arXiv subjects

Guido Bertolini

Publications and source records attributed to Guido Bertolini.

3 recordsLinked to original sources

arXiv2026

eCREAM-MedCorpus A Large-Scale Corpus of Clinical Notes for Italian

We present eCREAM-MedCorpus, a new and unique large-scale dataset of clinical notes produced in Emergency Departments of Italian hospitals. The corpus, in its current version, is composed of approximately 4 million clinical notes fully anonymized, covering diverse phases of patient care during the stay in the emergency department. In addition, a subset of about six thousand notes has been manually annotated by clinical experts through a structured Case Report Form (CRF) containing 132 items relevant for two patient situations in emergency departments, dyspnea and loss of consciousness. Items may assume numerical values (e.g., for blood saturation), categorical (e.g., for level of consciousness ), binary (e.g., for presence of traumas), and mixed value types. The annotation process involved multiple clinicians and underwent iterative revision to resolve ambiguities in item formulation, resulting in a richly structured (although high imbalanced) resource. The dataset aims to fill a relevant gap of data able to support both the development and the use of Large Language Models in concrete medical applications. We describe the data collection protocol, the on-site anonymisation pipeline, corpus statistics, and the annotation scheme. Finally, we propose CRF-filling as a novel structured information extraction benchmark, and provide zero-shot baseline resulting from Gemma-27B and MedGemma-27B. To the best of our knowledge, eCREAM-MedCorpus is the largest freely available dataset of clinical notes existing for the Italian language.

cs.CL↗

arXiv2022

Automatic lesion analysis for increased efficiency in outcome prediction of traumatic brain injury

The accurate prognosis for traumatic brain injury (TBI) patients is difficult yet essential to inform therapy, patient management, and long-term after-care. Patient characteristics such as age, motor and pupil responsiveness, hypoxia and hypotension, and radiological findings on computed tomography (CT), have been identified as important variables for TBI outcome prediction. CT is the acute imaging modality of choice in clinical practice because of its acquisition speed and widespread availability. However, this modality is mainly used for qualitative and semi-quantitative assessment, such as the Marshall scoring system, which is prone to subjectivity and human errors. This work explores the predictive power of imaging biomarkers extracted from routinely-acquired hospital admission CT scans using a state-of-the-art, deep learning TBI lesion segmentation method. We use lesion volumes and corresponding lesion statistics as inputs for an extended TBI outcome prediction model. We compare the predictive power of our proposed features to the Marshall score, independently and when paired with classic TBI biomarkers. We find that automatically extracted quantitative CT features perform similarly or better than the Marshall score in predicting unfavourable TBI outcomes. Leveraging automatic atlas alignment, we also identify frontal extra-axial lesions as important indicators of poor outcome. Our work may contribute to a better understanding of TBI, and provides new insights into how automated neuroimaging analysis can be used to improve prognostication after TBI.

cs.CV↗

arXiv2022

A predictive model for planning emergency events rescue during COVID-19 in Lombardy, Italy

Italy, particularly the Lombardy region, was among the first countries outside of Asia to report cases of COVID-19. The emergency medical service called Regional Emergency Agency (AREU) coordinates the intra- and inter-regional non-hospital emergency network and the European emergency number service in Lombardy. AREU must deal with daily and seasonal variations of call volume. The number and type of emergency calls changed dramatically during the COVID-19 pandemic. A model to predict incoming calls and how many of these turn into events, i.e., dispatch of transport and equipment until the rescue is completed, was developed to address the emergency period. We used the generalized additive model with a negative binomial family to predict the number of events one, two, five, and seven days ahead. The over-dispersion of the data was tackled by using the negative binomial family and the nonlinear relationship between the number of events and covariates (e.g., seasonal effects) by smoothing splines. The model coefficients show the effect of variables, e.g., the day of the week, on the number of events and how these effects change during the pre-COVID-19 period. The proposed model returns reasonable mean absolute errors for most of the 2020-2021 period.

stat.AP↗