Intelligent Clinical Decision Support Research Articles

Clinical and social characterization of patients hospitalized for COPD exacerbation using Machine Learning tools

Objective: This study aims to employ machine learning (ML) tools to cluster patients hospitalized for acute exacerbations of chronic obstructive pulmonary disease (COPD) based on their diverse social and clinical characteristics. This clustering is intended to facilitate the subsequent analysis of differences in clinical outcomes.Methods: We analysed a cohort of patients with severe COPD from two Pulmonary Departments in north-western Spain using the k-prototypes algorithm, incorporating demographic, clinical, and social data. The resulting clusters were correlated with metrics such as readmissions, mortality, and place of death. Additionally, we developed an Intelligent Clinical Decision Support System (ICDSS) using a supervised ML model (Random Forest) to assign new patients to these clusters based on a reduced set of variables.Results: The cohort consisted of 524 patients, with an average age of 70.30 ± 9.35 years, 77.67% male, and an average FEV1 of 44.43 ± 15.4. Four distinct clusters (A-D) were identified with varying clinical-demographic and social profiles. Cluster D showed the highest levels of dependency, social isolation, and increased rates of readmissions and mortality. Cluster B was characterized by prevalent cardiovascular comorbidities. Cluster C included a younger demographic, with a higher proportion of women and significant psychosocial challenges. The ICDSS, using five key variables, achieved areas under the ROC curve of at least 0.91.Conclusions: ML tools effectively facilitate the social and clinical clustering of patients with severe COPD, closely related to resource utilization and prognostic profiles. The ICDSS enhances the ability to characterize new patients in clinical settings.

Development of Artificial Intelligence for Clinical Decision Support to Optimizes Erythropoietin-Stimulating Agents in Patients with CKD

Development of Artificial Intelligence for Clinical Decision Support to Optimizes Erythropoietin-Stimulating Agents in Patients with CKD

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

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

Multimodal risk classifier for cardiorespiratory diseases taking into account concomitant diseases and synergy effect

The purpose of the research is to develop a methodology for classifying complexly structured halftone images based on a multimodal approach using methods of morphological analysis, spectral analysis and neural network modeling.Methods. A method for classifying the contours of the boundaries of segments of a complexly structured image is described. The method is based on the fact that in chronic diseases of the pancreas, there is a violation of the integrity of the contour of its border and its waviness increases due to retractions and bulges caused by an alterative inflammatory process. The method includes the stages of normalization of ultrasound images and image segmentation with the selection of the contour of the object of interest. To classify the contour of a segment boundary, it is proposed to use Fourier analysis and neural network technologies. The method is illustrated using the example of classifying the contour of the border of the pancreas on its transcutaneous acoustic image.Results. Experimental studies of the proposed methods and means for classifying medical risk were carried out on diagnostic tasks according to the following classes: "chronic pancreatitis" – "without pathology". For experimental studies, video sequences of ultrasound images of the pancreas provided by an endoscopist were used. The purpose of the experimental studies was to analyze the classification quality indicators of image classifiers with class segments "Chronic pancreatitis" and "Without pathology". The training sample of video images (frames of video sequences) included 200 examples, one hundred from each class. The quality indicator "Sensitivity" of classification for two classes is 85,7%, the indicator "Specificity" is 87,1%.Conclusion. The use of the contour analysis method in classifiers of ultrasound images of the pancreas opens up new opportunities for accessible and objective diagnosis of pancreatic diseases, expanding the capabilities of intelligent clinical decision support systems.

A novel soft computing based efficient feature selection approach for timely identification of COVID-19 infection using chest computed tomography images: a human centered intelligent clinical decision support system

A novel soft computing based efficient feature selection approach for timely identification of COVID-19 infection using chest computed tomography images: a human centered intelligent clinical decision support system

B-34 | Effects of Artificial Intelligence Clinical Decision Support Tools (Terumo Health Outcomes, ePRISM) on Contrast Induced Acute Kidney Injury (CI-AKI) in Patients Undergoing Percutaneous Coronary Intervention (PCI)

B-34 | Effects of Artificial Intelligence Clinical Decision Support Tools (Terumo Health Outcomes, ePRISM) on Contrast Induced Acute Kidney Injury (CI-AKI) in Patients Undergoing Percutaneous Coronary Intervention (PCI)

