Medical Decision Support System Research Articles

Analysis of the effectiveness of using U-net architecture for classification and segmentation of glioma in MRI images

The purpose of the research is to analyze the efficiency of the U-net neural network architecture in decision support systems for glioma diagnostics and segmentation of brain areas affected by it on MRI images.Methods. To conduct experimental studies, a training dataset was generated and the data was normalized. A software implementation of the U-Net neural network architecture was performed using the Keras framework in the Python programming language. The neural network model was trained.Results. A series of experiments were conducted, during which error and classification matrices were obtained, the efficiency of classification of the trained neural network model for the "Tumor" and "No tumor" classes was assessed using metrics such as Recall, Precision and F1-measure, and the quality of segmentation of glioma-affected areas on the test data set was assessed. The quality of segmentation was assessed using the IoU metric, which reflects the ratio of the areas of the bounding boxes and is used to assess the accuracy of the spatial correspondence of the predicted segmented areas highlighted on the masks. Based on the results of testing the neural network model in solving the problem of segmenting brain areas affected by glioma, the average value of the IoU metric was 0.812, which is an acceptable result.Conclusion. The testing results showed that the neural network model based on the U-net architecture is able to effectively diagnose the presence of glioma with acceptable values of the classification and segmentation quality metrics, which indicates the possibility of using this neural network model in medical decision support systems for glioma diagnostics, as well as its segmentation on MRI images. However, it is advisable to refine this neural network model to reduce the number of false negative classification results, which is critically important in medical diagnostics.

Temporal Lobe Epilepsy Focus Detection Based on the Correlation Between Brain MR Images and EEG Recordings with a Decision Tree

Background/Objectives: In this study, a medical decision support system is presented to assist physicians in epileptic focus detection by correlating MRI and EEG data of temporal lobe epilepsy patients. Methods: By exploiting the asymmetry in the hippocampus in MRI images and using voxel-based morphometry analysis, gray matter reduction in the temporal and limbic lobes is detected, and epileptic focus prediction is realized. In addition, an epileptic focus is also determined by calculating the asymmetry score from EEG channels. Finally, epileptic focus detection was performed by associating MRI and EEG data with a decision tree. Results: The results obtained from the proposed algorithm provide 100% overlap with the physician’s finding on the EEG data. Conclusions: MRI and EEG correlation in epileptic focus detection was improved compared with physicians. The proposed algorithm can be used as a medical decision support system for epilepsy diagnosis, treatment, and surgery planning.

The system of support for medical decision-making in the formation of a rehabilitation diagnosis in the categories of the International Classification of Functioning, Disability and Health

Introduction. Comprehensive rehabilitation of patients is a key task of the national healthcare system. To create an effective rehabilitation program, it is necessary to establish an accurate rehabilitation diagnosis. Designing an individual medical rehabilitation plan based on a biopsychosocial approach, the practical tool of which was the “International Classification of Functioning, Disability and Health” (ICF), has changed the global concept of rehabilitation. A priority in the field of digital healthcare is the introduction of medical decision support systems, one of the tasks of which is to help in diagnosis and minimize medical errors aimed at improving the quality of medical care. The purpose of the work: to create a software system for making medical decisions in the formation of a rehabilitation diagnosis in the ICF categories. Materials and methods. content analysis was used to study and analyze medical decision-making systems (foreign and domestic) and the ICF; scientific and methodological work was carried out on the development of software for medical decision-making systems. Results. based on the work carried out, software has been developed to automate, systematize, and optimize the process of establishing a rehabilitation diagnosis in patients with ischemic stroke, standardize approaches to individualizing diagnosis formulation taking into account the severity of functional disorders. Research limitations. During the development and description of the software, the capabilities of medical information systems applied in the healthcare of Moscow were used. Conclusion. The biopsychosocial approach is one of the main principles of modern rehabilitation. Digital healthcare facilitates the integration of decision support systems that improve the quality of medical care. The implementation of the developed program into digital platforms of the healthcare system will reduce time costs and unify the procedure for making a rehabilitation diagnosis and all related processes, consistently preserving the necessary information about the patient.

