Adapting SAM to Histopathology Images for Tumor Bud Segmentation in Colorectal Cancer

Ziyu Su, Wei Chen, Sony Annem, Usama Sajjad, Mostafa Rezapour, Wendy L. Frankel, Metin N. Gurcan, M. Khalid Khan Niazi

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

2 Scopus citations

Abstract

Colorectal cancer (CRC) is the third most common cancer in the United States. Tumor Budding (TB) detection and quantification are crucial yet labor-intensive steps in determining the CRC stage through the analysis of histopathology images. To help with this process, we adapt the Segment Anything Model (SAM) on the CRC histopathology images to segment TBs using SAM-Adapter. In this approach, we automatically take task-specific prompts from CRC images and train the SAM model in a parameter-efficient way. We compare the predictions of our model with the predictions from a trained-from-scratch model using the annotations from a pathologist. As a result, our model achieves an intersection over union (IoU) of 0.65 and an instance-level Dice score of 0.75, which are promising in matching the pathologist’s TB annotation. We believe our study offers a novel solution to identify TBs on H&E-stained histopathology images. Our study also demonstrates the value of adapting the foundation model for pathology image segmentation tasks.

Original languageEnglish
Title of host publicationMedical Imaging 2024
Subtitle of host publicationDigital and Computational Pathology
EditorsJohn E. Tomaszewski, Aaron D. Ward
PublisherSPIE
ISBN (Electronic)9781510671706
DOIs
StatePublished - 2024
EventMedical Imaging 2024: Digital and Computational Pathology - San Diego, United States
Duration: Feb 19 2024Feb 21 2024

Publication series

NameProgress in Biomedical Optics and Imaging - Proceedings of SPIE
Volume12933
ISSN (Print)1605-7422

Conference

ConferenceMedical Imaging 2024: Digital and Computational Pathology
Country/TerritoryUnited States
CitySan Diego
Period02/19/2402/21/24

Keywords

  • Colorectal cancer
  • attention mechanism
  • cross-attention network
  • deep learning
  • histopathology images
  • multiple instance learning
  • tumor budding
  • weakly supervised learning

Fingerprint

Dive into the research topics of 'Adapting SAM to Histopathology Images for Tumor Bud Segmentation in Colorectal Cancer'. Together they form a unique fingerprint.

Cite this