Health, early diagnosis of lung cancer thanks to AI

An artificial intelligence software to obtain a more effective and efficient early diagnosis of lung cancer.

It’s called “Deep Lung” and it’s a program based on Deep Learning algorithms developed by the Catalan Foundation Eurecat in order to monitor and study the development of suspicious nodules detected via lung CT scans in patients undergoing complete screenings, with the aim of the detection of malignancy with greater accuracy and in shorter time, and thus contributing to the timely diagnosis of cancer. 

Thanks to an agreement jointly signed by Emilia-Romagna Region, Eurecat Foundation and Irccs Sant’Orsola Polyclinic, this software will be tried out both in Bologna and at the University Clinic of Barcelona starting from June 2024 in order to determine the impact that this program could have on the early diagnosis and screening of lung cancer and the consequent benefits for the protection of patients health.

The final goal, once the testing phase is over, is to activate an innovative and free service for every Emilia-Romagna public healthcare institution.

The necessary computational power will be offered, for free, by the supercomputer called “MarghERita”, that the Emilia-Romagna Region has acquired with a 3.5 million euros funding offered by the Development and Cohesion Fund (DCF), with the aim of allowing local public administrations and healthcare companies to make use of the potential of new technologies.

The agreement between the Regional administration, Eurecat and Irccs Sant’Orsola Polyclinic in Bologna is the result of an important and positive example of european territorial cooperation between Emilia-Romagna and Catalonia with the project “Digital Transformation for Regions-DT4Regions”,driven by Emilia-Romagna, aimed at creating a European platform for Regions to optimize collective AI and Big Data solutions.

A collaboration that started in 2022  in Bologna with an agreement signedby Stefano Bonaccini (Emilia-Romagna Region President) and Pere Aragonès i Garcia (Generalitat de Catalunya President), at the regional headquarters. This experimentation won’t entail financial burdens for Emilia-Romagna, thanks to regional and Sant’Orsola Polyclinic competences and Eurocat effort, that will offer its software for free.

“Early diagnosis and timely cures are very important on an oncological point of view and can effectively save lives” underline Raffaele Donini and Paolo Calvano, advisors to Healthcare Politics and to European Cooperation respectively. From this perspective, artificial intelligence and Big Data analysis can be a precious support for diagnosis activities, in order to make more precise and timely clinical decisions and contribute to the improvement of patients’ health.

Both advisors also add that “The possibility to test the Deep Lung project represents an important chance of international collaboration to verify the efficiency of an innovative technology, able to increase the quality of public services given by the Health department. All of this would not have been possible without the extraordinarily competent people who already work in the Region and in Sant’Orsola Polyclinic. We can face the new challenges to which Public Health is called thanks to the cooperation of institutions and research centers, the big amount of data now available for the Public and the exchange of knowledge and competences”.

The “Deep Lung” project

The “Deep Lung” project aims to test an innovative software for an early lung cancer diagnosis that implements interconnected algorithms architectures of Artificial Intelligence (siamese convolutional learning) which were developed to increment the efficiency and promptness of lung cancer diagnosis, simplifying the monitoring of suspicious nodules detected through computerized axial tomography (CT) and making the comparison more precise and reliable.

The nodule, which was previously identified with the CT scan, gets checked, and the artificial intelligence analyzes physical details and growth potential – possible symptoms of a malignant tumor – accelerating a possible diagnosis. This is one of the most complex AI tests ever realized in Italy for the healthcare department and aims to understand how the big amount of neural nets that compose the software are able to learn and gradually support with ever greater accuracy radiologists and oncologists that work in both Countries.

The project is the result of a collaboration between the General Directions of Resources, Europe, Institutions and Innovation and personal Care, Healthcare and Welfare of the Emilia-Romagna region, the Irccs of the Sant’Orsola Polyclinic, which will use the data and lung scans which are already available for the Public Health Department to start its experimentation, and the Eurecat Foundation, which has created the software and will allow its use for free. Signing the agreement that kicks off the experimentation was the general director of Resources, Europe, Innovation and Institutions of the Emilia-Romagna Region, Francesco Raphael Frieri, the general director of Personal Care, Health and Welfare, Luca Baldino, the general director of the Irccs Sant'Orsola Polyclinic, Chiara Gibertoni, and Xavier López Luján, general director of corporate and Eurecat operations.

The Eurecat Foundation 

Eurecat – Centre Tecnològic de Catalunya is a private, non-profit Foundation which aims to effectively contribute to the improvement of competitivity, technological and sustainable development of companies and to the welfare of both citizens and society. The Foundation is the leader of the Centre of Innovation of Data Technologies and Artificial Intelligence – CIDAI, which promotes innovation based on AI and Big Data in Catalonia.