Experience in using artificial intelligence services for diagnosing compression fracture of vertebral body based on computed tomography – from testing to trials

Backgraund:Osteoporosis (OP) is often diagnosed at the stage of complications. Compression fractures (CF) of the vertebral bodies, complications of OP, and predictors of subsequent fractures are often asymptomatic. CF can be detected on CT scan performed for other indications. We have analyzed approaches to using artificial intelligence services (AI services) generated for diagnosing CF of the vertebral bodies. Aims:Testing AI services to identify CF based on chest CT and to assess the possibility of their implementation into the practice of medical facilities of the Moscow Healthcare Department. Materials and methods:For setting a clinical task for AI services, basic diagnostic requirements in the direction of “Osteoporosis” are formed. AI services go through the following stages: self-testing, functional and calibration testing, trial and trial operation. The first three stages of testing are carried out on previously generated datasets. At the stages of trial and trial operation, AI services analyze CT examinations performed in medical facilities. An expert group of radiologists evaluates the diagnostic accuracy and functional capacity of AI services at all stages. The obtained quantitative metrics of the accuracy of AI services are compared with the target values ​​specified in the requirements. Results:From June 2021 to June 2022, two AI services (AI service №1 and AI service №2) which used different methods for identifying a presence of CF were tested. AI services successfully passed the self-testing stage (6), functional (5) and calibration (100) testing. The ROC AUC for AI service №1 was 0.99 and for AI service №2–0.91. The AI service №1 passed the trial stage without any significant remarks, while AI service №2 was sent for revision. After the trial operation stage, the following accuracy metrics were obtained: the ROC AUC for AI service №1 was 0.93; for AI service №2–0.92. At all stages, both AI services showed sufficient metrics for clinical validation. Conclusions:A testing of the performance of AI services in the direction of “Osteoporosis” was carried out. A high quality of diagnosis was demonstrated. AI services can act as an auxiliary tool in the medical decision support system.

An electrocardiogram signal classification using a hybrid machine learning and deep learning approach

An electrocardiogram (ECG) is a diagnostic tool that captures the electrical activity of the heart. Any irregularity in the heart's electrical system is referred to as an arrhythmia, which can be identified through the analysis of ECG signals. Timely diagnosis of cardiac arrhythmias is crucial in order to mitigate their potentially harmful consequences. However, manual analysis of ECG signals is time-consuming and prone to inaccuracies. Therefore, researchers have developed medical decision support systems that utilize machine learning techniques to automate the analysis of ECG signals. In this study, we propose a novel method for classifying ECG signals into four distinct types of heartbeats: normal, supraventricular, ventricular, and fusion. Our method consists of two subsystems that integrate both machine learning and deep learning approaches. The first subsystem uses a residual network block to extract features from the input ECG signal, followed by an LSTM network for learning and classification of these features. The second subsystem uses several feature extraction methods and a random forest to classify the ECG signals. Furthermore, it employs a Synthetic Minority Over-Sampling Technique to improve dataset balance and overall performance. The ultimate result is achieved by merging the results of both subsystems together. An assessment of our approach was carried out on the MIT-BIH dataset, which acts as a recognized ECG signal classification benchmark. Our technique attained an impressive accuracy rate of 99.26%, ranking it as one of the most superior methods in the current literature. Our findings demonstrate the effectiveness and efficiency of our approach in accurately classifying ECG signals for arrhythmia detection.

