MLDSS model is conducted using a benchmark dataset, and the comparative results reported the enhanced outcomes of the QBWO-MLDSS model over the recent approaches.

A New Medical Decision Support System for Diagnosing HFrEF and HFpEF Using ECG and Machine Learning Techniques

As heart failure (HF) is a growing epidemic, no case should be overlooked in the diagnosis of HF. Two subtypes by left ventricular ejection fraction (LVEF) of HF are HF with reduced ejection fraction (HFrEF) (LVEF ≤ 40%) and HF with preserved ejection fraction (HFpEF) (LVEF ≥ 50%). HFrEF is easier to diagnose. However, the diagnosis of HFpEF is more complex and difficult even for specialists. The diagnosis of HFpEF is a problem that is being tried to be solved in medicine. Since LVEF appears normal (LVEF ≥ 50% in healthy individuals), HFpEF can be confused with chest diseases due to some similar symptoms. The diagnosis of HF subtypes is ideally made using echocardiography. Echocardiography should be performed in all patients with HF; however, it is expensive and requires specialists. Even in high-resource regions, this test is not always performed, and treatment may need to be initiated before the echocardiographic data are obtained. For such situations, economical and practical systems are required. In this study, a medical decision support system was developed to detect HFrEF and HFpEF cases using only 3-lead ECG. From the ECG data of 61 volunteers, 37 features were extracted, of which 16 were Yule-Walker and Burg’s method parameters, and 21 were in the time domain. Consequently, 37 features were reduced by feature selection and triple classification was performed with only 4 features with maximum accuracy. The aim of this study was to determine whether the individuals with HF symptoms were HFrEF, HFpEF, or healthy. Four machine learning algorithms were used for classification. The best classification accuracy rate was 100% for k-NN, and significant results were obtained from the other three algorithms: SVMs, Decision Trees, and Ensemble Bagged Trees.

BMRMIA: A Platform for Radiologists to Systematically Learn Automated Medical Image Analysis by Three Dimensional Medical Decision Support System

Contribution: This paper designs a learning and training platform that can systematically help radiologists learn automated medical image analysis technology. The platform can help radiologists master deep learning theories and medical applications such as the three-dimensional medical decision support system, and strengthen the teaching practice of deep learning related courses in hospitals, so as to help doctors better understand deep learning knowledge and improve the efficiency of auxiliary diagnosis. Background: In recent years, deep learning has been widely used in academia, industry, and medicine. An increasing number of companies are starting to recruit a large number of professionals in the field of deep learning. Increasing numbers of colleges and universities also offer courses related to deep learning to help radiologists learn automated medical image analysis techniques. For now, however, there is no practical training platform that can help radiologists learn automated medical image analysis systematically. Application Design: The platform proposes the basic learning, model combat, business application (BMR) concept, including the learning guidance system and the assessment training system, which constitutes a closed-loop learning guidance mode of “learning-assessment-training-learning”. Findings: The survey results show that most of radiologists met their learning expectations by using this platform. The platform can help radiologists master deep learning techniques quickly, comprehensively and firmly.

Developing an Explainable Machine Learning-Based Thyroid Disease Prediction Model

Healthcare and medicine are key areas where machine learning algorithms are widely used. The medical decision support systems thus created are accurate enough, however, they suffer from the lack of transparency in decision making and shows a black box behavior. However, transparency and trust are significant in the field of health and medicine and hence, a black box system is sub optimal in terms of widespread applicability and reach. Hence, the explainablility of the research make the system reliable and understandable, thereby enhancing its social acceptability. The presented work explores a thyroid disease diagnosis system. SHAP, a popular method based on coalition game theory is used for interpretability of results. The work explains the system behavior both locally and globally and shows how machine leaning can be used to ascertain the causality of the disease and support doctors to suggest the most effective treatment of the disease. The work not only demonstrates the results of machine learning algorithms but also explains related feature importance and model insights.

