LusoClin: Dataset of synthetic clinical notes in European Portuguese generated using an open-source large language model, along with prompting and evaluation data

The dataset can also be used for related information extraction tasks—such as named entity recognition, document classification, and summarization—which often serve as components of clinical IR pipelines. In addition, LusoClin provides a foundation for training, fine-tuning, and evaluating domain-adapted language and retrieval models for Portuguese clinical text. **About the dataset:** XML files comprising 54,518 fully synthetic clinical notes in European Portuguese, divided into 4 types: 14,253 admission notes, 13,561 ambulatory notes, 9,032 discharge summaries, and 17,672 nursing notes; CSV file with prompts and responses from prompt engineering; CSV files with prompts and responses from synthetic dataset generation; TXT files containing 1,000 clinical notes (250 of each type) taken from the synthetic dataset and used during automatic evaluation; TXT files detailing the improvements made to LusoClin's old version, which led to its current version.