On the Ethics of AI-based Algorithmic decision-making in Healthcare.

Assessing Trustworthy AI in times of COVID-19: Deep Learning for predicting a multi-regional score conveying the degree of lung compromise in COVID-19 patients.

In cooperation with

Department of Information Engineering and Department of Medical and Surgical Specialties, Radiological Sciences, and Public Health – University of Brescia, Brescia, Italy


Worldwide, the saturation of healthcare facilities, due to the high contagiousness of Sars-Cov-2 virus and the significant rate of respiratory complications is indeed one among the most critical aspects of the ongoing COVID-19 pandemic

The team of Alberto Signoroni, Davide Farina and colleagues implemented an end-to-end deep learning architecture, designed for predicting, on Chest X-rays images (CXR), a multi-regional score conveying the degree of lung compromise in COVID-19 patients.

The AI system has been experimentally deployed in the radiology department of the Brescia Public Hospital (ASST Spedali Civili) , Italy, since December 2020 during pandemic time.

We investigated the research question of what does  “trustworthy AI” mean at the time of the COVID-19 pandemic.

We started in April 2021 and conducted a Trustworthy AI self-assessment together with the key stakeholders, Alberto Signoroni (Department of Information Engineering), Davide Farina (Department of Medical and Surgical Specialties, Radiological Sciences, and Public Health) and their team. We have used the Z-inspection® process to assess the ethical, technical, medical and legal implications of using such AI system at the Brescia Public hospital (“ASST Spedali Civili di Brescia“). The results of our assessment – completed in December 2021- have been submitted for publication to a peer-reviewed journal.


Brixia score COVID-19 dataset: https://brixia.github.io

A large dataset of CXR images corresponding to the entire amount of images taken for both triage and patient monitoring in sub-intensive and intensive care units during one month (between March 4th and April 4th 2020) of pandemic peak at the ASST Spedali Civili di Brescia, and contains all the variability originating from a real clinical scenario. It includes 4,707 CXR images of COVID-19 subjects, acquired with both CR and DX modalities, in AP or PA projection, and retrieved from the facility RIS-PACS system.

Keywords: COVID-19 severity assessment, ChestX-Rays, semi-quantitative rating, End-to-end learning, Convolutional Neural Networks.

BS-Net: learning COVID-19 pneumonia severity on a large Chest X-Ray dataset,
Alberto Signoroni, Mattia Savardi, Sergio Benini, Nicola Adami, Riccardo Leonardi, Paolo Gibellini, Filippo Vaccher, Marco Ravanelli, Andrea Borghesi, Roberto Maroldi, Davide Farina, Medical Image Analysis,2021,102046, ISSN 1361-8415,


In this work we design an end-to-end deep learning architecture for predicting, on Chest X-rays images (CXR), a multi-regional score conveying the degree of lung compromise in COVID-19 patients. Such semi-quantitative scoring system, namely Brixia score, is applied in serial monitoring of such patients, showing significant prognostic value, in one of the hospitals that experienced one of the highest pandemic peaks in Italy. To solve such a challenging visual task, we adopt a weakly supervised learning strategy structured to handle different tasks (segmentation, spatial alignment, and score estimation) trained with a “from-the-part-to-the-whole” procedure involving different datasets.

In particular, we exploit a clinical dataset of almost 5,000 CXR annotated images collected in the same hospital. Our BS-Net demonstrates self-attentive behavior and a high degree of accuracy in all processing stages. Through inter-rater agreement tests and a gold standard comparison, we show that our solution outperforms single human annotators in rating accuracy and consistency, thus supporting the possibility of using this tool in contexts of computer-assisted monitoring. Highly resolved (super-pixel level) explainability maps are also generated, with an original technique, to visually help the understanding of the network activity on the lung areas. We also consider other scores proposed in literature and provide a comparison with a recently proposed non-specific approach. We eventually test the performance robustness of our model on an assorted public COVID-19 dataset, for which we also provide Brixia score annotations, observing good direct generalization and fine-tuning capabilities that highlight the portability of BS-Net in other clinical settings. The CXR dataset along with the source code and the trained model are publicly released for research purposes.

Keywords: COVID-19 severity assessment; Chest X-Rays; semi-quantitative rating; End-to-end learning; Convolutional Neural Networks

Alberto Signoroni, Department of Information Engineering, University of Brescia, Italy