OR12-6 | Effects of Artificial Intelligence Clinical Decision Support Tools (Terumo Health Outcomes, ePRISM) on Post-Procedure Bleeding Events in Patients Undergoing Percutaneous Coronary Intervention (PCI)

OR12-6 | Effects of Artificial Intelligence Clinical Decision Support Tools (Terumo Health Outcomes, ePRISM) on Post-Procedure Bleeding Events in Patients Undergoing Percutaneous Coronary Intervention (PCI)

Automated Classification of Abdominal Ultrasound lmages of the Pancreas Based on the Spectral Representation of the Border’s Contours

The purpose of the research is to develop a methodology for classifying complexly structured halftone images based on a multimodal approach using methods of morphological analysis, spectral analysis and neural network modeling.Methods. А method for classifying the contours of the boundaries of segments of a complexly structured image is described. Тhe method is based on the fact that in chronic diseases of the pancreas, there is a violation of the integrity of the contour of its border and its waviness increases due to retractions and bulges caused by an alterative inflammatory process. Тhe method includes the stages of normalization of ultrasound images and image segmentation with the selection of the contour of the object of interest. Тo classify the contour of a segment boundary, it is proposed to use Fourier analysis and neural network technologies. Тhe method is illustrated using the example of classifying the contour of the border of the pancreas on its transcutaneous acoustic image.Results. Еxperimental studies of the proposed methods and means for classifying medical risk were carried out on diagnostic tasks according to the following classes: "chronic pancreatitis" – "without pathology". For experimental studies, video sequences of ultrasound images of the pancreas provided by an endoscopist were used. Тhe purpose of the experimental studies was to analyze the classification quality indicators of image classifiers with class segments "Chronic pancreatitis" and "Without pathology". Тhe training sample of video images (frames of video sequences) included 200 examples, one hundred from each class. Тhe quality indicator "Sensitivity" of classification for two classes is 85,7%, the indicator "Specificity" is 87,1%.Тhe use of the contour analysis method in classifiers of ultrasound images of the pancreas opens up new opportunities for accessible and objective diagnosis of pancreatic diseases, expanding the capabilities of intelligent clinical decision support systems.

Proposal and Definition of an Intelligent Clinical Decision Support System Applied to the Prediction of Dyspnea after 12 Months of an Acute Episode of COVID-19.

Long COVID is a condition that affects a significant proportion of patients who have had COVID-19. It is characterised by the persistence of associated symptoms after the acute phase of the illness has subsided. Although several studies have investigated the risk factors associated with long COVID, identifying which patients will experience long-term symptoms remains a complex task. Among the various symptoms, dyspnea is one of the most prominent due to its close association with the respiratory nature of COVID-19 and its disabling consequences. This work proposes a new intelligent clinical decision support system to predict dyspnea 12 months after a severe episode of COVID-19 based on the SeguiCovid database from the Álvaro Cunqueiro Hospital in Vigo (Galicia, Spain). The database is initially processed using a CART-type decision tree to identify the variables with the highest predictive power. Based on these variables, a cascade of expert systems has been defined with Mamdani-type fuzzy-inference engines. The rules for each system were generated using the Wang-Mendel automatic rule generation algorithm. At the output of the cascade, a risk indicator is obtained, which allows for the categorisation of patients into two groups: those with dyspnea and those without dyspnea at 12 months. This simplifies follow-up and the performance of studies aimed at those patients at risk. The system has produced satisfactory results in initial tests, supported by an AUC of 0.75, demonstrating the potential and usefulness of this tool in clinical practice.

Development and evaluation of a clinical decision support system for early diagnosis of acute appendicitis

