Search Results (18)

Objectives: This study aimed to evaluate the performance of four AI models, (GPT-5, GPT-4, Microsoft Copilot, and Google Gemini), in detecting medication errors through pediatric case scenarios. Methods: A total of 60 pediatric cases were analyzed for the presence of medication errors, of which only half contained errors. The cases covered four therapeutic systems (respiratory, endocrine, neurology, and infectious). The four models were exposed to the cases in both English and Arabic using a unified prompt. The responses for each model were used to calculate various performance metric cover accuracy, sensitivity, specificity and reproducibility. Analysis was carried out using SPSS version 22. Results: Microsoft Copilot demonstrated relatively higher accuracy (86.7% in English, 85.0% in Arabic) compared to other models in this dataset, followed by GPT-5 (81.7% in English, 75.0% in Arabic). GPT-4 and Google Gemini had less accuracy, with Gemini having the lowest accuracy across all languages (76.7% in English, and 73.3% in Arabic). Microsoft Copilot showed comparatively higher sensitivity and specificity, particularly in cases of respiratory and infectious diseases. The accuracy in Arabic was lower compared to that of English for the majority of models. Microsoft Copilot exhibited relatively higher reproducibility and inter-run agreement (Cohen’s Kappa = 0.836 English, 0.815 Arabic, p < 0.001 for both), while Gemini showed the lowest reproducibility. For inter-language agreement in general, Copilot showed the highest Cohen’s Kappa of 0.701 for English and Arabic (p < 0.001). Conclusions: In our evaluation, Microsoft Copilot demonstrated relatively higher performance in pediatric drug error detection compared to the other AI models. The decreased performance in Arabic points toward the requirement of improved multilingual training for supporting equal AI aid across languages. This study highlights the importance of human oversight and domain-based training for AI tools in pediatric pharmacotherapy. Full article

Advancing health equity requires rigorous analysis of how research initiatives incorporate and address structural disparities across populations. In this study, we apply large language models (LLMs) to systematically analyze research projects registered on the All of Us platform, with a focus on identifying patterns and institutional dynamics associated with health equity research. We examine the relationship between projects that explicitly pursue health equity goals and their use of available demographic data, their institutional composition (e.g., single- vs. multi-institutional teams), and the research tier of participating institutions (R1 vs. R2). Using the capabilities of an established LLM, we automate key tasks including the extraction of relevant attributes from unstructured project descriptions, classification of institutional affiliations, and the summarization of project content into standardized keywords from the Unified Medical Language System vocabulary. This LLM-assisted pipeline enabled scalable, replicable analysis of hundreds of projects with minimal manual overhead. Our findings suggest a strong association between the use of demographic data and health equity aims, and indicate nuanced differences in equity-oriented research participation by institution type and collaborative structure. More broadly, our approach demonstrates how LLMs can support equity-focused computational social science by transforming free-text administrative data into analyzable structures, enabling novel insights in public health, team science, and science-of-science studies. Full article

