Videomics of the Upper Aero-Digestive Tract Cancer: Deep Learning Applied to White Light and Narrow Band Imaging for Automatic Segmentation of Endoscopic Images

作者：Muhammad Adeel Azam, Claudio Sampieri, Alessandro Ioppi, Pietro Benzi, Giorgio Gregory Giordano, Marta De Vecchi, Valentina Campagnari, Shunlei Li, Luca Guastini, Alberto Paderno, Sara Moccia, Cesare Piazza, Leonardo S. Mattos, Giorgio Peretti · 发表于：Frontiers in Oncology · 年份：2022 · DOI：10.3389/fonc.2022.900451 · 被引用次数：37 · 研究领域：Esophageal Cancer Research and Treatment、Photodynamic Therapy Research Studies、Head and Neck Cancer Studies

Introduction Narrow Band Imaging (NBI) is an endoscopic visualization technique useful for upper aero-digestive tract (UADT) cancer detection and margins evaluation. However, NBI analysis is strongly operator-dependent and requires high expertise, thus limiting its wider implementation. Recently, artificial intelligence (AI) has demonstrated potential for applications in UADT videoendoscopy. Among AI methods, deep learning algorithms, and especially convolutional neural networks (CNNs), are particularly suitable for delineating cancers on videoendoscopy. This study is aimed to develop a CNN for automatic semantic segmentation of UADT cancer on endoscopic images. Materials and Methods A dataset of white light and NBI videoframes of laryngeal squamous cell carcinoma (LSCC) was collected and manually annotated. A novel DL segmentation model (SegMENT) was designed. SegMENT relies on DeepLabV3+ CNN architecture, modified using Xception as a backbone and incorporating ensemble features from other CNNs. The performance of SegMENT was compared to state-of-the-art CNNs (UNet, ResUNet, and DeepLabv3). SegMENT was then validated on two external datasets of NBI images of oropharyngeal (OPSCC) and oral cavity SCC (OSCC) obtained from a previously published study. The impact of in-domain transfer learning through an ensemble technique was evaluated on the external datasets. Results 219 LSCC patients were retrospectively included in the study. A total of 683 videoframes composed the LSCC da...