The most frequent reason for individuals experiencing abdominal discomfort to be referred to emergency departments of hospitals is acute appendicitis, and the most frequent emergency surgery performed is an appendectomy. The purpose of this study was to design and develop an intelligent clinical decision support system for the timely and accurate diagnosis of acute appendicitis. The number of participants which is equal to 181 was chosen as the sample size for developing and evaluating neural networks. The information was gathered from the medical files of patients who underwent appendicectomies at Shahid Modarres Hospital as well as from the findings of their appendix samples' pathological tests. The diagnostic outcomes were then ascertained by the development and comparison of a Multilayer Perceptron network (MLP) and a Support Vector Machine (SVM) system in the MATLAB environment. The SVM algorithm functioned as the central processing unit in the Clinical Decision Support System (CDSS) that was built. The intelligent appendicitis diagnostic system was subsequently developed utilizing the Java programming language. Technical evaluation and system usability testing were both done as part of the software evaluation process. Comparing the output of the optimized artificial neural network of the SVM with the pathology result showed that the network's sensitivity, specificity, and accuracy were 91.7%, 96.2%, and 95%, respectively, in diagnosing acute appendicitis. Based on the existing standards and the opinions of general surgeons, and also comparing the results with the diagnostic accuracy of general surgeons, findings indicated the proper functioning of the network for the diagnosis of acute appendicitis. The use of this system in medical centers is useful for purposes such as timely diagnosis and prevention of negative appendectomy, reducing patient hospital stays and treatment costs, and improving the patient referral system.

Effectiveness of an artificial intelligence clinical assistant decision support system to improve the incidence of hospital-associated venous thromboembolism: a prospective, randomised controlled study

BackgroundThromboprophylaxis has been determined to be safe, effective and cost-effective for hospitalised patients at venous thromboembolism (VTE) risk. However, Chinese medical institutions have not yet fully used or improperly used...

Intelligent Clinical Decision Support System for Managing COPD Patients.

Chronic obstructive pulmonary disease (COPD) is the third leading cause of death worldwide. Health remote monitoring systems (HRMSs) play a crucial role in managing COPD patients by identifying anomalies in their biometric signs and alerting healthcare professionals. By analyzing the relationships between biometric signs and environmental factors, it is possible to develop artificial intelligence models that are capable of inferring patients' future health deterioration risks. In this research work, we review recent works in this area and develop an intelligent clinical decision support system (CIDSS) that is capable of providing early information concerning patient health evolution and risk analysis in order to support the treatment of COPD patients. The present work's CIDSS is composed of two main modules: the vital signs prediction module and the early warning score calculation module, which generate the patient health information and deterioration risks, respectively. Additionally, the CIDSS generates alerts whenever a biometric sign measurement falls outside the allowed range for a patient or in case a basal value changes significantly. Finally, the system was implemented and assessed in a real case and validated in clinical terms through an evaluation survey answered by healthcare professionals involved in the project. In conclusion, the CIDSS proves to be a useful and valuable tool for medical and healthcare professionals, enabling proactive intervention and facilitating adjustments to the medical treatment of patients.

Usability and performance expectancy govern spine surgeons' use of a clinical decision support system for shared decision-making on the choice of treatment of common lumbar degenerative disorders.

Quantitative survey study is the study design. The study aims to develop a model for the factors that drive or impede the use of an artificial intelligence clinical decision support system (CDSS) called PROPOSE, which supports shared decision-making on the choice of treatment of ordinary spinal disorders. A total of 62 spine surgeons were asked to complete a questionnaire regarding their behavioral intention to use the CDSS after being introduced to PROPOSE. The model behind the questionnaire was the Unified Theory of Acceptance and Use of Technology. Data were analyzed using partial least squares structural equation modeling. The degree of ease of use associated with the new technology (effort expectancy/usability) and the degree to which an individual believes that using a new technology will help them attain gains in job performance (performance expectancy) were the most important factors. Social influence and trust in the CDSS were other factors in the path model. r2 for the model was 0.63, indicating that almost two-thirds of the variance in the model was explained. The only significant effect in the multigroup analyses of path differences between two subgroups was for PROPOSE use and social influence (p = 0.01). Shared decision-making is essential to meet patient expectations in spine surgery. A trustworthy CDSS with ease of use and satisfactory predictive ability promoted by the leadership will stand the best chance of acceptance and bridging the communication gap between the surgeon and the patient.

An Intelligent Clinical Decision Support System based on Attentive Federated Learning

An Intelligent Clinical Decision Support System based on Attentive Federated Learning

