At the San Camillo hospital in Rome, artificial intelligence reviews x-rays. A virtual assistant that ‘takes a look’ to provide a second look, a second reading, in real time. The emergency radiology department of the Roman hospital company – the Ao informs – has been experimenting with the implementation of AI software in the field of radiodiagnostics for about 6 months, capable of offering support for the diagnosis of osteoarticular lesions and pulmonary pathologies.
The Emergency/Urgent Radiology Uosd of San Camillo, directed by Michele Galluzzo and affiliated with the Department of Emergency, Acceptance and Specialist Surgery led by Pierluigi Marini – explains a note – manages extremely intense work flows, both in terms of the number of patients and the complexity of the exams. It is active 24 hours a day, 365 days a year and performs a total annual number of services equal to 84,561, of which 41,792 CT exams and 35,084 X-ray exams. This requires particular attention in reporting, even at the most unfavorable times, such as at night.
Starting from the end of 2023, therefore, a team composed of Galluzzo, Riccardo Ferrari, medical director of Emergency/Urgency Radiology and elected president of the Informatics and AI section of Sirm (Italian Society of Medical and Interventional Radiology), and Annarita Caporaso, director of Clinical Engineering, has started feasibility studies to implement the technological equipment of Emergency Radiology with software powered by artificial intelligence, with the aim of “optimizing and improving, thanks to new technologies, outcomes, times and use of resources”. A software has therefore been acquired that works as a ‘second reader’ of radiographic images, capable of identifying osteotraumatic lesions such as fractures, dislocations and effusions, as well as focal lesions of the bone, and of providing information to support the radiologist in diagnosing some lung pathologies in emergency conditions, even in radiograms performed in difficult situations, for example in bedridden patients.
“Artificial intelligence is a technology with a strong impact on diagnostic imaging – says Galluzzo – In emergency conditions it is a useful diagnostic support for the radiologist. It is important to underline – he specifies – that the technology does not replace the doctor, but represents a valid tool with a view to improving diagnostic and management performance”.
“In a highly technological field such as radiodiagnostics – adds Ferrari – the implementation of technologies based on artificial intelligence allows the radiologist to use automation and diagnostic support to correctly interpret images and diagnose pathologies. The use of diagnostic support technologies in emergency radiology, and in a high-traffic department such as ours, means providing colleagues and patients with the most cutting-edge technologies for diagnosis and treatment”. Specific training of medical and technical staff was necessary – the note highlights – with precise instructions on the advantages and limits of using these technologies.
“We believe that the acquired system has an excellent quality/price ratio – Caporaso points out – this aspect is also relevant in terms of guaranteeing the quality of the services provided while still respecting the containment of health expenditure. We are ready to acquire other AI-based software that allows us to support the radiologist even in time-dependent pathologies, with a pre-reading and priority reporting of findings of particular interest and urgency”. A project that is part of the broader one strongly desired by the strategic management – we read – to bring the San Camillo-Forlanini hospital company to be the first among the Ao of Lazio in the use of technologies based on artificial intelligence.
#radiologist #San #Camillo #Rome #diagnosis #virtual #assistant
