Medical Diagnostic Decision Support System Research Articles

Agent-oriented data clustering for medical monitoring

Medical data processing is one of the priority machine learning areas. Usually, data obtained in the process of medical patient monitoring are complex and have a different nature. Solving the problem of clustering, classification, or forecasting problem these data requires the creation of new methods or improvement of existing methods to improve the decision accuracy and effectiveness. The classical clustering approaches and the c-means fuzzy clustering method were analyzed. Based on the multiagent systems theory, it is proposed to use in the c-means method the separate rules for selecting elites when forming clusters and selecting the best of them in accordance with the chosen intra-cluster distance measures. The result of solving such a problem is the number of clusters, as well as the number of elements in them. The method quality was tested on Fisher iris data set using three measures of intra-cluster distance: Mahalanobis distance, Mahalanobis distance considering the membership function, and Kullbak-Leibler entropy. The highest accuracy of 98% was obtained for the distance measured by the Kullbak-Leibler entropy. Therefore, this measure was chosen to solve the clustering problem of medical monitoring data for prostate disease. Medical monitoring data were divided into four classes of patient states: “healthy persons”, “non-metastatic patients”, “metastatic patients” and “hormone-resistant patients”. The accuracy of clustering according to medical data was 95,6%. In addition to accuracy, the confusion matrix, ROC- and LF-curves were used to assess the method quality. The minimum value of the ROC-curve was 0.96 for Fisher's irises and 0.95 for medical monitoring data, which characterizes the high quality of the proposed clustering method. The loss function value is also quite small (-0.056 and -0.0176 for each considered data set), which means that the optimal cluster number and the distribution of data over them are obtained. Based on the obtained results analysis, the proposed method can be recommended for use in medical information and diagnostic decision support systems for clustering monitoring data.

Medical Diagnostic Decision Support Systems Based on Artificial Intelligence Algorithms

The idea of using AI algorithms for diagnosing and detecting pathologies in the human body is currently very popular. The reason is that the artificial intelligence technologies available to researchers make it possible to analyze information almost intuitively, to detect similarities between non-obvious patterns, and find an effective solution through training on the existing research and practical experience of specialists. As a result, diagnostic decision support systems based on artificial intelligence algorithms are increasingly being used in practice. This review examines the main artificial intelligence algorithms used in modern diagnostic decision support systems and illustrates their application for the diagnosis of several key socially significant diseases. Key issues related to the application of artificial intelligence systems in disease diagnosis are also discussed. Keywords—clinical decision support systems, cognitive computing, deep learning, diagnostic decision support systems, machine learning.

Virtual Care Technologies, Wearable Health Monitoring Sensors, and Internet of Medical Things-based Smart Disease Surveillance Systems in the Diagnosis and Treatment of COVID-19 Patients

Digital epidemiological surveillance in monitoring, detection, and prevention of COVID-19 is optimized by use of medical artificial intelligence, clinical and diagnostic decision support systems, machine learning-based real-time data sensing and processing, and smart healthcare devices and applications. Study Design, Survey Methods, and Materials The interviews were conducted online and data were weighted by five variables (age, race/ethnicity, gender, education, and geographic region) using the Census Bureau's American Community Survey to reflect reliably and accurately the demographic composition of the United States. (Pustokhina et al., 2020) Body sensor networks integrate interconnected bio-sensors and wearable healthcare devices (Kovacova and Lazaroiu, 2021;Lyons and Lazaroiu, 2020) that assess abnormal alterations in vital physiological signs and share medical imaging data for patient diagnosis and monitoring, being instrumental in chronic diseases by use of deep learning-based applications. Conclusions, Implications, Limitations, and Further Research Directions Artificial intelligence-enabled wearable medical devices, virtualized care systems, and wireless biomedical sensing devices are pivotal in COVID-19 screening, testing, and treatment.

