To train the model, a database consisting of 10,840 unique sentences was extracted from all chest CT reports performed in 2021 (totalling 2001 reports) at the Radiology Service of the San Luigi Gonzaga University Teaching Hospital. 

Reports of CT studies that included other body parts, such as the abdomen and brain, were excluded.

These sentences were manually annotated with labels across three main categories: findings, anatomical sites, and study descriptors. 

The dataset was split into training and testing sets, and a BERT model pre-trained on Italian text (dbmdz/bert-base-italian-uncased) was fine-tuned for multi-label classification.