An Interdisciplinary Collaboration to Improve Hypertension Medication Adherence for Older Adults

Hypertension is highly prevalent in older adults (74.5% in ≥60 years) with dire consequences, and adherence to hypertension medications is low (approximately 50%). With increased smartphone use among older adults (81% for 60-69 years, 62% for ≥70 years), technology innovations can improve medication adherence. This symposium highlights the efforts of an innovative interdisciplinary team of experts (clinical, cognitive aging, human factors, health technology) to develop and implement the Medication Education, Decision Support, Reminding, and Monitoring (MEDSReM) system to improve hypertension medication adherence for older adults. MEDSReM is a theory-based, integrated mobile application (app) and companion web portal that educates, supports missed dose decisions, reminds, monitors adherence, and incorporates blood pressure feedback. In this symposium, we describe the interdisciplinary development efforts. Insel et al. will present the theory-based intervention, technology translation, and advancement of the MEDSReM system. Lee et al. will describe the interdisciplinary team and describe the work by the decision support subteam that created the medication formulary and generated an algorithm to guide missed-dose decisions based on pharmacology of aging. Rogers et al. will discuss the education subteam’s development of educational information about hypertension, medications, and adherence for the MEDSReM system. Mitzner et al. will illustrate the instructional support sub-team’s efforts to ensure older adults can interact with both the smartphone app and online portal. Lastly, Hale et al. will describe the user testing subteam’s usability processes including the integration of blood pressure self-monitoring. These efforts will provide insights for other interdisciplinary teams developing technology interventions for older adults.

Accuracy versus reliability-based modelling approaches for medical decision making

Forecasting in the medical domain is critical to the quality of decisions made by physicians, patients, and health planners. Modeling is one of the most important components of decision support systems, which are frequently used to simulate and analyze under-studied systems in order to make more appropriate decisions in medical science. In the medical modeling literature, various approaches with varying structures and characteristics have been proposed to cover a wide range of application categories and domains. Regardless of the differences between modeling approaches, all of them aim to maximize the accuracy or reliability of the results in order to achieve the most generalizable model and, as a result, a higher level of profitability decisions. Despite the theoretical significance and practical impact of reliability on generalizability, particularly in high-risk decisions and applications, a significant number of models in the fields of medical forecasting, classification, and time series prediction have been developed to maximize accuracy in mind. In other words, given the volatility of medical variables, it is also necessary to have stable and reliable forecasts in order to make sound decisions. The quality of medical decisions resulting from accuracy and reliability-based intelligent and statistical modeling approaches is compared and evaluated in this paper in order to determine the relative importance of accuracy and reliability on the quality of made decisions in decision support systems. For this purpose, 33 different case studies from the UCI in three categories of supervised modeling, namely causal forecasting, time series prediction, and classification, were considered. These cases were chosen from various domains, such as disease diagnosis (obesity, Parkinson's disease, diabetes, hepatitis, stenosis of arteries, orthopedic disease, autism) and cancer (lung, breast, cervical), experiments, therapy (immunotherapy, cryotherapy), fertility prediction, and predicting the number of patients in the emergency room and ICU. According to empirical findings, the reliability-based strategy outperformed the accuracy-based strategy in causal forecasting cases by 2.26%, classification cases by 13.49%, and time series prediction cases by 3.08%. Furthermore, compared to similar accuracy-based models, the reliability-based models can generate a 6.28% improvement. As a result, they can be considered an appropriate alternative to traditional accuracy-based models for medical decision support systems modeling purposes.