Enhancing Medication Safety with ML-Enhanced Decision Support Systems: A Comparative Analysis of Prescription Error Detection

Background: Medication prescription errors are a global issue, leading to significant morbidity and mortality. Traditional rule-based Medical Decision Support Systems (MDSS) are often ineffective, generating numerous false alerts and failing to detect all potential errors. This study assesses a new anomaly detection system integrated with Electronic Health Records (EHR) to improve the accuracy and utility of medication error warnings. Methods: Anomalous prescription detection was implemented alongside an existing MDSS in a real-world inpatient setting over 18 months. The new system utilized Machine Learning (ML) combined with a rule-based MDSS to analyze historical EHR data. It aimed to identify and flag high-risk prescriptions through real-time anomaly detection. The performance of this hybrid system was compared against traditional MDSS and multicriteria query (MQ) methods. A clinical pharmacist reviewed 415 patients (3401 prescriptions) to validate the effectiveness of the system, assessing notifications for accuracy, clinical relevance, and practicality. Results: The ML-enhanced MDSS demonstrated superior performance compared to traditional systems. It achieved a 76% interception rate for prescriptions needing pharmacist review and a precision rate of 75%. The hybrid system outperformed traditional MDSS and MQ methods, with areas under the ROC and PRC curves of 0.84 and 0.79, respectively, compared to 0.66 and 0.57 for MDSS and 0.7 and 0.58 for MQ approaches. Conclusion: Integrating ML with rule-based MDSS significantly improves the detection of high-risk medication prescriptions, reducing false alerts and enhancing accuracy. This hybrid approach offers a more effective tool for identifying potential medication errors and improving patient safety in inpatient settings.

A novel deer hunting optimized cat boost approach for interactive medical decision-making system

The need for precise and effective medical decision-making mechanisms is driving constant change in the healthcare sector. In this regard, algorithms based on machine learning are becoming prominent due to their potential to help medical practitioners diagnose and cure a wide range of illnesses. All over the world, one significant public health issue is chronic kidney disease (CKD), which requires prompt and precise diagnosis to manage and cure. A new Deer Hunting Optimized CatBoost (DHO-CB) method for CKD seriousness prediction is presented in this research. Our approach takes advantage of CatBoost's built-in flexibility and resilience by incorporating an optimization framework inspired by deer hunting. This framework dynamically modifies model parameters to traverse intricate medical data environments. The standard CKD dataset is used to verify the suggested DHO-CB approach, and Python is the primary language used in this study. In terms of kappa coefficient (92.6%), f1-score metrics (97.7%), accuracy (97.2%), and specificity (95.5%), the effectiveness of the suggested DHO-CB approach is examined. Based on this analyzed data, our suggested DHO-CB approach outperforms other current approaches in CKD prediction as well as severity assessment. The suggested DHO-CB offers a viable path toward the creation of complex yet approachable medical decision support systems, which will promote increased accuracy and effectiveness in the identification and treatment of CKDs.

Expert gaze as a usability indicator of medical AI decision support systems: a preliminary study

Given the current state of medical artificial intelligence (AI) and perceptions towards it, collaborative systems are becoming the preferred choice for clinical workflows. This work aims to address expert interaction with medical AI support systems to gain insight towards how these systems can be better designed with the user in mind. As eye tracking metrics have been shown to be robust indicators of usability, we employ them for evaluating the usability and user interaction with medical AI support systems. We use expert gaze to assess experts’ interaction with an AI software for caries detection in bitewing x-ray images. We compared standard viewing of bitewing images without AI support versus viewing where AI support could be freely toggled on and off. We found that experts turned the AI on for roughly 25% of the total inspection task, and generally turned it on halfway through the course of the inspection. Gaze behavior showed that when supported by AI, more attention was dedicated to user interface elements related to the AI support, with more frequent transitions from the image itself to these elements. When considering that expert visual strategy is already optimized for fast and effective image inspection, such interruptions in attention can lead to increased time needed for the overall assessment. Gaze analysis provided valuable insights into an AI’s usability for medical image inspection. Further analyses of these tools and how to delineate metrical measures of usability should be developed.

Trust me if you can: a survey on reliability and interpretability of machine learning approaches for drug sensitivity prediction in cancer.