Artificial Intelligence-enabled Wearable Medical Devices, Clinical and Diagnostic Decision Support Systems, and Internet of Things-based Healthcare Applications in COVID-19 Prevention, Screening, and Treatment

Building our argument by drawing on data collected from Accenture, GlobalWebIndex, GoMo Health, KPMG, McKinsey, Oracle, Sermo, STAT, Statista, and Workplace Intelligence, we performed analyses and made estimates regarding how predictive big data analytics, body sensor networks, medical wearable devices, decision support systems, and wireless sensing applications can be harnessed in real-time continuous remote monitoring of patients vital signs configuring clinical data in pervasive mobile patient-centric healthcare. Introduction The extensive data of COVID-19 patients can be assimilated and inspected by cutting-edge machine learning algorithms to grasp the pattern of viral transmission, optimize diagnostic swiftness and precision, advance adequate therapeutic methods, and identify the most vulnerable individuals according to personalized genetic and physiological features. Methodology and Empirical Analysis Building our argument by drawing on data collected from Accenture, Global-WebIndex, GoMo Health, KPMG, McKinsey, Oracle, Sermo, STAT, Statista, and Workplace Intelligence, we performed analyses and made estimates regarding how predictive big data analytics, body sensor networks, medical wearable devices, decision support systems, and wireless sensing applications can be harnessed in real-time continuous remote monitoring of patients' vital signs configuring clinical data in pervasive mobile patient-centric healthcare. Study Design, Survey Methods, and Materials The interviews were conducted online and data were weighted by five variables (age, race/ethnicity, gender, education, and geographic region) using the Census Bureau's American Community Survey to reflect reliably and accurately the demographic composition of the United States.

Evolutionary computing for clinical dataset classification using a novel feature selection algorithm

The medical diagnostic decision support system uses machine learning and data mining algorithms to detect and diagnose diseases. Several deaths can be avoided if the diseases are detected and cured in the early stages of infection. Feature selection is a major pre-processing method used to obtain the most significant features, thereby enhancing the data mining model's classification accuracy. This work proposes a new feature selection algorithm to perform feature selection as a multi-objective optimization problem. The minimization of classification error rate and minimization of the feature subset's cardinality are the two contradictory objectives that need to be optimized simultaneously. The proposed work is applied for five clinical datasets such as lung cancer, breast cancer, diabetes, fertility, and immunotherapy and the results are compared with existing techniques based on 6 other datasets. This work converts the real-valued Jaya Optimization Algorithm into binary space. It also handles premature convergence and sensitivity–specificity trade-off. The proposed algorithm's efficiency is assessed and analyzed based on average classification accuracy, sensitivity, specificity, number of features selected, percentage feature selection, and CPU computation time. The proposed algorithm improves the effectiveness of data mining based medical diagnostic decision support system.

Development of medical diagnostic decision support systems and their economic efficiency

The object of research is the diagnostic decision support system (DSS). One of the most problematic areas in medical diagnostic systems is the formation of a knowledge base based on expert rules, which provides a recommendation for the disease. The methods of designing medical diagnostic systems have been studied. Methods for applying the potential of artificial intelligence in medicine in the form of fuzzy rules or conducting diagnostics on the basis of Bayesian networks are considered. Intellectual computing tools in the form of expert systems based on rules and fuzzy logic, applied to neural networks and genetic algorithms performed in medical diagnostics are considered.To develop a decision support system for a pediatrician, a method to build a knowledge base on the basis of logical rules «If ..., then ...» was chosen. Using this method allows to create initial conditions for input data in the system, and speed up their processing in the knowledge base. Although the knowledge base is quite cumbersome, this does not reduce the performance of the system.In the process of research, the development of a medical diagnostic system for decision support by a pediatrician for the design stages is described. The application of this system allows to automate the process of document circulation for a pediatrician and to speed up the stage of preliminary assessment of the patient's condition.The built-in pediatrician electronic pediatric module not only automates the workflow process, reduces the doctor's work time with papers, but also allows to obtain complete information about the patient.The calculation of economic efficiency from the DSS introduction by a pediatrician is performed. The system cost is to be recouped within 1 year.The prospect of adding modules to the system for individual diseases and forming an electronic record from the moment of birth with the prospect of transferring data to the system for adults are advantages over analogues of this software product.