Standardization in regulating artificial intelligence systems in Russian healthcare

Artificial intelligence technologies in medical practice are a promising direction in the world. Artificial intelligence medical decision support systems, diagnostic and screening programs can help medical personnel in routine and complex tasks and improve the level of medical care provided to patients. At the same time, the development, production and distribution of artificial intelligence systems must be regulated without fail. Registration and subsequent control (post-registration monitoring) of artificial intelligence systems in medicine require the creation, adjustment of the legal framework and technological regulation. The Russian Federation has developed a promising development strategy in this area. Seven national standards have been developed by experts in the field of Artificial intelligence in healthcare. These standards establish the procedures for conducting clinical and technical trials, performance requirements and the concept of life cycle, a quality management system and risk management. Aseparate standards is devoted to dataset creation for training and testing the developed algorithms, requirements for them and a metadata format. There are plans to bring the developed national standards to the international level, which will allow Russian manufacturers of artificial intelligence systems implemented these national standards to comply with foreign counterparts and become more competitive at the international level. The international community has already supported the development of an ISO standard based on the national standard for clinical trials. The development will be performed based on the technical committee ISO/TC215 (Health informatics) in conjunction with ISO/IEC JTC1/SC42 (Artificial intelligence), this will allow bringing the national requirements for the Artificial intelligence to the international level. The cycle of these standards will summarize recognized methodologies, helping both manufacturers and medical organizations, doctors and patients to produce and use aquality, safe and effective product.

Computer-Aided Medical Diagnosis System Using Logistics Regression Algorithms (LRA) Supervised Learning Approach

This work focused on the designing of medical diagnosis system using Supervised Machine Learning. Logistics Regression Algorithms (LRA) was adopted, the label inputs for the data set which the symptoms were trained and mapped with the input of the user. Diagnosis of malaria was considered in this work; the system verified the value of the logical regression in the medical decision support system. Medical practitioners and other health workers can use this system to make better decisions in medical diagnosis for malaria. Adoption of this system will reduce stress of diagnoses malaria from patient and reduce congestion in our hospitals.

СВЕРТОЧНЫЕ НЕЙРОННЫЕ СЕТИ В ДИАГНОСТИКЕ НОВООБРАЗОВАНИЙ КОЖИ

The problem of using artificial intelligence technologies in the diagnosis of skin neoplasms is considered. Dermatoscopic images for 8 nosologies were considered as the object of research. The melanoma was one of them. Melanoma is responsible for the most deaths of all skin cancers. The aim of the study was to evaluate the effectiveness of the use of pre-trained convolutional neural networks for the classification of skin neoplasms. A classification algorithm for an ensemble of convolutional networks was proposed. Pre-trained neural networks have been studied to form the ensemble. Neural network samples were selected from a set of neural networks that have proven themselves in the ImageNet Large Scale Visual Recognition Challenge. According to the results of the experiment the best three of the eight convolutional neural networks were selected for inclusion in the ensemble – MobileNet_v2, ResNet_152, ResNeXt_101_32x8d. The experiment was conducted on a sample of 10015 images representing 8 nosologies. The average classification accuracy for all nosologies was 79%. The paper highlights the features of ensuring information security when using telemedicine diagnostic technologies using the proposed approach in the recognition of images of skin neoplasms. The results of the work can be used in the design of medical decision support systems for the diagnosis of malignant skin neoplasms (including melanoma).

Information system for monitoring of hypopituitarism in childhood and adolescence

The article is devoted to the development of a medical information system (MIS) designed to monitor and predict hypopituitarism in children and teenagers. MIS combines the medical information storage and management system with an intellectual system of medical decision-making support. The information system is based on a database with information obtained from impersonal medical discharge papers of children and adolescents of the Altai Territory diagnosed with hypopituitarism. Medical discharge papers provide information about the patient’s health status while in the hospital, the results of examinations and treatment. Discharges form a complete medical history of the patient, and the information system allows assessing the disease in dynamics. The purpose of the intelligent information system is to predict the patient's growth in order to select the optimal treatment strategy. The intelligent decision support system is based on a machine-learning model trained on data from a database. The model was trained in the Python programming language. The MIS interface is developed using the C # programming language. The use of MIS in medical institutions will allow doctors to carry out a personalized approach to the monitoring and predicting the growth for patients, and to choose the optimal trajectory of treatment.

