Instance segmentation of upper aerodigestive tract cancer: site-specific outcomes.

Objective: To achieve instance segmentation of upper aerodigestive tract (UADT) neoplasms using a deep learning (DL) algorithm, and to identify differences in its diagnostic performance in three different sites: larynx/hypopharynx, oral cavity and oropharynx. Methods: A total of 1034 endoscopic images from 323 patients were examined under narrow band imaging (NBI). The Mask R-CNN algorithm was used for the analysis. The dataset split was: 935 training, 48 validation and 51 testing images. Dice Similarity Coefficient (Dsc) was the main outcome measure. Results: Instance segmentation was effective in 76.5% of images. The mean Dsc was 0.90 ± 0.05. The algorithm correctly predicted 77.8%, 86.7% and 55.5% of lesions in the larynx/hypopharynx, oral cavity, and oropharynx, respectively. The mean Dsc was 0.90 ± 0.05 for the larynx/hypopharynx, 0.60 ± 0.26 for the oral cavity, and 0.81 ± 0.30 for the oropharynx. The analysis showed inferior diagnostic results in the oral cavity compared with the larynx/hypopharynx (p < 0.001). Conclusions: The study confirms the feasibility of instance segmentation of UADT using DL algorithms and shows inferior diagnostic results in the oral cavity compared with other anatomic areas.

Regional disease in head and neck cutaneous squamous cell carcinoma: the role of primary tumor characteristics and number of nodal metastases.

PURPOSE: To identify potential risk factors impacting on overall survival (OS) of patients affected by lymph node metastasis from cutaneous squamous cell carcinoma (cSCC) of the head and neck (HN), with special emphasis on primary tumor characteristics and pattern of nodal recurrence (intraparotid and/or cervical). METHODS: A bi-institutional retrospective study on consecutive patients affected by cervical and/or intraparotid NM from HN cSCC and surgically treated with curative intent from May 2010 to January 2020 was conducted. OS was considered the outcome of interest. RESULTS: The study included 89 patients (M:F = 3.4:1; median age, 78 years; range, 22-99). Among the primary tumor characteristics, the most relevant prognostic factors were diameter ≥ 4 cm (hazard ratio [HR] = 2.56, p = 0.010) and depth of infiltration ≥ 6 mm (HR = 3.54, p = 0.027). Cervical NM was associated with worse OS (HR = 2.09, p = 0.016) compared to purely intraparotid NM (5-year OS: 60.9% vs. 28.1%, p = 0.014). At multivariable analysis, age, immunosuppression, pT3-T4 categories and a high burden of nodal disease (> 2 NM) confirmed to be independent risk factors, whereas adjuvant radiotherapy was independently associated with better outcome. CONCLUSION: This study confirms the association of several independent prognosticators related to the patient, primary tumor, and nodal burden status. Patients with cervical NM should be considered at risk for harboring a higher number of metastatic lymph nodes.