Premium
Fully convolutional networks in multimodal nonlinear microscopy images for automated detection of head and neck carcinoma: Pilot study
Author(s) -
Rodner Erik,
Bocklitz Thomas,
von Eggeling Ferdinand,
Ernst Günther,
Chernavskaia Olga,
Popp Jürgen,
Denzler Joachim,
GuntinasLichius Orlando
Publication year - 2019
Publication title -
head and neck
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.012
H-Index - 127
eISSN - 1097-0347
pISSN - 1043-3074
DOI - 10.1002/hed.25489
Subject(s) - artificial intelligence , convolutional neural network , computer science , segmentation , pattern recognition (psychology) , head and neck cancer , histopathology , head and neck , computer vision , cancer , pathology , medicine , surgery
Background A fully convolutional neural networks (FCN)‐based automated image analysis algorithm to discriminate between head and neck cancer and noncancerous epithelium based on nonlinear microscopic images was developed. Methods Head and neck cancer sections were used for standard histopathology and co‐registered with multimodal images from the same sections using the combination of coherent anti‐Stokes Raman scattering, two‐photon excited fluorescence, and second harmonic generation microscopy. The images analyzed with semantic segmentation using a FCN for four classes: cancer, normal epithelium, background, and other tissue types. Results A total of 114 images of 12 patients were analyzed. Using a patch score aggregation, the average recognition rate and an overall recognition rate or the four classes were 88.9% and 86.7%, respectively. A total of 113 seconds were needed to process a whole‐slice image in the dataset. Conclusion Multimodal nonlinear microscopy in combination with automated image analysis using FCN seems to be a promising technique for objective differentiation between head and neck cancer and noncancerous epithelium.