Personalized content recommendation in online health communities

PurposeRecommending suitable content for users of online health communities (OHCs) is critical for overcoming information overload problem and facilitate medical decision making, but remains not fully investigated. This study aims to provide a content recommendation approach to automatically match valuable health-related information for OHC members.Design/methodology/approachA framework of health-related content recommendation was proposed by leveraging rich social information in online communities. The authors constructed user influence relationship (UIR) utilizing users' interaction records, user profiles and user-generated content. The initial user rating matrix and the user post matching matrix were then created by analyzing text content of posts. Finally, the user rating matrix and the recommended content were generated for community members. Datasets were collected from an OHC to evaluate the effectiveness of the proposed approach.FindingsThe experimental results revealed that the proposed method statistically outperformed baseline models in content recommendation for users of OHCs.Research limitations/implicationsThe incorporation of social information can significantly enhance the performance of content recommendation in OHCs. The user post matching degree based on text analysis can improve the effectiveness of recommendation.Practical implicationsThis study potentially contributes to the social support exchange and medical decision making of community members and the sustainable prosperity of OHCs.Originality/valueThis study proposes a novel social content recommendation method for online health consumers based on UIRs by leveraging social information in OHCs. The results indicate the significance of social information in content recommendation of healthcare social media.

Information technologies’ role in optimizing the application of drugs in clinical practice: a view of a clinical pharmacologist

Currently, the development and implementation of computerized medical decision support systems (CMDSS) is an effective tool aimed at optimizing drug therapy in clinical practice. It has been proven in clinical studies their use can increase the efficacy and safety of pharmacotherapy for a number of socially significant diseases. An active integration of CMDSS into medical information systems of medical organizations is required.

Interpretable SincNet-based Deep Learning for Emotion Recognition from EEG brain activity.

Machine learning methods, such as deep learning, show promising results in the medical domain. However, the lack of interpretability of these algorithms may hinder their applicability to medical decision support systems. This paper studies an interpretable deep learning technique, called SincNet. SincNet is a convolutional neural network that efficiently learns customized band-pass filters through trainable sinc-functions. In this study, we use SincNet to analyze the neural activity of individuals with Autism Spectrum Disorder (ASD), who experience characteristic differences in neural oscillatory activity. In particular, we propose a novel SincNet-based neural network for detecting emotions in ASD patients using EEG signals. The learned filters can be easily inspected to detect which part of the EEG spectrum is used for predicting emotions. We found that our system automatically learns the high-α (9-13 Hz) and β (13-30 Hz) band suppression often present in individuals with ASD. This result is consistent with recent neuroscience studies on emotion recognition, which found an association between these band suppressions and the behavioral deficits observed in individuals with ASD. The improved interpretability of SincNet is achieved without sacrificing performance in emotion recognition.

Applied aspects of data mining for decision support at the regional health system

The article presents the results of research of Data Mining methods with Microsoft SQL Server. Microsoft Clustering algorithm was used for improving the effectiveness of medical prevention and treatment in a cohort of patients with arterial hypertension. There are rationales for monitoring of cardiovascular risk and desire to correct the risk with Data Mining at medical decision support systems. Authors used medical and sociological monitoring data from regional clinical hospital. The segmentation of arterial hypertension patients was performed using Microsoft Clustering algorithm. As a result, a quantitative assessment of the population profile for patients with arterial hypertension was obtained. The authors presented diagrams and profiles of clusters. They were compared. The developed approach is applied for decision support at regional health information management system for reduce of cardiovascular risk.

Control model of parallel functioning production modules as fuzzy Petri nets

The article presents the results of research of Data Mining methods with Microsoft SQL Server. Microsoft Clustering algorithm was used for improving the effectiveness of medical prevention and treatment in a cohort of patients with arterial hypertension. There are rationales for monitoring of cardiovascular risk and desire to correct the risk with Data Mining at medical decision support systems. Authors used medical and sociological monitoring data from regional clinical hospital. The segmentation of arterial hypertension patients was performed using Microsoft Clustering algorithm. As a result, a quantitative assessment of the population profile for patients with arterial hypertension was obtained. The authors presented diagrams and profiles of clusters. They were compared. The developed approach is applied for decision support at regional health information management system for reduce of cardiovascular risk.