#3815 A DEEP LEARNING APPROACH TO PERSONALISED ANTI-HYPERTENSIVE MEDICATION TITRATION

Abstract Background and Aims Hypertension is the number one risk factor for premature death worldwide. Artificial Intelligence (AI) Clinical Decision Support Systems are an important next step for hypertension management but require rigorous evaluation before assimilation into routine clinical practice. Our aim is to develop and evaluate an AI clinical decision support tool for hypertension management trained on randomised clinical trial data. Method The Systolic Blood Pressure Intervention Trial (SPRINT) trial showed that intensive BP control to SBP &amp;lt;120 mm Hg results in significant cardiovascular benefit in high-risk patients with hypertension compared with routine BP control to &amp;lt;140 mm Hg. We trained a feed-forward neural network using Keras and Tensorflow for R and data from 9361 persons in the SPRINT randomized clinical trial to predict the probability of increase, reduce or no-change in total number of anti-hypertensive medications at each visit. The network is designed to model SPRINT investigator deviations from the protocol. Six baseline patient variables (age, sex, race, aspirin use, eGFR, group assignment (intensive or standard)), two visit patient variables (Systolic and Diastolic Blood Pressure), and two previous visit variables (Systolic and Diastolic Blood Pressure) were inputted into the model after data normalisation and centering. Hyperparameters were tuned using a grid search method. We conducted internal validation using a 20% validation set. Clinical validation was performed using an unseen test set of typical hypertension scenarios (n = 50). An R Shiny app was developed to enter new patient information and display a sensitivity analysis of probabilities calculated after changing each baseline variable by plus and minus 10 (Figure 1). To avoid poor performance on new out of distribution cases we tested five methods for Out of Distribution (OOD) detector, 1. Autoencoders, 2. Normalizing flow, 3. Local Outlier Factor (LOF), 4. Probabilistic Principal Component Analysis (PPCA) and Kernel Density Estimation (KDE). Results The accuracy was 76.7% on the validation dataset after 1000 Epochs. Clinical validation revealed suboptimal performance for unseen Out of Distribution data (e.g. recommending reduction of medication when SBP &amp;gt;160 and number of medications was zero). This poor performance on OOD samples prompted the creation of an OOD detector to use in series with the AI Decision Support System for Hypertension. The AUROCs were 0.77 for AE, 0.96 for Normalizing Flow, 0.82 for LOF, 0.88 for PPCA and 0.96 for the Kernel Density Estimation. Conclusion Artificial Intelligence (AI) Clinical Decision Support Systems for Hypertension are feasible and an important next step for hypertension management but generalisability and safety require rigorous evaluation before assimilation into routine clinical practice.

Intelligent clinical decision support for small patient datasets

Intelligent clinical decision support for small patient datasets

Integration of the Wang &amp; Mendel Algorithm into the Application of Fuzzy Expert Systems to Intelligent Clinical Decision Support Systems

The use of intelligent systems in clinical diagnostics has evolved, integrating statistical learning and knowledge-based representation models. Two recent works propose the identification of risk factors for the diagnosis of obstructive sleep apnea (OSA). The first uses statistical learning to identify indicators associated with different levels of the apnea-hypopnea index (AHI). The second paper combines statistical and symbolic inference approaches to obtain risk indicators (Statistical Risk and Symbolic Risk) for a given AHI level. Based on this, in this paper we propose a new intelligent system that considers different AHI levels and generates risk pairs for each level. A learning-based model generates Statistical Risks based on objective patient data, while a cascade of fuzzy expert systems determines a Symbolic Risk using symptom data from patient interviews. The aggregation of risk pairs at each level involves a fuzzy expert system with automatically generated fuzzy rules using the Wang-Mendel algorithm. This aggregation produces an Apnea Risk indicator for each AHI level, allowing discrimination between OSA and non-OSA cases, along with appropriate recommendations. This approach improves variability, usefulness, and interpretability, increasing the reliability of the system. Initial tests on data from 4400 patients yielded AUC values of 0.74–0.88, demonstrating the potential benefits of the proposed intelligent system architecture.

