TY - JOUR
T1 - Enhanced Pathology Image Quality with Restore–Generative Adversarial Network
AU - Rong, Ruichen
AU - Wang, Shidan
AU - Zhang, Xinyi
AU - Wen, Zhuoyu
AU - Cheng, Xian
AU - Jia, Liwei
AU - Yang, Donghan M.
AU - Xie, Yang
AU - Zhan, Xiaowei
AU - Xiao, Guanghua
N1 - Publisher Copyright:
© 2023 American Society for Investigative Pathology
PY - 2023/4
Y1 - 2023/4
N2 - Whole slide imaging is becoming a routine procedure in clinical diagnosis. Advanced image analysis techniques have been developed to assist pathologists in disease diagnosis, staging, subtype classification, and risk stratification. Recently, deep learning algorithms have achieved state-of-the-art performances in various imaging analysis tasks, including tumor region segmentation, nuclei detection, and disease classification. However, widespread clinical use of these algorithms is hampered by their performances often degrading due to image quality issues commonly seen in real-world pathology imaging data such as low resolution, blurring regions, and staining variation. Restore–Generative Adversarial Network (GAN), a deep learning model, was developed to improve the imaging qualities by restoring blurred regions, enhancing low resolution, and normalizing staining colors. The results demonstrate that Restore-GAN can significantly improve image quality, which leads to improved model robustness and performance for existing deep learning algorithms in pathology image analysis. Restore-GAN has the potential to be used to facilitate the applications of deep learning models in digital pathology analyses.
AB - Whole slide imaging is becoming a routine procedure in clinical diagnosis. Advanced image analysis techniques have been developed to assist pathologists in disease diagnosis, staging, subtype classification, and risk stratification. Recently, deep learning algorithms have achieved state-of-the-art performances in various imaging analysis tasks, including tumor region segmentation, nuclei detection, and disease classification. However, widespread clinical use of these algorithms is hampered by their performances often degrading due to image quality issues commonly seen in real-world pathology imaging data such as low resolution, blurring regions, and staining variation. Restore–Generative Adversarial Network (GAN), a deep learning model, was developed to improve the imaging qualities by restoring blurred regions, enhancing low resolution, and normalizing staining colors. The results demonstrate that Restore-GAN can significantly improve image quality, which leads to improved model robustness and performance for existing deep learning algorithms in pathology image analysis. Restore-GAN has the potential to be used to facilitate the applications of deep learning models in digital pathology analyses.
UR - http://www.scopus.com/inward/record.url?scp=85151235604&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85151235604&partnerID=8YFLogxK
U2 - 10.1016/j.ajpath.2022.12.011
DO - 10.1016/j.ajpath.2022.12.011
M3 - Article
C2 - 36669682
AN - SCOPUS:85151235604
SN - 0002-9440
VL - 193
SP - 404
EP - 416
JO - American Journal of Pathology
JF - American Journal of Pathology
IS - 4
ER -