Medical Specialists' Perspectives on the Influence of Electronic Medical Record Use on the Quality of Hospital Care: Semistructured Interview Study.

BackgroundNumerous publications show that electronic medical records (EMRs) may make an important contribution to increasing the quality of care. There are indications that particularly the medical specialist plays an important role in the use of EMRs in hospitals.ObjectiveThe aim of this study was to examine how, and by which aspects, the relationship between EMR use and the quality of care in hospitals is influenced according to medical specialists.MethodsTo answer this question, a qualitative study was conducted in the period of August-October 2018. Semistructured interviews of around 90 min were conducted with 11 medical specialists from 11 different Dutch hospitals. For analysis of the answers, we used a previously published taxonomy of factors that can influence the use of EMRs.ResultsThe professional experience of the participating medical specialists varied between 5 and 27 years. Using the previously published taxonomy, these medical specialists considered technical barriers the most significant for EMR use. The suboptimal change processes surrounding implementation were also perceived as a major barrier. A final major problem is related to the categories “social” (their relationships with the patients and fellow care providers), “psychological” (based on their personal issues, knowledge, and perceptions), and “time” (the time required to select, implement, and learn how to use EMR systems and subsequently enter data into the system). However, the medical specialists also identified potential technical facilitators, particularly in the assured availability of information to all health care professionals involved in the care of a patient. They see promise in using EMRs for medical decision support to improve the quality of care but consider these possibilities currently lacking.ConclusionsThe 11 medical specialists shared positive experiences with EMR use when comparing it to formerly used paper records. The fact that involved health care professionals can access patient data at any time they need is considered important. However, in practice, potential quality improvement lags as long as decision support cannot be applied because of the lack of a fully coded patient record.

인공지능시대의 경혈 주치 연구를 위한 제언

Artificial intelligence technology sheds light on new ways of innovating acupuncture research. As acupoint selection is specific to target diseases, each acupoint is generally believed to have a specific indication. However, the specificity of acupoint selection may be not always same with the specificity of acupoint indication. In this review, we propose that the specificity of acupoint indication can be inferred from clinical data using reverse inference. Using forward inference, the prescribed acupoints for each disease can be quantified for the specificity of acupoint selection. Using reverse inference, targeted diseases for each acupoint can be quantified for the specificity of acupoint indication. It is noteworthy that the selection of an acupoint for a particular disease does not imply the acupoint has specific indications for that disease. Electronic medical record includes various symptoms and chosen acupoint combinations. Data mining approach can be useful to reveal the complex relationships between diseases and acupoints from clinical data. Combining the clinical information and the bodily sensation map, the spatial patterns of acupoint indication can be further estimated. Interoperable medical data should be collected for medical knowledge discovery and clinical decision support system. In the era of artificial intelligence, machine learning can reveal the associations between diseases and prescribed acupoints from large scale clinical data warehouse.

Impact of a CE-marked medical software sensor on SARS-CoV-2 disease pandemic progression estimation