On the Algorithm of the Medical Diagnostic Decision Support System under the Mobile Platform

A mobile platform usually refers to a complete platform that provides a reference design, a hardware chipset and upper communication protocols and supporting software and development tools. The downstream vendors can design and manufacture appropriate mobile products in a shorter time as required. The app developers mustn’t pay attention to the content of platform hardware, but systematically understand the types of mobile devices, supporting software and appropriate development tools in mobile platform. The key for the medical experts to deal with the patients is decisionmaking. They make the decisions in diagnosing the patient’s status, adjusting the therapeutic plan and monitoring the changing disease status. In recent decades, a series of methods and tools have been developed to assist the clinicians for decision. Typically, these tools and methods are not intended to substitute the human’s job, but used for assistant and supportive purpose. The doctors, healthcare professionals and other healthcare staffs have been being aware of the importance of these algorithms and techniques in order to support the decision-making process as the information and communication technology has become increasingly important for the infrastructure construction in healthcare organizations.

An improved hybrid feature reduction for increased breast cancer diagnostic performance

Purpose Early and correct diagnosis of a disease is vital for the success of treatment. Medical diagnostic decision support system can be used to improve the accuracy of the traditional diagnosis. As such, various pattern recognition methods are studied and applied to develop medical diagnostic decision support system. In this study, the effects of dimensionality reduction techniques with probabilistic neural network (PNN) on breast cancer classification are examined.

A Refined Continuous Ant Colony Optimization Based FP-Growth Association Rule Technique on Type 2 Diabetes

In recent years, Diabetes mellitus almost tops the list of chronic diseases worldwide among the major public health challenges. Diagnosing diabetes at the preliminary stage is undoubtedly challenging as it involves varying complexities and inter relation and dependence on several factors that affect it directly or indirectly. Due to large number of diabetic patients in recent years, desperate measures have to be devised and developed to facilitate medical diagnostic decision support systems that help doctors, researchers and medical practitioners during and after the process of diagnosing Diabetes. In this paper, the Association Rule Mining and Enhanced FP-Growth Algorithm has been used as a reference to propose a new algorithm that is basically has similar functionalities as that of the Ant Colony Optimization algorithm and one that improvises association rule mining results. The Refined Continuous Ant Colony Optimization or CACO deploys a meta-heuristic approach and has been devised and enthused by actual ant colonies behavior along with the sustained continuous domains. Preliminarily association rules so produced by the Enhanced FP-Growth algorithm are deployed thereafter which rules from weakest set are found on the basis of threshold value and then further used by the Ant Colony algorithm so that association rules are reduced and a better quality of rules are discovered as a result of the efforts. The study as well the research presented here aims at reducing database scanning by optimizing as well by improving the quality of rules that are produced for CACO.

Using experts feedback in clinical case resolution and arbitration as accuracy diagnosis methodology

This paper proposes a new methodology for assessing the efficiency of medical diagnostic systems and clinical decision support systems by using the feedback/opinions of medical experts. The methodology behind this work is based on a comparison between the expert feedback that has helped solve different clinical cases and the expert system that has evaluated these same cases. Once the results are returned, an arbitration process is carried out in order to ensure the correctness of the results provided by both methods. Once this process has been completed, the results are analyzed using Precision, Recall, Accuracy, Specificity and Matthews Correlation Coefficient (MCC) (PRAS-M) metrics. When the methodology is applied, the results obtained from a real diagnostic system allow researchers to establish the accuracy of the system based on objective facts. The methodology returns enough information to analyze the system's behavior for each disease in the knowledge base or across the entire knowledge base. It also returns data on the efficiency of the different assessors involved in the evaluation process, analyzing their behavior in the diagnostic process. The proposed work facilitates the evaluation of medical diagnostic systems, having a reliable process based on objective facts. The methodology presented in this research makes it possible to identify the main characteristics that define a medical diagnostic system and their values, allowing for system improvement. A good example of the results provided by the application of the methodology is shown in this paper. A diagnosis system was evaluated by means of this methodology, yielding positive results (statistically significant) when comparing the system with the assessors that participated in the evaluation process of the system through metrics such as recall (+27.54%) and MCC (+32.19%). These results demonstrate the real applicability of the methodology used.

