Integrating SAM priors with U-Net for enhanced multiclass cell detection in digital pathology

Abstract In digital pathology, the accurate detection, segmentation, and classification of cells are pivotal for precise pathological diagnosis. Traditionally, pathologists manually segment cells from pathological images to facilitate diagnosis based on these results and other critical indicators. H...

Full description

Saved in:
Bibliographic Details
Main Authors: Zheng Wu, Ji-Yun Yang, Chang-Bao Yan, Cheng-Gui Zhang, Hai-Chao Yang
Format: Article
Language:English
Published: Nature Portfolio 2025-05-01
Series:Scientific Reports
Subjects:
Online Access:https://doi.org/10.1038/s41598-025-99278-0
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849312042029154304
author Zheng Wu
Ji-Yun Yang
Chang-Bao Yan
Cheng-Gui Zhang
Hai-Chao Yang
author_facet Zheng Wu
Ji-Yun Yang
Chang-Bao Yan
Cheng-Gui Zhang
Hai-Chao Yang
author_sort Zheng Wu
collection DOAJ
description Abstract In digital pathology, the accurate detection, segmentation, and classification of cells are pivotal for precise pathological diagnosis. Traditionally, pathologists manually segment cells from pathological images to facilitate diagnosis based on these results and other critical indicators. However, this manual approach is not only time-consuming but also prone to subjective biases, which significantly hampers its efficiency and consistency in large-scale applications. While classic segmentation networks like U-Net have gained widespread adoption in medical imaging, their integration with external prior features remains limited, thereby restricting the potential enhancement of segmentation accuracy. Although the large model SAM, renowned for its capability to “segment anything”, has shown promise, its application in the specialized field of medical image processing presents considerable challenges. Direct application of SAM to medical scenarios often fails to yield optimal results. To overcome these limitations, this paper proposes a novel prior-guided joint attention mechanism. This method effectively integrates the prior features generated by SAM with the foundational features extracted by U-Net, achieving superior cell segmentation and classification. Extensive experiments on public datasets reveal that the proposed method significantly surpasses both standalone U-Net and approaches that merely augment inputs by overlaying prior features onto color channels. This advancement not only enhances the utility of large models in medical applications but also lays the groundwork for the evolution of intelligent pathological diagnostic technologies.
format Article
id doaj-art-c9d51f5fb2344ab8a027e519426c243d
institution Kabale University
issn 2045-2322
language English
publishDate 2025-05-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-c9d51f5fb2344ab8a027e519426c243d2025-08-20T03:53:12ZengNature PortfolioScientific Reports2045-23222025-05-0115111810.1038/s41598-025-99278-0Integrating SAM priors with U-Net for enhanced multiclass cell detection in digital pathologyZheng Wu0Ji-Yun Yang1Chang-Bao Yan2Cheng-Gui Zhang3Hai-Chao Yang4College of Mathematics and Computer Science, Dali UniversityPeople’s Hospital of Dali Bai Autonomous PrefecturePeople’s Hospital of Dali Bai Autonomous PrefectureNational-Local Joint Engineering Research Center of EntomoceuticsYunnan Provincial Key Laboratory of Entomological Biopharmaceutical R&D, College of Mathematics and Computer Science, Dali UniversityAbstract In digital pathology, the accurate detection, segmentation, and classification of cells are pivotal for precise pathological diagnosis. Traditionally, pathologists manually segment cells from pathological images to facilitate diagnosis based on these results and other critical indicators. However, this manual approach is not only time-consuming but also prone to subjective biases, which significantly hampers its efficiency and consistency in large-scale applications. While classic segmentation networks like U-Net have gained widespread adoption in medical imaging, their integration with external prior features remains limited, thereby restricting the potential enhancement of segmentation accuracy. Although the large model SAM, renowned for its capability to “segment anything”, has shown promise, its application in the specialized field of medical image processing presents considerable challenges. Direct application of SAM to medical scenarios often fails to yield optimal results. To overcome these limitations, this paper proposes a novel prior-guided joint attention mechanism. This method effectively integrates the prior features generated by SAM with the foundational features extracted by U-Net, achieving superior cell segmentation and classification. Extensive experiments on public datasets reveal that the proposed method significantly surpasses both standalone U-Net and approaches that merely augment inputs by overlaying prior features onto color channels. This advancement not only enhances the utility of large models in medical applications but also lays the groundwork for the evolution of intelligent pathological diagnostic technologies.https://doi.org/10.1038/s41598-025-99278-0Large modelCell segmentationModel integrationPrior features
spellingShingle Zheng Wu
Ji-Yun Yang
Chang-Bao Yan
Cheng-Gui Zhang
Hai-Chao Yang
Integrating SAM priors with U-Net for enhanced multiclass cell detection in digital pathology
Scientific Reports
Large model
Cell segmentation
Model integration
Prior features
title Integrating SAM priors with U-Net for enhanced multiclass cell detection in digital pathology
title_full Integrating SAM priors with U-Net for enhanced multiclass cell detection in digital pathology
title_fullStr Integrating SAM priors with U-Net for enhanced multiclass cell detection in digital pathology
title_full_unstemmed Integrating SAM priors with U-Net for enhanced multiclass cell detection in digital pathology
title_short Integrating SAM priors with U-Net for enhanced multiclass cell detection in digital pathology
title_sort integrating sam priors with u net for enhanced multiclass cell detection in digital pathology
topic Large model
Cell segmentation
Model integration
Prior features
url https://doi.org/10.1038/s41598-025-99278-0
work_keys_str_mv AT zhengwu integratingsampriorswithunetforenhancedmulticlasscelldetectionindigitalpathology
AT jiyunyang integratingsampriorswithunetforenhancedmulticlasscelldetectionindigitalpathology
AT changbaoyan integratingsampriorswithunetforenhancedmulticlasscelldetectionindigitalpathology
AT chengguizhang integratingsampriorswithunetforenhancedmulticlasscelldetectionindigitalpathology
AT haichaoyang integratingsampriorswithunetforenhancedmulticlasscelldetectionindigitalpathology