Abstract Background To address the current and any future pandemic, we need a robust, real-time and population-scale collection of data. Rapid and comprehensive knowledge on the trends in reported symptoms in populations provides an earlier window into our progress against the viral spread, and helps predict the need and timing of professional healthcare. We thus used a CE-marked medical online symptom checker service and validated the data against national demand of care to estimate the progression of the pandemic in Finland. Methods Our data comprised real-time ©Omaolo Covid-19 symptom checker responses (414 477 in total) and daily admission counts in nationwide inpatient and outpatient registers provided by the Finnish Institute for Health and Welfare 16th March-15th June, 2020 (the first wave of the pandemic in Finland). The questionnaires responses provide self-triage information input to a medically qualified algorithm that produces a personalized probability of having the Covid-19 and provides graded recommendations for further actions. We trained linear regression and XGBoost models together with F-Score and mutual information feature pre-selectors to predict the admissions 1 week ahead. Results Our models reached a MAPE (Mean Absolute Percentage Error) between 24.2% and 36.4% and MdAPE (Median Average Percentage Error) between 20.1% and 25.1% in predicting national daily patient admissions. The best result was achieved by combining both ©Omaolo and historical patient admission counts. Our best predictor was linear regression with mutual information as the pre-selector with slight advantage over XGBoost. Conclusions Accurate short-term predictions of Covid-19 patient admissions can be made, and both the symptom check questionnaires and the daily admissions data contribute to the accuracy of the predictions. Thus, symptoms checkers can be used to estimate the progression of the pandemic, which can be considered when predicting the health care burden in a future pandemic. Key messages Digital surveillance of symptoms enables rapid and contactless tracing of epidemic progression. Prediction models can help support medical decision making in an epidemic to promote and protect public health and prevent health care system overload.

Development of Segmentation and Classification Algorithms for Computed Tomography Images of Human Kidney Stone

Computed tomography (CT) scan diagnostics procedures adopt the use of image information retrieval system with the help of radiographer’s expertise. However, this technique is prone to errors. Significant height of accuracy is required in healthcare decision support, as 20% of CT scans are associated with error. The application of artificial intelligence (AI) can improve performance level, mitigate human error, and enhance clinical decision support in the context of time and accuracy. The study introduced machine learning algorithm to analyze stream of anonymous CT scans of kidney. The research adopted deep learning approach for segmentation and classification of kidney stone (renal calculi) images in Python (with Keras and TensorFlow) environment. A control volume of data along with 336 kidney stone images were used to train the deep learning network with 10 testing images. The training images were divided into two sets (folders) as follows; one was labeled as STONE (containing 167 images) and the other as NO-STONE (containing 169 images); 10 iterations were performed for model training. The network layers were structured as input layer in the following with 2-D convolutional neural network machine learning (CNN-ML), ReLU activation, Maxpooling, and fully connected (dense) layer including the sigmoid activation layer. The training adopted a batch size of 8 with 10% validation. The output result, upon testing the model, has an accuracy of 90%, sensitivity value of 80% and effectiveness of 89%. The segmentation and classification algorithm model could be embedded in future CT diagnostic procedure to enhance medical decision support and accuracy.

Utilization of Chest Pain Decision Aids in a Community Hospital Emergency Department: A Mixed-methods Implementation Study.

Chest pain is a common reason for emergency department (ED) visits. Evidence-based decision aids assessing risk for an adverse cardiac event are underused in community hospital emergency care. This study explored the acceptability, barriers, facilitators, and potential strategies for implementation of the HEART Score risk stratification tool, accelerated diagnostic pathway, and shared decision-making visual aid with physicians and chest pain patients ages >45 in a community hospital ED. Single center, mixed-methods study. (1) Physician semistructured interviews using The Consolidated Framework for Implementation Research for systematic analysis. (2) Patient and physician surveys. (3) 16-week intervention of physician training and pilot testing of decision aids with ED patients. Physician interviews (n = 19); key facilitators: electronic medical record decision support, ease of use, risk stratification and disposition support, and shared decision-making training. Key barriers: time constraints, patient ability, and/or willingness to participate in shared decision-making, lack of integration with medical record and change in practice workflow. Patient study participants (n = 184) with a survey response rate of 92% (n = 170). Most patients (85%) were satisfied with the shared decision-making visual aid. Physicians surveyed (n = 84) with a response rate of 50% (n = 42). Most physicians, 95% (n = 40), support use of the HEART Score, with limited acceptance of the shared decision-making visual aid of 57% (n = 24). Using evidence-based chest pain decision aids in a community hospital ED is feasible and acceptable. Key barriers and facilitators for implementation were identified. Further research in community hospitals is needed to verify findings, examine generalizability, and test implementation strategies.