With the ever-increasing number of artificial intelligence (AI) systems, mitigating risks associated with their use has become one of the most urgent scientific and societal issues. To this end, the European Union passed the EU AI Act, proposing solution strategies that can be summarized under the umbrella term trustworthiness. In anti-cancer drug sensitivity prediction, machine learning (ML) methods are developed for application in medical decision support systems, which require an extraordinary level of trustworthiness. This review offers an overview of the ML landscape of methods for anti-cancer drug sensitivity prediction, including a brief introduction to the four major ML realms (supervised, unsupervised, semi-supervised, and reinforcement learning). In particular, we address the question to what extent trustworthiness-related properties, more specifically, interpretability and reliability, have been incorporated into anti-cancer drug sensitivity prediction methods over the previous decade. In total, we analyzed 36 papers with approaches for anti-cancer drug sensitivity prediction. Our results indicate that the need for reliability has hardly been addressed so far. Interpretability, on the other hand, has often been considered for model development. However, the concept is rather used intuitively, lacking clear definitions. Thus, we propose an easily extensible taxonomy for interpretability, unifying all prevalent connotations explicitly or implicitly used within the field.

Prediction of diabetes mellitus using XGBoost model

With the number of health-threatening diseases and deaths on the rise, medical decision support systems continue to prove effective in improving the efficiency of physicians and other healthcare providers and supporting clinical decisions. Diabetes remains one of the leading diseases responsible for many deaths worldwide. Diabetes is characterized by elevated blood glucose levels, which can have serious consequences for other organs in the body. According to the International Diabetes Alliance (IDA), 382 million people currently have diabetes and this number is expected to double to 592 million by 2035. This paper proposes a clinical prediction model for diabetes based on machine learning (ML).As traditional ML models, we consider the most commonly used classifiers, the Logic Tree (LR) and XGBoost. We compare these ML models. On the other hand, we also use deep learning (DL) models and apply a fully convolutional neural network (DNN) for diabetes prediction and recognition. The proposed models were evaluated in the public Kaggle database: prediction performance at each level was analyzed using the Precision, Recall, AUC, accuracy and F1 metrics, and overall prediction efficiency was assessed using accuracy and its macro-average: DL, LR, The overall accuracy obtained by XGBoost was 91.88%, 88.75% and 97.12% respectively. The experimental results show that XGBoost is more effective at predicting diabetes than deep learning and LR methods.

Prediction of diabetes mellitus using XGBoost model

With the number of health-threatening diseases and deaths on the rise, medical decision support systems continue to prove effective in improving the efficiency of physicians and other healthcare providers and supporting clinical decisions. Diabetes remains one of the leading diseases responsible for many deaths worldwide. Diabetes is characterized by elevated blood glucose levels, which can have serious consequences for other organs in the body. According to the International Diabetes Alliance (IDA), 382 million people currently have diabetes and this number is expected to double to 592 million by 2035. This paper proposes a clinical prediction model for diabetes based on machine learning (ML).As traditional ML models, we consider the most commonly used classifiers, the Logic Tree (LR) and XGBoost. We compare these ML models. On the other hand, we also use deep learning (DL) models and apply a fully convolutional neural network (DNN) for diabetes prediction and recognition. The proposed models were evaluated in the public Kaggle database: prediction performance at each level was analyzed using the Precision, Recall, AUC, accuracy and F1 metrics, and overall prediction efficiency was assessed using accuracy and its macro-average: DL, LR, The overall accuracy obtained by XGBoost was 91.88%, 88.75% and 97.12% respectively. The experimental results show that XGBoost is more effective at predicting diabetes than deep learning and LR methods.