Objective: Medical Visual Question Answering (Med-VQA), a key technology that integrates computer vision and natural language processing to assist in clinical diagnosis, possesses significant potential for enhancing diagnostic efficiency and accuracy. However, its development is constrained by two major bottlenecks: weak few-shot generalization capability stemming from the scarcity of high-quality annotated data and the problem of catastrophic forgetting when continually learning new knowledge. Existing research has largely addressed these two challenges in isolation, lacking a unified framework. Methods: To bridge this gap, this paper proposes a novel Evolvable Clinical-Semantic Alignment (ECSA) framework, designed to synergistically solve these two challenges within a single architecture. ECSA is built upon powerful pre-trained vision (BiomedCLIP) and language (Flan-T5) models, with two innovative modules at its core. First, we design a Clinical-Semantic Disambiguation Module (CSDM), which employs a novel debiased hard negative mining strategy for contrastive learning. This enables the precise discrimination of “hard negatives” that are visually similar but clinically distinct, thereby significantly enhancing the model’s representation ability in few-shot and long-tail scenarios. Second, we introduce a Prompt-based Knowledge Consolidation Module (PKC), which acts as a rehearsal-free non-parametric knowledge store. It consolidates historical knowledge by dynamically accumulating and retrieving task-specific “soft prompts,” thus effectively circumventing catastrophic forgetting without relying on past data. Results: Extensive experimental results on four public benchmark datasets, VQA-RAD, SLAKE, PathVQA, and VQA-Med-2019, demonstrate ECSA’s state-of-the-art or highly competitive performance. Specifically, ECSA achieves excellent overall accuracies of 80.15% on VQA-RAD and 85.10% on SLAKE, while also showing strong generalization with 64.57% on PathVQA and 82.23% on VQA-Med-2019. More critically, in continual learning scenarios, the framework achieves a low forgetting rate of just 13.50%, showcasing its significant advantages in knowledge retention. Conclusions: These findings validate the framework’s substantial potential for building robust and evolvable clinical decision support systems. Full article

Deep learning (DL) has revolutionized medical image analysis (MIA), enabling early anomaly detection, precise lesion segmentation, and automated disease classification. However, its clinical integration faces two major challenges: reliance on limited, narrowly annotated datasets that inadequately capture real-world patient diversity, and the inherent “black-box” nature of DL decision-making, which complicates physician scrutiny and accountability. Eye tracking (ET) technology offers a transformative solution by capturing radiologists’ gaze patterns to generate supervisory signals. These signals enhance DL models through two key mechanisms: providing weak supervision to improve feature recognition and diagnostic accuracy, particularly when labeled data are scarce, and enabling direct comparison between machine and human attention to bridge interpretability gaps and build clinician trust. This approach also extends effectively to multimodal learning models (MLMs) and vision–language models (VLMs), supporting the alignment of machine reasoning with clinical expertise by grounding visual observations in diagnostic context, refining attention mechanisms, and validating complex decision pathways. Conducted in accordance with the PRISMA statement and registered in PROSPERO (ID: CRD42024569630), this review synthesizes state-of-the-art strategies for ET-DL integration. We further propose a unified framework in which ET innovatively serves as a data efficiency optimizer, a model interpretability validator, and a multimodal alignment supervisor. This framework paves the way for clinician-centered AI systems that prioritize verifiable reasoning, seamless workflow integration, and intelligible performance, thereby addressing key implementation barriers and outlining a path for future clinical deployment. Full article

Reinforcement learning (RL) is being used more in medical imaging for segmentation, detection, registration, and classification. This survey provides a comprehensive overview of RL techniques applied in this domain, categorizing the literature based on clinical task, imaging modality, learning paradigm, and algorithmic design. We introduce a unified taxonomy that supports reproducibility, highlights design guidance, and identifies underexplored intersections. Furthermore, we examine the integration of Large Language Models (LLMs) for automation and interpretability, and discuss privacy-preserving extensions using Differential Privacy (DP) and Federated Learning (FL). Finally, we address deployment challenges and outline future research directions toward trustworthy and scalable medical RL systems. Full article

