Synthetic data offers a compelling solution to the challenges associated with acquiring high-quality medical data, which is often constrained by privacy concerns and limited accessibility. This study explores the efficacy of synthetic data generated using diffusion models for training deep learning models within a self-supervised learning framework. The primary objective is to evaluate whether synthetic data can effectively preserve critical medical biomarkers and support reliable downstream tasks such as classification and segmentation. Using chest X-ray images as a case study, the results reveal that models pretrained on synthetic data achieve performance comparable to or surpassing those pretrained on real data. Specifically, in pneumonia classification tasks, the model trained on synthetic data outperformed established benchmarks, achieving an Area Under the Curve (AUC) of 99.1 and an F1-score of 96.1%. Similarly, for segmentation tasks, the model trained on synthetic data demonstrated robust performance, attaining a Dice score of 0.85. These findings underscore a significant advancement in the generation of synthetic medical images, providing a viable approach to creating realistic, biomarker-preserving datasets that ensure patient confidentiality and enable diverse applications in medical imaging.

Objectives: Medical imaging is indispensable for diagnosis, with abdominal imaging playing a pivotal role in generating medical reports and informing clinical decision-making. Recent works in artificial intelligence (AI), particularly in multimodal approaches such as vision-language models, have demonstrated significant potential to enhance medical image analysis by seamlessly integrating visual and textual data. While 2D imaging has been the main focus of many studies, the enhanced spatial detail and volumetric consistency offered by 3D images, such as CT scans, remain relatively underexplored. This gap underscores the need for innovative approaches to unlock the potential of 3D imaging in clinical workflows. Methods: In this study, we utilized a multimodal AI pipeline, Phi3-V, to address 2 key challenges in abdominal imaging: generating clinically coherent medical reports from 3D CT images and performing visual question answering based on these images. Results: Our optimized model attained an average GREEN score of 0.409 for medical report generation and an accuracy of 79% for multiple-choice visual question answering on the validation cases. Conclusions: These findings demonstrate the potential of multimodal AI in advancing the analysis of 3D medical imaging, paving the way for more robust and efficient applications in healthcare. Advances in knowledge: This study advances the use of multimodal AI for 3D CT imaging, achieving improvements in medical report generation and visual question answering.

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.

Dialectal variation presents a major challenge for deploying medical language models in real-world healthcare settings, where patient–clinician communication often occurs in regional vernaculars rather than standardized language forms. This challenge is particularly pronounced in the Arabic-speaking world, where clinical interactions frequently take place in diverse dialects that differ substantially from Modern Standard Arabic. Fine-tuning and maintaining separate models for each dialect is computationally inefficient and difficult to scale, motivating more integrated approaches. In this work, we present MENARA, an Arabic medical language model constructed by merging Egyptian Arabic, Moroccan Darija, and medical-domain specialists through model merging. We extend prior feasibility findings through comprehensive evaluation of cross-dialect performance, medical safety, and cross-lingual knowledge retention. Specifically, we introduce a fine-grained dialect composition analysis to quantify lexical purity and structured code-switching behavior, benchmark against state-of-the-art Arabic LLMs, conduct subject-matter-expert assessment of both dialectal fidelity and medical appropriateness. The results show that model merging preserves core medical competence while enabling robust dialectal adaptation, achieving strong cross-dialect fidelity while substantially reducing storage and deployment overhead compared to maintaining separate models. These findings establish model merging as a potentially practical and resource-efficient paradigm for dialect-aware medical NLP in linguistically fragmented healthcare environments.

Clinical reasoning over 3D CT scans is inherently compositional, requiring the integration of anatomical measurement, pathology assessment, spatial comparison, and clinical interpretation. We introduce MedToolica, a finetuning-free, role-based agentic framework for quantitative 3D abdominal CT reasoning that decomposes complex queries into structured sub-tasks coordinated through specialized expert tools. Empirical evaluation across quantitative reasoning benchmarks demonstrates that MedToolica is particularly effective in organ-centric measurement tasks when supported by reliable expert tools, achieving strong quantitative agreement (e.g., $CCC=0.99$ for organ HU estimation versus $0.46$ for finetuned baselines) and notable gains on multi-step visual reasoning tasks. In contrast, lesion-oriented tasks remain constrained by upstream tool limitations, indicating that reasoning sophistication alone cannot compensate for unreliable perception. Furthermore, we observe that the capability of the core language model substantially influences orchestration quality: smaller LLM orchestrators exhibit reduced overall accuracy due to higher execution failure rates ($25\%$ vs. $79\%$) and increased susceptibility to hallucination ($43\%$ vs. $2\%$). Collectively, these findings identify expert tool reliability and orchestration capability as critical determinants of performance in compositional medical AI and highlight both the promise and current limitations of finetuning-free agentic reasoning for quantitative 3D CT analysis.