Effectiveness and safety of an artificial intelligence-based medical decision support system for adjusting insulin pump settings in children with type 1 diabetes mellitus: randomized controlled trial

BACKGROUND: Previously, we presented the process of developing a clinical decision support system (CDSS) for adjusting insulin pump (IP) settings in children with type 1 diabetes mellitus (T1D) and assessing the agreement of the recommendations it generates with the expert opinion. The CDSS demonstrated satisfactory forecasting of glucose profile and agreement rates between recommendations CDSS and experts.AIM: To evaluate the effectiveness and safety of using CDSS in children with T1D, testing the hypothesis of non-inferiority (with a limit of -5%) of relative increase of glucose time in range (TIR) over 6 months.MATERIALS AND METHODS: The trial included 80 children with T1D, divided into two comparable groups of 40 children using the minimization method. Patients in the main group received recommendations for adjusting the IP settings from a physician who uses the CDSS; patients in the control group received recommendations from a physician (control group). Patients were observed for 6 months with remote consultations once a month (7 consultations in total) and monitoring of glycated hemoglobin (HbA1c) at 1, 4 and 7 consultations. The primary outcome is the difference in group mean relative changes in TIR (%), secondary outcomes are TIR (%), HbA1c concentration. RESULTS: The trial was completed by 63 patients 32 in the main group, 31 in the control group. The difference in the mean relative increase in TIR at the 7th consultation in the groups was 3.02%, one-sided 95% CI (-4.55%; inf ). Thus, the lower bound of this CI is greater than the noninferiority limit of -5%, and the noninferiority hypothesis can be accepted. There were no statistically significant differences between groups for all outcomes. The dynamics of the indicators were positive in the main group and had a statistical tendency towards positive changes in the control group.CONCLUSION: The use of CDSS was no less effective in terms of the TIR than the management of the patient by a physician. The use of CDSS in clinical practice can help in regular and frequent monitoring of children with T1D, and standardize at a high level the approach to correction of IP parameters, supplemented with CGM.

The effectiveness of endoscopic diagnostics of colon tumors using artificial intelligence: prospective tandem study

AIM: to estimate the effectiveness of a medical decision support system based on artificial intelligence in the endoscopic diagnosis of benign tumors during tandem study.PATIENTS AND METHODS: from October to December 2023, a single-center comparative tandem study of medical decision support system based on artificial intelligence “ArtInCol” was done. The first stage was a traditional colonoscopy under sedation, the second one — colonoscopy using AI. A pairwise comparison of the main indicators of the effectiveness was made.RESULTS: in the primary endpoint, the polyp detection rate (PDR) in the traditional colonoscopy group was 40.6% vs 56.4% in the AI group, p = 0.0001 (RR = 1.39; 95% CI: 1.04–1.87). The mean number of lesions detected (MPP) was 1.63 (± 1.2) vs 2.47 (± 1.8) in the AI group (mean difference = 0.84; (95% CI: 1.07–0.61).CONCLUSION: the study demonstrated the effectiveness of the original medical decision support system for benign colon tumors detection in real clinical practice. The further stage, a multicenter randomized trial is needed.

AUTOMATED SYSTEMS FOR DIAGNOSING DISEASES: A REVIEW OF EXISTING TOOLS

Diagnostic expert systems are computer-based decision-support systems that are designed to assist healthcare professionals in diagnosing diseases. They are designed to provide a list of possible diagnoses based on a patient’s symptoms, medical history, and other relevant information. The basic operation of a diagnostic expert system involves capturing information from a patient, such as symptoms and medical history, and using that information to generate a list of potential diagnoses. The system then uses knowledge-based techniques, such as decision trees and rule-based systems, to narrow down the list of potential diagnoses to arrive at a final diagnosis. This research paper provides an overview of expert systems, including their history, architecture, knowledge representation. It also discusses the advantages and limitations of expert systems, as well as their applications in different fields. There are many research papers studied and research analysis done. There have been numerous research studies in the field of systems for diagnosing diseases, particularly in the area of medical expert systems and decision support systems.