Impact of a Computer-Based Diagnostic Decision Support Tool on the Differential Diagnoses of Medicine Residents

Computer-based medical diagnostic decision support systems have been used for decades, initially as stand-alone applications. More recent versions have been tested for their effectiveness in enhancing the diagnostic ability of clinicians. To determine if viewing a rank-ordered list of diagnostic possibilities from a medical diagnostic decision support system improves residents' differential diagnoses or management plans. Twenty first-year internal medicine residents at Massachusetts General Hospital viewed 3 deidentified case descriptions of real patients. All residents completed a web-based questionnaire, entering the differential diagnosis and management plan before and after seeing the diagnostic decision support system's suggested list of diseases. In all 3 exercises, the actual case diagnosis was first on the system's list. Each resident served as his or her own control (pretest/posttest). For all 3 cases, a substantial percentage of residents changed their primary considered diagnosis after reviewing the system's suggested diagnoses, and a number of residents who had not initially listed a "further action" (laboratory test, imaging study, or referral) added or changed their management options after using the system. Many residents (20% to 65% depending on the case) improved their differential diagnosis from before to after viewing the system's suggestions. The average time to complete all 3 cases was 15.4minutes. Most residents thought that viewing the medical diagnostic decision support system's list of suggestions was helpful. Viewing a rank-ordered list of diagnostic possibilities from a diagnostic decision support tool had a significant beneficial effect on the quality of first-year medicine residents' differential diagnoses and management plans.

Does Bayes have it? Decision Support Systems in diagnostic medicine

This paper aims to show how psychological research can contribute to the design of medical diagnostic Decision Support Systems (DSSs). A brief overview of some popular Artificial Intelligence (AI)-based algorithms typically applied in such DSSs is presented first. The concept of diagnosticity is then introduced, which leads into a brief discussion of DSSs in medicine. A set of biases found in human decision-making is then outlined. This is followed by a summary of five psychological experiments investigating information integration in the presence of nondiagnostic case-specific evidence. The results revealed a robust tendency of medical diagnosticians to pay a disproportionate amount of attention to the first item presented. Finally, the design of a DSS in which the principles uncovered in the five experiments are included is discussed. Relevance to industry The design of a Bayesian Decision Support System (DSS) for training medical decision-making under uncertainty is described. Research revealing systematic, robust judgmental biases informed the design. Results showed that diagnosticity is crucial for useful DSSs, and that designers should quantify the variables to be applied in their DSS algorithms. The notion of diagnosticity, the biases uncovered, and the DSS model are applicable to a wide variety of areas of human endeavour characterized by complex decisions and ambiguous information.

Feature saliency using signal-to-noise ratios in automated diagnostic systems developed for Doppler ultrasound signals

Artificial neural networks (ANNs) have been used in a great number of medical diagnostic decision support system applications and within feedforward ANNs framework there are a number of established measures such as saliency measures for identifying important input features. By identifying a set of salient features, the noise in a classification model can be reduced, resulting in more accurate classification. In this study, a signal-to-noise ratio (SNR) saliency measure was employed to determine saliency of input features of multilayer perceptron neural networks (MLPNNs) used in classification of Doppler signals. The SNR saliency measure determines the saliency of a feature by comparing it to that of an injected noise feature and the SNR screening method utilizes the SNR saliency measure to select a parsimonious set of salient features. Ophthalmic and internal carotid arterial Doppler signals were decomposed into time–frequency representations using discrete wavelet transform. Input feature vectors were extracted using statistics over the set of the wavelet coefficients. The MLPNNs used in classification of the ophthalmic and internal carotid arterial Doppler signals were trained for the SNR screening method. The application results of the SNR screening method to the ophthalmic and internal carotid arterial Doppler signals demonstrated that classification accuracies of the MLPNNs with salient input features are higher than that of the MLPNNs with salient and non-salient input features.

