Using artificial intelligence to help doctors and dentists recognize suspicious oral lesions early and refer patients more quickly for specialist evaluation. This is the goal of an international study coordinated by the University of Sassari and published in the journal Otolaryngology–Head and Neck Surgery , the official journal of the American Academy of Otolaryngology–Head and Neck Surgery Foundation. The study, titled "Image-Based Diagnosis of Oral Lesions: Performance of a Vision-Language Model versus Human Clinicians," evaluated the capabilities of a specially trained multimodal artificial intelligence system based on Gemini 2.5 Pro in analyzing images of oral mucosal lesions.

The research represents one of the results of the investment made by the University of Sassari and the Sassari University Hospital within the e.INS – Ecosystem of Innovation for Next Generation Sardinia project and, in particular, within Spoke 01 "A new route to preventive medicine: genomics, digital innovation, and telemedicine," of which the University of Sassari is the implementing entity and Spoke Leader. The program is funded under the National Recovery and Resilience Plan, with resources from the European Union – NextGenerationEU. The Telemaco-S project, a collaboration between the University and the University of Sassari and dedicated to the development of telemedicine and precision medicine solutions, also fits into this context.

"Oral cancer continues to be a significant problem because too many patients are diagnosed when the disease is already in an advanced stage," explains Professor Giacomo De Riu, Professor of Maxillofacial Surgery at the University of Sassari and Director of the Maxillofacial Surgery Unit at the University Hospital of Sassari. "The diagnostic delay may be due to the patient, who sometimes underestimates an initial lesion, but also to the difficulty in recognizing manifestations that in the early stages may resemble benign conditions. Reducing the time between the appearance of the lesion, its recognition, and referral to a specialist is therefore one of the most important goals in the fight against these tumors."

The prospective, multicenter study involved 350 patients , collected from 20 university and hospital centers in Italy, Belgium, France, Spain, and Israel. Lesion diagnoses were verified through histological examination, which served as a reference to evaluate the system's accuracy.

The results showed an overall accuracy of 97.1% in distinguishing between malignant and benign lesions. The study also included an exploratory comparison with individual professionals with varying levels of experience. In distinguishing between malignant and benign lesions, the accuracy was 72.5% for general practitioners, 78.2% for dentists, 87.4% for oral pathologists, and 99.4% for experienced head and neck surgeons, compared to 97.1% for the Gemini-based system.

© Riproduzione riservata