Prevalence of signs of professional burnout among medical workers involved in providing medical care for circulatory system diseases

BACKGROUND: Among 13 thousand doctors interviewed in 2021, 42% displayed signs of professional burnout. This phenomenon negatively affects the frequency of repeated hospitalizations, the number of adverse events, and increased employee turnover. However, studies concerning burnout in doctors are rare and do not focus on specialists who provide medical care for patients with circulatory system diseases (CSD). AIM: was to assess the level of burnout in medical workers involved in the provision of medical care for patients with CSD and to explore staff perceptions of its prevention. MATERIALS AND METHODS: A sociological survey (based on the standardized Maslach Burnout Inventory questionnaire) was conducted among medical personnel (women represented the majority of the sample — 76%, men — 24%). The sample is represented by doctors (n=89), nursing staff and auxillary nurses (NSAN) (n=128). The mean age for the whole group of respondents was 40.7±0.7, professional experience lasted 12.0±0.6 years. Standard methods of descriptive statistics were used for the data analysis. The age characteristics of the group had no significant differences by professional categories. Doctors had 39.0±1.2 years (min — 24, max — 63), NSAN had 41.0±0.6 years (min — 19, max — 67). Professional experience lasted differed doctors 39.0±1.2 years, in NSAN — 12.3±0.8. Men were characterized by a mean age of 36.7±1.6 years and professional experience lasted 10.0±1.2 years and women 42.0±0.7 years and 12.6±0.7 years, respectively. The character of distribution of quantitative data in the presented study was analyzed using the Kolmogorov–Smirnov and Shapiro–Wilk criterion. RESULTS: 80.2% of respondents presented with a high level of reduced professional accomplishments (RPA), 53.4% — a moderate average level of depersonalization. The emotional exhaustion was similar in all respondents, although average level of emotional exhaustion was the most prevalent (39.6%). 43.8% of doctors had an average level of emotional exhaustion, which is higher compared with NSAN — 37.5%. Clinical and paraclinical units are associated with a medium level of emotional exhaustion and a high RPA. At the same time, paraclinical units are characterized by a medium level of depersonalization, in contrast to clinical units, which are found to have a high level of this factor. According to the respondents, possible measures to prevent emotional burnout should include a set of measures including, rational distribution of responsibilities within departments, workload among the staff, as well as the formation of special skills to overcome stressful situations and optimization of activities with the help of digital solutions (support system for medical decisions, etc.). CONCLUSION: Working as medical staff and providing medical care for patients with CSD is strongly associated with a high level of reduced professional accomplishments (80.2% of respondents). Proper measures for the prevention of emotional burnout should primarily include organizational solutions focused on optimizing internal processes in medical staff’ work. According to the respondents, among the measures aimed at the prevention of emotional burnout, the most demanded may be: rational distribution of responsibilities within departments, workload among staff, as well as the formation of special skills to overcome stressful situations and optimization of activities with the help of digital solutions (medical decision support system, etc.).

Improved Kepler Optimization Algorithm for enhanced feature selection in liver disease classification

Liver diseases represent a significant healthcare challenge, impacting millions globally and posing complexities in diagnosis. To address this global health concern, this paper introduces a groundbreaking enhancement to the Kepler Optimization Algorithm, termed I-KOA, designed specifically for feature selection in high-dimensional datasets. By harnessing the synergies of Opposition-Based Learning and a Local Escaping Operator grounded in the k-nearest Neighbor (kNN) classifier, I-KOA asserts itself as a potent tool for local exploitation, balanced exploration, and evasion of local optima. To our knowledge, this is the first work to exploit KOA as a feature selection method. Pioneering the utilization of KOA as a feature selection method, the paper rigorously tests I-KOA in two extensive experiments, tackling the complex CEC’22 benchmark suite functions and the intricate landscape of five liver disease datasets. Results underscore I-KOA’s unparalleled performance, validated through the Friedman test, where it surpasses seven rival optimization algorithms. Achieving an outstanding overall classification accuracy of 93.46%, Feature selection size of 0.1042, sensitivity of 97.46%, precision of 94.37%, and F1-score of 90.35% across the liver disease datasets, I-KOA’s randomized algorithm ensures robust feature selection, striking a compelling balance between subset size and classification efficacy. Acknowledging computational demands and generalization nuances, I-KOA is a formidable tool ready to revolutionize medical diagnosis and decision support systems. The open source codes of the proposed I-KOA are available at https://www.mathworks.com/matlabcentral/fileexchange/161376-improved-kepler-optimization-algorithm.