The increasing complexity and scale of electronic health records (EHRs) demand advanced tools for efficient data retrieval, summarization, and comparative analysis in clinical practice. MERA (Medical Electronic Records Assistant) is a Retrieval-Augmented Generation (RAG)-based AI system that addresses these needs by integrating domain-specific retrieval with large language models (LLMs) to deliver robust question answering, similarity search, and report summarization functionalities. MERA is designed to overcome key limitations of conventional LLMs in healthcare, such as hallucinations, outdated knowledge, and limited explainability. To ensure both privacy compliance and model robustness, we constructed a large synthetic dataset using state-of-the-art LLMs, including Mistral v0.3, Qwen 2.5, and Llama 3, and further validated MERA on de-identified real-world EHRs from the MIMIC-IV-Note dataset. Comprehensive evaluation demonstrates MERA’s high accuracy in medical question answering (correctness: 0.91; relevance: 0.98; groundedness: 0.89; retrieval relevance: 0.92), strong summarization performance (ROUGE-1 F1-score: 0.70; Jaccard similarity: 0.73), and effective similarity search (METEOR: 0.7–1.0 across diagnoses), with consistent results on real EHRs. The similarity search module empowers clinicians to efficiently identify and compare analogous patient cases, supporting differential diagnosis and personalized treatment planning. By generating concise, contextually relevant, and explainable insights, MERA reduces clinician workload and enhances decision-making. To our knowledge, this is the first system to integrate clinical question answering, summarization, and similarity search within a unified RAG-based framework. Full article

The integration of artificial intelligence into healthcare has advanced rapidly in recent years, with multimodal approaches emerging as promising tools for improving diagnostic accuracy and clinical decision making. These approaches combine heterogeneous data sources such as medical images, electronic health records, physiological signals, and clinical notes to better capture the complexity of disease processes. Despite this progress, only a limited number of studies offer a unified view of multimodal AI applications in medicine. In this review, we provide a comprehensive and up-to-date analysis of machine learning and deep learning-based multimodal architectures, fusion strategies, and their performance across a range of diagnostic tasks. We begin by summarizing publicly available datasets and examining the preprocessing pipelines required for harmonizing heterogeneous medical data. We then categorize key fusion strategies used to integrate information from multiple modalities and overview representative model architectures, from hybrid designs and transformer-based vision-language models to optimization-driven and EHR-centric frameworks. Finally, we highlight the challenges present in existing works. Our analysis shows that multimodal approaches tend to outperform unimodal systems in diagnostic performance, robustness, and generalization. This review provides a unified view of the field and opens up future research directions aimed at building clinically usable, interpretable, and scalable multimodal diagnostic systems. Full article

Understanding and interpreting medical images, particularly radiology images, is a time-consuming task that requires specialized expertise. In this study, we developed a deep learning-based system capable of automatically assigning multiple standardized medical concepts to radiology images, leveraging deep learning models. These concepts are based on Unified Medical Language System (UMLS) Concept Unique Identifiers (CUIs) and describe the radiology images in detail. Each image is associated with multiple concepts, making it a multilabel classification problem. We implemented several deep learning models, including DenseNet121, ResNet101, and VGG19, and evaluated them on the ImageCLEF 2020 Medical Concept Detection dataset. This dataset consists of radiology images with multiple CUIs associated with each image and is organized into seven categories based on their modality information. In this study, transfer learning techniques were applied, with the models initially pre-trained on the ImageNet dataset and subsequently fine-tuned on the ImageCLEF dataset. We present the evaluation results based on the F1-score metric, demonstrating the effectiveness of our approach. Our best-performing model, DenseNet121, achieved an F1-score of 0.89 on the classification of the twenty most frequent medical concepts, indicating a significant improvement over baseline methods. Full article

Background: Lumbar foraminal stenosis (LFS) involves the narrowing of neural foramina, leading to nerve compression, significant lower back pain and radiculopathy, particularly in the aging population. Management includes physical therapy, medications and potentially invasive surgeries such as foraminotomy. Advances in diagnostic and treatment strategies are essential due to LFS’s complexity and prevalence, which underscores the importance of a multidisciplinary approach in optimizing patient outcomes. Method: This literature review on LFS employed a systematic methodology to gather and synthesize recent scientific data. A comprehensive search was conducted across PubMed, Scopus and Cochrane Library databases using specific keywords related to LFS. The search, restricted to English language articles from 1 January 2000 to 31 December 2023, focused on peer-reviewed articles, clinical trials and reviews. Due to the heterogeneity among the studies, data were qualitatively synthesized into themes related to diagnosis, treatment and pathophysiology. Results: This literature review on LFS analyzed 972 articles initially identified, from which 540 remained after removing duplicates. Following a rigorous screening process, 20 peer-reviewed articles met the inclusion criteria and were reviewed. These studies primarily focused on evaluating the diagnostic accuracy, treatment efficacy and pathophysiological insights into LFS. Conclusion: The comprehensive review underscores the necessity for precise diagnostic and management strategies for LFS, highlighting the role of a multidisciplinary approach and the utility of a unified classification system in enhancing patient outcomes in the face of this condition’s increasing prevalence. Full article

