Super resolution of pathology images with hierarchical feature integration and local image patterns
The Journal of Pathology · 2025-12 · 267 (4) · 464-478
DOI: 10.1002/path.6482 ↗Kurz gefasst
Fachartikel in der Zeitschrift The Journal of Pathology, online erschienen am 13.10.2025. Verfasst von Feng Xu und 8 weiteren Autor:innen. Beteiligte Einrichtung: Department of Gastroenterology, The Affiliated Lihuili Hospital of Ningbo University, Zhejiang, PR China. Hauptthemen laut MEDLINE: Deep Learning, Image Interpretation, Computer-Assisted, Image Processing, Computer-Assisted. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 55 Literaturangaben. Gefördert von: Ningbo public welfare project, Medical and Health Research Project of Zhejiang Province, the Huili Fund of Li Huili Hospital of Ningbo Medical Center.
Abstract (Originalsprache)
Recent advancements in pathological imaging have facilitated single‐cell and subcellular‐level analysis based on high‐resolution images for tumor subtyping, cytomorphological assessment, and infection detection. As high‐resolution imaging is often limited by cost, super‐resolution methods provide a practical alternative with only low‐resolution data. However, existing methods generally suffer from artifacts, oversmoothing, and slow inference speed. In this study, we developed a hierarchal deep learning framework based on local pathological image patterns, named Hierarchical Local Image Patterns (HLIP), to achieve accurate, high‐fidelity, and real‐time super resolution with flexible magnifications. HLIP integrates semantic features with both pixel‐ and morphology‐level features and reconstructs super‐resolution images by the recognized local pathological image patterns. Benchmark analysis showed HLIP achieved the best performance and robustness on both internal and external test datasets. The generated super‐resolution images contain abundant pathological details and maintain high fidelity. HLIP can be used for the enhancement of other models across multiple clinical scenarios, including gland segmentation, cell segmentation, Helicobacter pylori detection, and therapy response prediction. With its superior performance in pathology image super resolution, HLIP offers a versatile tool for image preprocessing in computer‐aided systems, thereby supporting accurate diagnosis in clinical practice. © 2025 The Pathological Society of Great Britain and Ireland.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Gastroenterology, The Affiliated Lihuili Hospital of Ningbo University, Zhejiang, PR China.
- Pathological Diagnosis Department, Meikang Medicalsystem Laboratory, Zhejiang, PR China.
- Department of Pathology, School of Basic Medical Sciences, Fudan University, Shanghai, PR China.
- Shanghai Key Laboratory of Medical Imaging Computing and Computer Assisted Intervention, Shanghai, PR China.
- Department of Research and Development, Shanghai Laizi Software Technology Co., LTD, Shanghai, PR China.
- Ningbo Clinical Pathology Diagnosis Center, Zhejiang, PR China.
ORCID-Profile: Jin Ling, Chaofei Gao
Förderung
- Ningbo public welfare project
- Medical and Health Research Project of Zhejiang Province
- the Huili Fund of Li Huili Hospital of Ningbo Medical Center
Zitieren
APA
Xu, F., Li, L., Wang, S., Ling, R., Zhang, X., Deng, X., Zhou, M., Ling, J., & Gao, C. (2025). Super resolution of pathology images with hierarchical feature integration and local image patterns. The Journal of Pathology, 267(4), 464-478. https://doi.org/10.1002/path.6482
Vancouver
Xu F, Li L, Wang S, Ling R, Zhang X, Deng X, et al. Super resolution of pathology images with hierarchical feature integration and local image patterns. The Journal of Pathology. 2025;267(4):464-478. doi:10.1002/path.6482
BibTeX
@article{Xu2025,
author = {Xu, Feng and Li, Lei and Wang, Shuyang and Ling, Ren and Zhang, Xie and Deng, Xi and Zhou, Mengzhe and Ling, Jin and Gao, Chaofei},
title = {Super resolution of pathology images with hierarchical feature integration and local image patterns},
journal = {The Journal of Pathology},
year = {2025},
volume = {267},
number = {4},
pages = {464--478},
doi = {10.1002/path.6482}
}
EUVIMED