PO-04-205 A NOVEL ARTIFICIAL INTELLIGENCE ALGORITHM TO IDENTIFY PATIENTS AT RISK OF SUDDEN CARDIAC ARREST

Sudden cardiac death (SCD) from cardiac arrest is a major international public health problem accounting for an estimated 15%-20% of all deaths. Implantable cardioverter defibrillators (ICD) have been demonstrated to improve survival in high-risk populations. While 500,000 patients experience SCD annually, there are only 75,000 ICDs implanted annually suggesting a significant treatment gap. To determine if a novel artificial intelligence enabled clinical decision support algorithm can improve identification of patients at risk for sudden cardiac arrest An evidence based algorithm was created to identify patients at risk for sudden cardiac arrest that had not been referred to electrophysiology. This algorithm was embedded in an electronic medical record and used natural language processing to review patient records, including progress notes, diagnostic studies and appointments. The algorithm would prompt clinicians to order additional diagnostic studies or refer patients to electrophysiology for consultation if appropriate. In 6 months of usage, the clinical decision support algorithm screened 27,112 patients, 34,360 echocardiograms and identified 1,112 patients at risk for SCA. Alerts were sent for 208 patients, out of which 88 alerts were acted upon while 120 were pending action. In 55 patients who received a follow up echocardiogram, the EF remained < 35% and 22 of these patients were referred to EP, out of which 4 patients received an ICD. A novel artificial intelligence clinical decision support algorithm is a feasible and effective means of identifying patients at risk of sudden cardiac arrest. Further study is needed to asses the relative impact of this intervention on clinical care.

Artificial intelligence for clinical decision support for monitoring patients in cardiovascular ICUs: A systematic review.

Artificial intelligence (AI) and machine learning (ML) models continue to evolve the clinical decision support systems (CDSS). However, challenges arise when it comes to the integration of AI/ML into clinical scenarios. In this systematic review, we followed the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA), the population, intervention, comparator, outcome, and study design (PICOS), and the medical AI life cycle guidelines to investigate studies and tools which address AI/ML-based approaches towards clinical decision support (CDS) for monitoring cardiovascular patients in intensive care units (ICUs). We further discuss recent advances, pitfalls, and future perspectives towards effective integration of AI into routine practices as were identified and elaborated over an extensive selection process for state-of-the-art manuscripts. Studies with available English full text from PubMed and Google Scholar in the period from January 2018 to August 2022 were considered. The manuscripts were fetched through a combination of the search keywords including AI, ML, reinforcement learning (RL), deep learning, clinical decision support, and cardiovascular critical care and patients monitoring. The manuscripts were analyzed and filtered based on qualitative and quantitative criteria such as target population, proper study design, cross-validation, and risk of bias. More than 100 queries over two medical search engines and subjective literature research were developed which identified 89 studies. After extensive assessments of the studies both technically and medically, 21 studies were selected for the final qualitative assessment. Clinical time series and electronic health records (EHR) data were the most common input modalities, while methods such as gradient boosting, recurrent neural networks (RNNs) and RL were mostly used for the analysis. Seventy-five percent of the selected papers lacked validation against external datasets highlighting the generalizability issue. Also, interpretability of the AI decisions was identified as a central issue towards effective integration of AI in healthcare.

Proposal and Definition of an Intelligent Clinical Decision Support System Applied to the Screening and Early Diagnosis of Breast Cancer.

Breast cancer is the most frequently diagnosed tumor pathology on a global scale, being the leading cause of mortality in women. In light of this problem, screening programs have been implemented on the population at risk in the form of mammograms, starting in the 20th century. This has considerably reduced the associated deaths, as well as improved the prognosis of the patients who suffer from this disease. In spite of this, the evaluation of mammograms is not without certain variability and depends, to a large extent, on the experience and training of the medical team carrying out the assessment. With the aim of supporting the evaluation process of mammogram images and improving the diagnosis process, this work presents the design, development and proof of concept of a novel intelligent clinical decision support system, grounded on two predictive approaches that work concurrently. The first of them applies a series of expert systems based on fuzzy inferential engines, geared towards the treatment of the characteristics associated with the main findings present in mammograms. This allows the determination of a series of risk indicators, the Symbolic Risks, related to the risk of developing breast cancer according to the different findings. The second one implements a classification machine learning algorithm, which using data related to mammography findings as well as general patient information determines another metric, the Statistical Risk, also linked to the risk of developing breast cancer. These risk indicators are then combined, resulting in a new indicator, the Global Risk. This could then be corrected using a weighting factor according to the BI-RADS category, allocated to each patient by the medical team in charge. Thus, the Corrected Global Risk is obtained, which after interpretation can be used to establish the patient's status as well as generate personalized recommendations. The proof of concept and software implementation of the system were carried out using a data set with 130 patients from a database from the School of Medicine and Public Health of the University of Wisconsin-Madison. The results obtained were encouraging, highlighting the potential use of the application, albeit pending intensive clinical validation in real environments. Moreover, its possible integration in hospital computer systems is expected to improve diagnostic processes as well as patient prognosis.