Recurrent neural networks employing Lyapunov exponents for EEG signals classification

There are a number of different quantitative models that can be used in a medical diagnostic decision support system including parametric methods, non-parametric methods and several neural network models. Unfortunately, there is no theory available to guide model selection. The aim of this study is to evaluate the diagnostic accuracy of the recurrent neural networks (RNNs) employing Lyapunov exponents trained with Levenberg-Marquardt algorithm on the electroencephalogram (EEG) signals. An approach based on the consideration that the EEG signals are chaotic signals was used in developing a reliable classification method for electroencephalographic changes. This consideration was tested successfully using the non-linear dynamics tools, like the computation of Lyapunov exponents. We explored the ability of designed and trained Elman RNNs, combined with the Lyapunov exponents, to discriminate the EEG signals (EEG signals recorded from healthy volunteers with eyes open, epilepsy patients in the epileptogenic zone during a seizure-free interval, and epilepsy patients during epileptic seizures). The RNNs achieved accuracy rates which were higher than that of the feedforward neural network models. The obtained results demonstrated that the proposed RNNs employing the Lyapunov exponents can be useful in analyzing long-term EEG signals for early detection of the electroencephalographic changes.

Ensemble strategies for a medical diagnostic decision support system: A breast cancer diagnosis application

The model selection strategy is an important determinant of the performance and acceptance of a medical diagnostic decision support system based on supervised learning algorithms. This research investigates the potential of various selection strategies from a population of 24 classification models to form ensembles in order to increase the accuracy of decision support systems for the early detection and diagnosis of breast cancer. Our results suggest that ensembles formed from a diverse collection of models are generally more accurate than either pure-bagging ensembles (formed from a single model) or the selection of a “single best model.” We find that effective ensembles are formed from a small and selective subset of the population of available models with potential candidates identified by a multicriteria process that considers the properties of model generalization error, model instability, and the independence of model decisions relative to other ensemble members.

Feature saliency using signal-to-noise ratios in automated diagnostic systems developed for ECG beats

Artificial neural networks (ANNs) have been used in a great number of medical diagnostic decision support system applications and within feedforward ANNs framework there are a number of established measures such as saliency measures for identifying important input features. By identifying a set of salient features, the noise in a classification model can be reduced, resulting in more accurate classification. In this study, a signal-to-noise ratio (SNR) saliency measure was employed to determine saliency of input features of multilayer perceptron neural networks (MLPNNs) used in classification of electrocardiogram (ECG) beats (normal beat, congestive heart failure beat, ventricular tachyarrhythmia beat, atrial fibrillation beat) obtained from the Physiobank database. The SNR saliency measure determines the saliency of a feature by comparing it to that of an injected noise feature and the SNR screening method utilizes the SNR saliency measure to select a parsimonious set of salient features. ECG signals were decomposed into time-frequency representations using discrete wavelet transform. Input feature vectors were extracted using statistics over the set of the wavelet coefficients. The MLPNNs used in the ECG beats-classification were trained for the SNR screening method. The application results of the SNR screening method to the ECG signals demonstrated that classification accuracies of the MLPNNs with salient input features are higher than that of the MLPNNs with salient and non-salient input features.

