Text size :

Cancers, Laghi (Humanitas): “Radiologist’s challenge is to predict treatment response”

ADNKRONOS

November 6, 2026

Rome, Nov 6. (Adnkronos Health) – “Predicting the response to therapy is the great challenge we are facing in oncology. It would be fundamental to know, already at the time of diagnosis, what the evolution of a tumor will be and how it will respond to the specific therapeutic option we choose. In research, we do not yet have definitive tools in this regard, but it is the work that large international consortia are carrying out, collecting millions of images of oncology patients and applying artificial intelligence algorithms to predict the response to therapy, thus offering the patient, from diagnosis, a targeted prognosis.” This was stated by Andrea Laghi, director of the Diagnostic Imaging Department at IRCCS Humanitas Clinical Institute and full professor of Radiology, Humanitas University, today in Milan, at the 52nd Sirm Congress – Italian Society of Medical and Interventional Radiology, which concludes on November 8.

The application of AI in the field of diagnostic imaging is bringing numerous benefits to specialists and patients. “There are various integrated systems within our equipment that allow, for example, in a simple radiographic examination, to better center the patient and ensure a reproducible quality standard during various examinations,” explains Laghi. “In magnetic resonance imaging, systems drastically reduce acquisition times, improving patient comfort as they do not have to remain inside the tube for too many minutes. In examinations involving the use of significant radiation, such as computed tomography (CT) – he adds – today we record a drastic reduction in dose, even by 60% compared to past examinations. This has an extremely important impact, especially for oncology patients who must repeatedly undergo CT scans to monitor their therapy.”

These tools “can support us, for example, by reducing diagnostic errors when the radiologist is fatigued, but we cannot delegate the final judgment to them,” warns Laghi. “We must maintain our critical spirit, even questioning what artificial intelligence suggests to us. It must always be remembered that these tools are not infallible: they are probabilistic systems with a very high probability of success and highly significant technical data, but not perfect. Guidance and responsibility – he concludes – must always remain in human hands.”

Jiji Press

Related

Page TOP