In recent years, Convolutional Neural Network (CNN) models have demonstrated notable advancements in various domains such as image classification and Natural Language Processing (NLP). Despite their success in image classification tasks, their potential impact on medical image retrieval, particularly in text-based medical image retrieval (TBMIR) tasks, has not yet been fully realized. This could be attributed to the complexity of the ranking process, as there is ambiguity in treating TBMIR as an image retrieval task rather than a traditional information retrieval or NLP task. To address this gap, our paper proposes a novel approach to re-ranking medical images using a Deep Matching Model (DMM) and Medical-Dependent Features (MDF). These features incorporate categorical attributes such as medical terminologies and imaging modalities. Specifically, our DMM aims to generate effective representations for query and image metadata using a personalized CNN, facilitating matching between these representations. By using MDF, a semantic similarity matrix based on Unified Medical Language System (UMLS) meta-thesaurus, and a set of personalized filters taking into account some ranking features, our deep matching model can effectively consider the TBMIR task as an image retrieval task, as previously mentioned. To evaluate our approach, we performed experiments on the medical ImageCLEF datasets from 2009 to 2012. The experimental results show that the proposed model significantly enhances image retrieval performance compared to the baseline and state-of-the-art approaches. Full article

This research proposes a new approach to improve information retrieval systems based on a multinomial naive Bayes classifier (MNBC), Bayesian networks (BNs), and a multi-terminology which includes MeSH thesaurus (Medical Subject Headings) and SNOMED CT (Systematized Nomenclature of Medicine of Clinical Terms). Our approach, which is entitled improving semantic information retrieval (IMSIR), extracts and disambiguates concepts and retrieves documents. Relevant concepts of ambiguous terms were selected using probability measures and biomedical terminologies. Concepts are also extracted using an MNBC. The UMLS (Unified Medical Language System) thesaurus was then used to filter and rank concepts. Finally, we exploited a Bayesian network to match documents and queries using a conceptual representation. Our main contribution in this paper is to combine a supervised method (MNBC) and an unsupervised method (BN) to extract concepts from documents and queries. We also propose filtering the extracted concepts in order to keep relevant ones. Experiments of IMSIR using the two corpora, the OHSUMED corpus and the Clinical Trial (CT) corpus, were interesting because their results outperformed those of the baseline: the P@50 improvement rate was +36.5% over the baseline when the CT corpus was used. Full article

The human body’s reaction to various therapeutic medications is critical to comprehend since it aids in the appropriate construction of automated decision support systems for healthcare. Healthcare Internet of Things (IoT) solutions are becoming more accessible and trusted, necessitating more testing before they are standardized for commercial usage. We have developed an activity diagram based on the Unified Modeling Language (UML) to represent acceptability testing in IoT systems. The activity flow graph is used to extract all of the necessary information by traversing the activity flow diagram from start to finish, displaying all its properties. In this paper, a test case is generated to compute the type of diabetes using blood sugar test results, estimate the kind of diabetes, and the probability that a person would get diabetes in the future. We have demonstrated how these test cases can function using a telehealth care case study. First, we offer a high-level overview of the topic as well as a design model working diagram. The test case creation method is then outlined using the activity diagram as a guide. Full article