Deep Learning Approaches for Medical Image Analysis and Diagnosis.

In addition to enhancing diagnostic accuracy, deep learning techniques offer the potential to streamline workflows, reduce interpretation time, and ultimately improve patient outcomes. The scalability and adaptability of deep learning algorithms enable their deployment across diverse clinical settings, ranging from radiology departments to point-of-care facilities. Furthermore, ongoing research efforts focus on addressing the challenges of data heterogeneity, model interpretability, and regulatory compliance, paving the way forseamless integration of deep learning solutions into routine clinical practice. As the field continues to evolve, collaborations between clinicians, data scientists, and industry stakeholders will be paramount in harnessing the full potential of deep learning for advancing medical image analysis and diagnosis. Furthermore, the integration of deep learning algorithms with other technologies, including natural language processing and computer vision, may foster multimodal medical data analysis and clinical decision support systems to improve patient care. The future of deep learning in medical image analysis and diagnosis is promising. With each success and advancement, this technology is getting closer to being leveraged for medical purposes. Beyond medical image analysis, patient care pathways like multimodal imaging, imaging genomics, and intelligent operating rooms or intensive care units can benefitfrom deep learning models.

Multimodal Machine Learning in Image-Based and Clinical Biomedicine: Survey and Prospects

Machine learning (ML) applications in medical artificial intelligence (AI) systems have shifted from traditional and statistical methods to increasing application of deep learning models. This survey navigates the current landscape of multimodal ML, focusing on its profound impact on medical image analysis and clinical decision support systems. Emphasizing challenges and innovations in addressing multimodal representation, fusion, translation, alignment, and co-learning, the paper explores the transformative potential of multimodal models for clinical predictions. It also highlights the need for principled assessments and practical implementation of such models, bringing attention to the dynamics between decision support systems and healthcare providers and personnel. Despite advancements, challenges such as data biases and the scarcity of “big data” in many biomedical domains persist. We conclude with a discussion on principled innovation and collaborative efforts to further the mission of seamless integration of multimodal ML models into biomedical practice.

Multi-criteria decision analysis method for differential diagnosis of tropical febrile diseases.

This paper employs the Analytical Hierarchy Process (AHP) to enhance the accuracy of differential diagnosis for febrile diseases, particularly prevalent in tropical regions where misdiagnosis may have severe consequences. The migration of health workers from developing countries has resulted in frontline health workers (FHWs) using inadequate protocols for the diagnosis of complex health conditions. The study introduces an innovative AHP-based Medical Decision Support System (MDSS) incorporating disease risk factors derived from physicians' experiential knowledge to address this challenge. The system's aggregate diagnostic factor index determines the likelihood of febrile illnesses. Compared to existing literature, AHP models with risk factors demonstrate superior prediction accuracy, closely aligning with physicians' suspected diagnoses. The model's accuracy ranges from 85.4% to 96.9% for various diseases, surpassing physicians' predictions for Lassa, Dengue, and Yellow Fevers. The MDSS is recommended for use by FHWs in communities lacking medical experts, facilitating timely and precise diagnoses, efficient application of diagnostic test kits, and reducing overhead expenses for administrators.

A medical decision support system using an artificial intelligence module for endoscopic examination of the stomach

A medical decision support system using an artificial intelligence module for endoscopic examination of the stomach