Deep Learning Prediction of TERT Promoter Mutation Status in Thyroid Cancer Using Histologic Images

objectives: Telomerase reverse transcriptase ( ) promoter mutation, found in a subset of patients with thyroid cancer, is strongly associated with aggressive biologic behavior. Predicting promoter mutation is thus necessary for the prognostic stratification of thyroid cancer patients. In this study,...

Full description

Saved in:
Bibliographic Details
Published in:Medicina (Kaunas, Lithuania) Lithuania), 2023-03, Vol.59 (3), p.536
Main Authors: Kim, Jinhee, Ko, Seokhwan, Kim, Moonsik, Park, Nora Jee-Young, Han, Hyungsoo, Cho, Junghwan, Park, Ji Young
Format: Article
Language:English
Subjects:
Classification
CNN
color transformation
CRNN
Datasets
Deep Learning
Humans
Mutation
Neural networks
Pathology
Prognosis
Promoter Regions, Genetic
Proto-Oncogene Proteins B-raf - genetics
Telomerase - genetics
TERT
Thyroid cancer
Thyroid Neoplasms - diagnostic imaging
Thyroid Neoplasms - genetics
Tumors
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:objectives: Telomerase reverse transcriptase ( ) promoter mutation, found in a subset of patients with thyroid cancer, is strongly associated with aggressive biologic behavior. Predicting promoter mutation is thus necessary for the prognostic stratification of thyroid cancer patients. In this study, we evaluate promoter mutation status in thyroid cancer through the deep learning approach using histologic images. Our analysis included 13 consecutive surgically resected thyroid cancers with promoter mutations (either C228T or C250T) and 12 randomly selected surgically resected thyroid cancers with a wild-type promoter. Our deep learning model was created using a two-step cascade approach. First, tumor areas were identified using convolutional neural networks (CNNs), and then promoter mutations within tumor areas were predicted using the CNN-recurrent neural network (CRNN) model. : Using the hue-saturation-value (HSV)-strong color transformation scheme, the overall experiment results show 99.9% sensitivity and 60% specificity (improvements of approximately 25% and 37%, respectively, compared to image normalization as a baseline model) in predicting mutations. : Highly sensitive screening for promoter mutations is possible using histologic image analysis based on deep learning. This approach will help improve the classification of thyroid cancer patients according to the biologic behavior of tumors.
ISSN:1648-9144
1010-660X
1648-9144
DOI:10.3390/medicina59030536