Comparative Performance of Four Staging Classifications to Select «High-Risk¼ Head and Neck Cutaneous Squamous Cell Carcinomas.

BACKGROUND: Many classifications exist to select patients with "high-risk" head and neck cutaneous squamous cell carcinoma (HNCSCC). OBJECTIVE: To compare the performance of the Brigham and Women's Hospital (BWH) classification with the performance of the American Joint Committee on Cancer 8th Edition (AJCC8), the Union for International Cancer Control 8th Edition (UICC8), and the National Comprehensive Cancer Network (NCCN) classifications. METHODS: In this single-center retrospective study, HNCSCC resected in a tertiary care center were classified as "low-risk" or "high-risk" tumors according to the four classifications. Rates of local recurrence (LR), lymph node recurrence (NR), and disease-specific death (DSD) were collected. The performance of each classification was then calculated in terms of homogeneity, monotonicity, and discrimination and compared. RESULTS: Two hundred and seventeen HNCSCC from 160 patients, with a mean age of 80 years, were included. For predicting the risk of any poor outcome and risk of NR, the BWH classification had the best specificity and positive predictive value. However, its concordance index was not significantly higher than that of the AJCC8 and UICC8 classifications. The NCCN classification was the least discriminant. CONCLUSIONS AND RELEVANCE: This study suggests that the BWH classification is the most appropriate for predicting the risk of poor outcomes in patients with HNCSCC when compared with the NCCN, UICC8, and AJCC8 classifications.

Validation and Improvement of a Convolutional Neural Network to Predict the Involved Pathology in a Head and Neck Surgery Cohort.

The selection of patients for the constitution of a cohort is a major issue for clinical research (prospective studies and retrospective studies in real life). Our objective was to validate in real life conditions the use of a Deep Learning process based on a neural network, for the classification of patients according to the pathology involved in a head and neck surgery department. 24,434 Electronic Health Records (EHR) from the first visit between 2000 and 2020 were extracted. More than 6000 EHR were manually classified in ten groups of interest according to the reason for consultation with a clinical relevance. A convolutional neural network (TensorFlow, previously reported by Hsu et al.) was then used to predict the group of patients based on their pathology, using two levels of classification based on clinically relevant criteria. On the first and second level of classification, macro-average performances were: 0.95, 0.83, 0.85, 0.97, 0.84 and 0.93, 0.76, 0.83, 0.96, 0.79 for accuracy, recall, precision, specificity and F1-score versus accuracy, recall and precision of 0.580, 580 and 0.582 for Hsu et al., respectively. We validated this model to predict the pathology involved and to constitute clinically relevant cohorts in a tertiary hospital. This model did not require a preprocessing stage, was used in French and showed equivalent or better performances than other already published techniques.