Current clinical approaches fail to fully utilize unstructured data in managing prostate cancer (PCa) and kidney cancer (KC), leading to inefficiencies in patient care and increased costs. Effective diagnostics and treatments depend on integrating multimodal data, yet progress is hampered by limited data accessibility and a lack of collaborative validation between clinicians and computer scientists. To address these challenges, the EU-funded COMFORT project aims to develop commercially viable, data-driven multimodal decision support systems. These systems will improve clinical prognostication, patient stratification, and personalized treatment while also assessing the trust that healthcare professionals and patients place in AI-driven tools.

OBJECTIVES: Large language models (LLMs) have shown potential in biomedical applications, leading to efforts to fine-tune them on domain-specific data. However, the effectiveness of this approach remains unclear. This study aims to critically evaluate the performance of biomedically fine-tuned LLMs against their general-purpose counterparts across a range of clinical tasks. MATERIALS AND METHODS: We evaluated the performance of biomedically fine-tuned LLMs against their general-purpose counterparts on clinical case challenges from NEJM and JAMA, and on multiple clinical tasks, such as information extraction, document summarization and clinical coding. We used a diverse set of benchmarks specifically chosen to be outside the likely fine-tuning datasets of biomedical models, ensuring a fair assessment of generalization capabilities. RESULTS: Biomedical LLMs generally underperformed compared to general-purpose models, especially on tasks not focused on probing medical knowledge. While on the case challenges, larger biomedical and general-purpose models showed similar performance (eg, OpenBioLLM-70B: 66.4% vs Llama-3-70B-Instruct: 65% on JAMA), smaller biomedical models showed more pronounced underperformance (OpenBioLLM-8B: 30% vs Llama-3-8B-Instruct: 64.3% on NEJM). Similar trends appeared across CLUE benchmarks, with general-purpose models often achieving higher scores in text generation, question answering, and coding. Notably, biomedical LLMs also showed a higher tendency to hallucinate. DISCUSSION: Our findings challenge the assumption that biomedical fine-tuning inherently improves LLM performance, as general-purpose models consistently performed better on unseen medical tasks. Retrieval-augmented generation may offer a more effective strategy for clinical adaptation. CONCLUSION: Fine-tuning LLMs on biomedical data may not yield the anticipated benefits. Alternative approaches, such as retrieval augmentation, should be further explored for effective and reliable clinical integration of LLMs.

This paper presents medBERT.de , a pre-trained German BERT model specifically designed for the German medical domain. The model has been trained on a large corpus of 4.7 Million German medical documents and has been shown to achieve new state-of-the-art performance on eight different medical benchmarks covering a wide range of disciplines and medical document types. In addition to evaluating the overall performance of the model, this paper also conducts a more in-depth analysis of its capabilities. We investigate the impact of data deduplication on the model's performance, as well as the potential benefits of using more efficient tokenization methods. Our results indicate that domain-specific models such as medBERT.de are particularly useful for longer texts, and that deduplication of training data does not necessarily lead to improved performance. Furthermore, we found that efficient tokenization plays only a minor role in improving model performance, and attribute most of the improved performance to the large amount of training data. To encourage further research, the pre-trained model weights and new benchmarks based on radiological data are made publicly available for use by the scientific community.

MOTIVATION: The development of deep, bidirectional transformers such as Bidirectional Encoder Representations from Transformers (BERT) led to an outperformance of several Natural Language Processing (NLP) benchmarks. Especially in radiology, large amounts of free-text data are generated in daily clinical workflow. These report texts could be of particular use for the generation of labels in machine learning, especially for image classification. However, as report texts are mostly unstructured, advanced NLP methods are needed to enable accurate text classification. While neural networks can be used for this purpose, they must first be trained on large amounts of manually labelled data to achieve good results. In contrast, BERT models can be pre-trained on unlabelled data and then only require fine tuning on a small amount of manually labelled data to achieve even better results. RESULTS: Using BERT to identify the most important findings in intensive care chest radiograph reports, we achieve areas under the receiver operation characteristics curve of 0.98 for congestion, 0.97 for effusion, 0.97 for consolidation and 0.99 for pneumothorax, surpassing the accuracy of previous approaches with comparatively little annotation effort. Our approach could therefore help to improve information extraction from free-text medical reports. Availability and implementationWe make the source code for fine-tuning the BERT-models freely available at https://github.com/fast-raidiology/bert-for-radiology. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.