Abstract
Pathologic diagnosis of nasopharyngeal carcinoma (NPC) can be challenging since most cases are nonkeratinizing carcinoma with little differentiation and many admixed lymphocytes. Our aim was to evaluate the possibility to identify NPC in nasopharyngeal biopsies using deep learning. A total of 726 nasopharyngeal biopsies were included. Among them, 100 cases were randomly selected as the testing set, 20 cases as the validation set, and all other 606 cases as the training set. All three datasets had equal numbers of NPC cases and benign cases. Manual annotation was performed. Cropped square image patches of 256 × 256 pixels were used for patch-level training, validation, and testing. The final patch-level algorithm effectively identified NPC patches, with an area under the receiver operator characteristic curve (AUC) of 0.9900. Using gradient-weighted class activation mapping, we demonstrated that the identification of NPC patches was based on morphologic features of tumor cells. At the second stage, whole-slide images were sequentially cropped into patches, inferred with the patch-level algorithm, and reconstructed into images with a smaller size for training, validation, and testing. Finally, the AUC was 0.9848 for slide-level identification of NPC. Our result shows for the first time that deep learning algorithms can identify NPC.
Highlights
Nasopharyngeal carcinoma (NPC) is a cancer with unique ethnic predisposition, Epstein–Barr virus (EBV) association, and morphologic features [1]
During patch-level learning, we noticed that patches of benign nasopharynx with certain morphologic features, including germinal centers and benign epithelial cells, tended to be misclassified as nasopharyngeal carcinoma (NPC) (Figure 1A–C)
We show for the first time that deep convolutional neural networks can identify NPC, a cancer with little differentiation and many admixed inflammatory cells
Summary
Nasopharyngeal carcinoma (NPC) is a cancer with unique ethnic predisposition, Epstein–Barr virus (EBV) association, and morphologic features [1]. In endemic areas, including Taiwan, most cases of NPC are of the nonkeratinizing type with EBV association [1,2,3]. Such nonkeratinizing NPC is characterized by undifferentiated or poorly differentiated carcinoma cells and a large number of admixed inflammatory cells, mainly small lymphocytes and plasma cells [1]. These morphologic features could pose difficulty in pathologic diagnosis, especially for pathologists with less experience or in nonendemic areas. The digital files of high-resolution pathology images are extremely large in size and highly complicated, and computer analysis of such images is very difficult
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.