Related Experiment Video For Computed tomography
Updated: Jun 5, 2026
Comparison of Predictive Performance of Three Lymph Node Staging Systems in Colorectal Signet Ring Cell Carcinoma Based on Machine Learning Model
Published on: April 18, 2025
Diagnostic performance of the NI-RADS system for detecting recurrence in head and neck squamous cell carcinoma: a
Sara Greco1, Ilaria Villanova1, Nicola Maggialetti2
1Interdisciplinary Department of Medicine, Section of Radiology and Radiation Oncology, University of Bari "Aldo Moro", 70124, Bari, Italy.
Background:
This pilot study evaluates the diagnostic performance and clinical relevance of the Neck Imaging Reporting and Data System (NI-RADS) in the post-treatment surveillance of patients with head and neck squamous cell carcinoma (HNSCC), using computed tomography (CT) and magnetic resonance imaging (MRI).
Methods:
This retrospective study included 25 patients with histologically confirmed HNSCC treated between August 2021 and December 2024. All patients underwent at least two post-treatment CT or MRI scans, independently reviewed by two radiologists and classified according to the NI-RADS system. Recurrence was confirmed by histopathology or clinical and imaging follow-up. Diagnostic performance metrics, including sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and accuracy, were calculated using a threshold of NI-RADS ≥ 3. Receiver operating characteristic (ROC) analysis was performed for primary and neck sites.
Results:
A total of 56 imaging studies and 112 anatomical sites were assessed. Recurrence was confirmed in 36% of primary sites and 27% of neck sites. NI-RADS category 1 showed 0% recurrence, whereas category 4 showed 100% recurrence for both anatomical sites. For the primary site, sensitivity was 80% (95% CI: 56-94%) and specificity 92% (95% CI: 78-98%), with an accuracy of 88% (95% CI: 76-95%). For neck sites, sensitivity was 80% (95% CI: 52-96%) and specificity 98% (95% CI: 87-100%), with an accuracy of 93% (95% CI: 82-98%). The area under the ROC curve was 0.86 for the primary site and 0.88 for the neck site.
Conclusions:
NI-RADS showed good diagnostic performance for detecting locoregional recurrence in post-treatment surveillance of HNSCC using CT and MRI. However, given the retrospective design and limited sample size, these findings should be considered preliminary and require confirmation in larger prospective studies.