Improving medical diagnostic accuracy of ultrasound Doppler signals by combining neural network models

There are a number of different quantitative models that can be used in a medical diagnostic decision support system including parametric methods (linear discriminant analysis or logistic regression), nonparametric models ( k nearest neighbor or kernel density) and several neural network models. The complexity of the diagnostic task is thought to be one of the prime determinants of model selection. Unfortunately, there is no theory available to guide model selection. This paper illustrates the use of combined neural network models to guide model selection for diagnosis of ophthalmic and internal carotid arterial disorders. The ophthalmic and internal carotid arterial Doppler signals were decomposed into time–frequency representations using discrete wavelet transform and statistical features were calculated to depict their distribution. The first-level networks were implemented for the diagnosis of ophthalmic and internal carotid arterial disorders using the statistical features as inputs. To improve diagnostic accuracy, the second-level networks were trained using the outputs of the first-level networks as input data. The combined neural network models achieved accuracy rates which were higher than that of the stand-alone neural network models.

Model selection for medical diagnosis decision support systems

In this paper, we examine the model section decision for a medical diagnostic decision support system (MDSS). Our purpose in doing this is to understand how model selection affects the accuracy of the decision support system. We explore two related research questions: (1) Do ensembles of models, acting as a single decision maker, perform more accurately than single models; and (2) How does model diversity affect the accuracy of the ensembles? Specifically, we compare 23 single models and bootstrap aggregating (i.e., bagging) models for their predictive abilities across five diverse medical data sets. We are able to reach important conclusions about our research objectives. Ensembles are more accurate than single models in their predictive ability. The best ensemble model achieves an error level significantly lower than the error of the best single model for four of the five medical applications analyzed. The magnitude of the error reduction ranges from 6.4% to 17.5%. Also, when designing an ensemble for an MDSS, the decision to diversify the model selection should be guided by the relationship between model instability and generalization error for the population of models under consideration.

Model selection for a medical diagnostic decision support system: a breast cancer detection case

There are a number of different quantitative models that can be used in a medical diagnostic decision support system (MDSS) including parametric methods (linear discriminant analysis or logistic regression), non-parametric models ( K nearest neighbor, or kernel density) and several neural network models. The complexity of the diagnostic task is thought to be one of the prime determinants of model selection. Unfortunately, there is no theory available to guide model selection. Practitioners are left to either choose a favorite model or to test a small subset using cross validation methods. This paper illustrates the use of a self-organizing map (SOM) to guide model selection for a breast cancer MDSS. The topological ordering properties of the SOM are used to define targets for an ideal accuracy level similar to a Bayes optimal level. These targets can then be used in model selection, variable reduction, parameter determination, and to assess the adequacy of the clinical measurement system. These ideas are applied to a successful model selection for a real-world breast cancer database. Diagnostic accuracy results are reported for individual models, for ensembles of neural networks, and for stacked predictors.

A diagnostic support system in general practice: Is it feasible?

A medical diagnostic decision support system (DDSS) has been developed for and tested in general practice. Two major issues have been addressed: diagnostic support and usefulness. The diagnostic support pertains to the ability of the system to generate diagnostic hypotheses from a set of patient data. The usefulness is approached by creating a computer system which can be used simultaneously with the doctor-patient consultation. The support function operates by matching symptoms from the patient data base with symptom configurations contained in the knowledge base. The support is presented as a list of diagnostic hypotheses ranked by degree of concordance. A user-friendly interface has been constructed with a comprehensive set of clinical terms within which the doctor can locate a desired symptom and store it with a single keystroke. With another keystroke the doctor can check the stored data and ask for support at any moment during the process. The overall purpose is to invite the doctor to rethink and re-examine his steps and to reconsider possible alternatives in the light of the presented diagnostic information. In our view it has to be the doctor who makes the final judgement. A test with the system in general practice revealed good performance of the system and an astonishing proficiency of the participating doctors in its use during the consultation. Twenty doctors solved five patient cases, entering 2000 clinical items within acceptable limits of consultation time. In 96% of the cases the correct diagnosis appeared in the differential diagnosis list. The doctors' diagnostic accuracy was 43%. The use of standardised terminology as an option for further development is discussed. The role of the doctor in computer-aided diagnostics remains open to debate. A computer-aided diagnostic support system in general practice appears to be feasible.