RESUMO
OBJECTIVE: The presence of occult nodal metastases in patients with squamous cell carcinoma (SCC) of the oral tongue has implications for treatment. Upwards of 30% of patients will have occult nodal metastases, yet a significant number of patients undergo unnecessary neck dissection to confirm nodal status. This study sought to predict the presence of nodal metastases in patients with SCC of the oral tongue using a convolutional neural network (CNN) that analyzed visual histopathology from the primary tumor alone. METHODS: Cases of SCC of the oral tongue were identified from the records of a single institution. Only patients with complete pathology data were included in the study. The primary tumors were randomized into 2 groups for training and testing, which was performed at 2 different levels of supervision. Board-certified pathologists annotated each slide. HALO-AI convolutional neural network and image software was used to perform training and testing. Receiver operator characteristic (ROC) curves and the Youden J statistic were used for primary analysis. RESULTS: Eighty-nine cases of SCC of the oral tongue were included in the study. The best performing algorithm had a high level of supervision and a sensitivity of 65% and specificity of 86% when identifying nodal metastases. The area under the curve (AUC) of the ROC curve for this algorithm was 0.729. CONCLUSION: A CNN can produce an algorithm that is able to predict nodal metastases in patients with squamous cell carcinoma of the oral tongue by analyzing the visual histopathology of the primary tumor alone.
Assuntos
Carcinoma de Células Escamosas , Neoplasias da Língua , Humanos , Inteligência Artificial , Neoplasias da Língua/patologia , Carcinoma de Células Escamosas/patologia , Língua/patologia , Esvaziamento Cervical/métodos , Estudos Retrospectivos , Linfonodos/patologia , Estadiamento de NeoplasiasRESUMO
BACKGROUND: The incidence of thyroid cancer in the United States has risen dramatically since the 1970s, driven by an increase in the diagnosis of small tumors. There is a paucity of published New Mexico (NM) specific data regarding thyroid cancer. We hypothesized that due to New Mexico's unique geographic and cultural makeup, the incidence of thyroid cancer and tumor size at diagnosis in this state would differ from that demonstrated on a national level. METHODS: The New Mexico Tumor Registry (NMTR) was queried to include all NM residents diagnosed with thyroid cancer between 1992 and 2019. For 2010 to 2019, age-adjusted incidence rates were calculated via direct method using the 2000 United States population as the adjustment standard. Differences in incidence rate and tumor size by race/ethnicity and residence (metropolitan vs non-metropolitan) were assessed with rate ratios between groups. For 1992 to 2019, temporal trends in age-adjusted incidence rates for major race/ethnic groups in NM [Non-Hispanic White (NHW), Hispanic, and American Indian (AI)] were assessed by joinpoint regression using National Cancer Institute software. RESULTS: Our study included 3,161 patients for the time period 2010 to 2019, including NHW (1518), Hispanic (1425), and AI (218) cases. The overall incidence rates for NM AIs were lower than those for Hispanics and NHWs because of a decreased incidence of very small tumors (<1.1 cm). The incidence rates for large tumors (>5.1 cm) was equivalent among groups. In the early 2000s, Hispanics also had lower rates of small tumors when compared to NHWs but this trend disappeared over time. CONCLUSION: AIs in New Mexico have been left out of the nationwide increase in incidental diagnosis of small thyroid tumors. This same pattern was noted for Hispanics in the early 2000s but changed over time to mirror incidence rates for NHWs. These data are illustrative of the health care disparities that exist among New Mexico's population and how these disparities have changed over time.