Background and aim: Artificial Intelligence (AI) in healthcare is rapidly expanding and researchers are exploring its possible role in assisting physicians in early diagnosis, accurate prognosis prediction, and efficient treatment planning. This systematic review aims to summarize the evidence about the role of AI, Machine Learning (ML), and Deep Learning (DL) in the diagnostic imaging of chronic rhinosinusitis (CRS). Methods: The search strategy was performed according to PRISMA guidelines for systematic reviews. The authors searched all articles in three major medical databases (PubMed, Scopus, Cochrane Library) using the following key terms: “Artificial Intelligence” or “Machine Learning” or “Deep Learning” or “Neural Convolution Learning” or “Knowledge Engineering” and “Nose” or “Nasal” or “Septum” or “Turbinate” or “Sinus” or “Rhinology” or “Sinusitis” or “Rhinosinusitis” or “Chronic Rhinosinusitis” or “Chronic Sinusitis” or “CRS” and “CT” or “MRI” or “Computed Tomography” or “Images” or “CBCT” or “Magnetic Resonance Imaging” or “Imaging” or “Radiographs” or “X-ray”. Results: Overall, 395 manuscripts were identified, and after duplicate removal (27 articles), excluding off-topic studies (298) and for other structural reasons (50) papers were assessed for eligibility; finally, only 20 papers were included and summarized in the review. Conclusions: Despite the growing interest in AI applications, due to the lack of standardized and unified validation procedures and the heterogeneity of patient cohorts, its practical role in rhinology, particularly in radiological image processing in CRS, is not yet well defined, and further research is needed. It should be crucial for physicians to use their knowledge and skills to critically assess the information provided by AI and make any final treatment decisions. (www.actabiomedica.it).
Diagnostic applications of artificial intelligence in chronic rhinosinusitis: a systematic review
Cavaliere C.;
2025-01-01
Abstract
Background and aim: Artificial Intelligence (AI) in healthcare is rapidly expanding and researchers are exploring its possible role in assisting physicians in early diagnosis, accurate prognosis prediction, and efficient treatment planning. This systematic review aims to summarize the evidence about the role of AI, Machine Learning (ML), and Deep Learning (DL) in the diagnostic imaging of chronic rhinosinusitis (CRS). Methods: The search strategy was performed according to PRISMA guidelines for systematic reviews. The authors searched all articles in three major medical databases (PubMed, Scopus, Cochrane Library) using the following key terms: “Artificial Intelligence” or “Machine Learning” or “Deep Learning” or “Neural Convolution Learning” or “Knowledge Engineering” and “Nose” or “Nasal” or “Septum” or “Turbinate” or “Sinus” or “Rhinology” or “Sinusitis” or “Rhinosinusitis” or “Chronic Rhinosinusitis” or “Chronic Sinusitis” or “CRS” and “CT” or “MRI” or “Computed Tomography” or “Images” or “CBCT” or “Magnetic Resonance Imaging” or “Imaging” or “Radiographs” or “X-ray”. Results: Overall, 395 manuscripts were identified, and after duplicate removal (27 articles), excluding off-topic studies (298) and for other structural reasons (50) papers were assessed for eligibility; finally, only 20 papers were included and summarized in the review. Conclusions: Despite the growing interest in AI applications, due to the lack of standardized and unified validation procedures and the heterogeneity of patient cohorts, its practical role in rhinology, particularly in radiological image processing in CRS, is not yet well defined, and further research is needed. It should be crucial for physicians to use their knowledge and skills to critically assess the information provided by AI and make any final treatment decisions. (www.actabiomedica.it).I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
