Research Appraisals
Evidence-based critical appraisals of the latest medical research, systematically evaluated using Oxford CEBM methodology.
Showing 3 appraisals
Studies in health technology and informatics
Leveraging Large Language Models with Retrieval-Augmented Generation for Semantic Mapping of Clinical Data Lakes to SNOMED CT
Mapping local clinical concepts to standardized terminologies such as SNOMED CT is essential for semantic interoperability and large-scale research, but manual mapping is labor-intensive and difficult to scale. We report a preliminary evaluation of a hybrid approach combining retrieval-augmented generation (RAG) and a large language model (LLM) to support the mapping of hospital datalake concepts to SNOMED CT. Using a dataset of 2,768 concepts annotated by the SIMED across five semantic categories (organisms, healthcare locations, laboratory samples, allergies, and clinical assessments), the pipeline consistently outperformed retrieval-only baselines. For organism concepts, strict accuracy reached 92.8% (weighted 98.0%, top-3 96.7%), while allergy and assessment concepts achieved 84.7% and 72.1% strict accuracy, respectively. Healthcare location and laboratory sample concepts remained more challenging, with strict accuracies of 57.5% and 57.1%, and weighted accuracy varied strongly depending on frequency distributions. Overall, strict accuracy was 72.8%, weighted accuracy 76.1%, and top-3 accuracy 83.1%. These findings suggest that combining RAG with LLM-based reasoning can reduce expert workload while maintaining accuracy, representing a step toward scalable, AI-assisted semantic interoperability in healthcare. The approach relies on a retrieval-augmented generation pipeline using embedding-based candidate retrieval followed by LLM-based disambiguation (GPT-4.1-mini), and is evaluated against a retrieval-only baseline. By improving semantic consistency in clinical data lakes, it indirectly supports interoperable access to patient-level health data.
24 May 2026
Read appraisal →Studies in health technology and informatics
Large Language Models for Health Knowledge Modelling in Data Interoperability: A Scoping Review of Methods, Standards, and Applications
Achieving interoperability in healthcare requires semantic alignment of data with standards such as SNOMED CT, LOINC, and FHIR. Manual ontology mapping is slow and inconsistent. Evolving Large Language Models (LLMs) can automate knowledge modelling and data harmonisation. This review explores how LLMs are applied to enhance or automate health knowledge modelling for data interoperability. Following PRISMA-ScR guidance, PubMed and Compendex were searched. Studies applying LLMs to ontology alignment, terminology mapping, schema integration, or knowledge-graph construction were included. From 166 records, 20 studies met the inclusion criteria. Most used GPT-4 or BERT-derived models with retrieval-augmented or embedding-based methods. Reported gains included automation, improved mapping precision, and scalability across standards such as FHIR, OMOP, and UMLS. LLMs show strong potential for automating semantic interoperability in health informatics. Standardised evaluation benchmarks and explainable frameworks are essential for trustworthy adoption.
24 May 2026
Read appraisal →Artificial intelligence in medicine
TADynFed: Dynamic modality-adaptive federated learning with tissue-aware disentanglement for cross-disease analysis
Federated learning (FL) enables collaborative medical image analysis across decentralized institutions while preserving data privacy. However, real-world deployment faces critical challenges: modality heterogeneity, where clients possess incomplete or varying medical image modalities, and cross-disease generalization, requiring models to adapt to unseen pathologies with anatomical consistency. Existing methods like FedAvg, FedProx, IOP-FL, and PntTranForFL assume uniform modality availability and static client participation, leading to poor performance under realistic clinical constraints. We propose TADynFed, a novel framework for the Heterogeneous Federated Learning (HFL) paradigm, which addresses both data and modality heterogeneity through, a tissue-aware disentanglement strategy that decouples modality-tailored and modality-shared features, a dynamic prototype memory bank for missing modality compensation, an adaptive aggregation mechanism that accounts for client reliability and tenure. We evaluate TADynFed using multi-disease MRI datasets including BraTS21 along with cross-domain imaging data from CheXpert (chest X-ray) and Hep-2 (microscopy), simulating a 13-client FL environment. The proposed framework achieves an average mDice score of 66.03%, significantly outperforming baseline methods such as PointTransformerFL at 58.60% and FedAvg at 53.06%. It also attains the lowest boundary alignment error with an ASD of 1.85 mm and HD95 of 8.70 mm, surpassing existing approaches by notable margins. In terms of calibration stability, TADynFed records an ECE of 0.09, indicating superior confidence reliability. Furthermore, it demonstrates high communication efficiency with only 76 MB of data exchanged per round, compared to 95-105 MB in other frameworks. These results validate TADynFed's ability to maintain high segmentation accuracy, boundary precision, and calibration stability while minimizing bandwidth usage. Outperforming existing frameworks by significant margins, TADynFed demonstrates robust boundary alignment, superior calibration, and efficient communication. It also exhibits strong cross-disease transferability without retraining. By integrating structured representation decomposition, prototype-guided fusion, and client-adaptive learning, TADynFed establishes a new benchmark in realistic, heterogeneous, and mix-modal federated medical imaging systems. Code supporting this study TADynFed.
5 May 2026
Read appraisal →