(1) Background: Aspect-based sentiment analysis (SA) is a natural language processing task, the aim of which is to classify the sentiment associated with a specific aspect of a written text. The performance of SA methods applied to texts related to health and well-being lags behind that of other domains. (2) Methods: In this study, we present an approach to aspect-based SA of drug reviews. Specifically, we analysed signs and symptoms, which were extracted automatically using the Unified Medical Language System. This information was then passed onto the BERT language model, which was extended by two layers to fine-tune the model for aspect-based SA. The interpretability of the model was analysed using an axiomatic attribution method. We performed a correlation analysis between the attribution scores and syntactic dependencies. (3) Results: Our fine-tuned model achieved accuracy of approximately $95\%$ on a well-balanced test set. It outperformed our previous approach, which used syntactic information to guide the operation of a neural network and achieved an accuracy of approximately $82\%$. (4) Conclusions: We demonstrated that a BERT-based model of SA overcomes the negative bias associated with health-related aspects and closes the performance gap against the state-of-the-art in other domains. Full article

Visual Question Generation (VQG) from images is a rising research topic in both fields of natural language processing and computer vision. Although there are some recent efforts towards generating questions from images in the open domain, the VQG task in the medical domain has not been well-studied so far due to the lack of labeled data. In this paper, we introduce a goal-driven VQG approach for radiology images called VQGRaD that generates questions targeting specific image aspects such as modality and abnormality. In particular, we study generating natural language questions based on the visual content of the image and on additional information such as the image caption and the question category. VQGRaD encodes the dense vectors of different inputs into two latent spaces, which allows generating, for a specific question category, relevant questions about the images, with or without their captions. We also explore the impact of domain knowledge incorporation (e.g., medical entities and semantic types) and data augmentation techniques on visual question generation in the medical domain. Experiments performed on the VQA-RAD dataset of clinical visual questions showed that VQGRaD achieves 61.86% BLEU score and outperforms strong baselines. We also performed a blinded human evaluation of the grammaticality, fluency, and relevance of the generated questions. The human evaluation demonstrated the better quality of VQGRaD outputs and showed that incorporating medical entities improves the quality of the generated questions. Using the test data and evaluation process of the ImageCLEF 2020 VQA-Med challenge, we found that relying on the proposed data augmentation technique to generate new training samples by applying different kinds of transformations, can mitigate the lack of data, avoid overfitting, and bring a substantial improvement in medical VQG. Full article

Clinical Practice Guidelines (CPGs) aim to optimize patient care by assisting physicians during the decision-making process. However, guideline adherence is highly affected by its unstructured format and aggregation of background information with disease-specific information. The objective of our study is to extract disease-specific information from CPG for enhancing its adherence ratio. In this research, we propose a semi-automatic mechanism for extracting disease-specific information from CPGs using pattern-matching techniques. We apply supervised and unsupervised machine-learning algorithms on CPG to extract a list of salient terms contributing to distinguishing recommendation sentences (RS) from non-recommendation sentences (NRS). Simultaneously, a group of experts also analyzes the same CPG and extract the initial patterns “Heuristic Patterns” using a group decision-making method, nominal group technique (NGT). We provide the list of salient terms to the experts and ask them to refine their extracted patterns. The experts refine patterns considering the provided salient terms. The extracted heuristic patterns depend on specific terms and suffer from the specialization problem due to synonymy and polysemy. Therefore, we generalize the heuristic patterns to part-of-speech (POS) patterns and unified medical language system (UMLS) patterns, which make the proposed method generalize for all types of CPGs. We evaluated the initial extracted patterns on asthma, rhinosinusitis, and hypertension guidelines with the accuracy of 76.92%, 84.63%, and 89.16%, respectively. The accuracy increased to 78.89%, 85.32%, and 92.07% with refined machine-learning assistive patterns, respectively. Our system assists physicians by locating disease-specific information in the CPGs, which enhances the physicians’ performance and reduces CPG processing time. Additionally, it is beneficial in CPGs content annotation. Full article