Search
Abstract
Accurate segmentation of cellular nuclei is imperative for various biological and medical applications, such as cancer diagnosis and drug discovery. Histopathology, a discipline employing microscopic examination of bodily tissues, serves as a cornerstone for cancer diagnosis. Nonetheless, the conventional histopathological diagnosis process is frequently marred by time constraints and potential inaccuracies. Consequently, there arises a pressing need for automated image analysis tools to augment medical practitioners’ efforts. In this paper, we present a novel approach utilising Transformer model, originally designed for natural language processing tasks, for automated cellular nuclei segmentation in whole-slide microscopic images. Specifically targeting cell nuclei, this methodology holds significance as the initial phase in diagnosing various illnesses, streamlining the analysis and quantification process. The study introduces a novel model that combines a U-Net architecture with a Transformer-based network functioning as a parallel encoder. This model was compared against three other popular architectures in the literature: U-Net, ResU-Net, and LinkNet-34. The impact of augmentation and colour normalisation techniques was investigated. The average Dice similarity coefficient for the considered images was found to be 0.8041.
Citations
-
0
CrossRef
-
0
Web of Science
-
0
Scopus
Authors (2)
-
Mateusz Erezman
Cite as
Full text
full text is not available in portal
Keywords
Details
- Category:
- Conference activity
- Type:
- publikacja w wydawnictwie zbiorowym recenzowanym (także w materiałach konferencyjnych)
- Language:
- English
- Publication year:
- 2024
- Bibliographic description:
- Erezman M., Dziubich T.: Deep Learning-Based Cellular Nuclei Segmentation Using Transformer Model// / : , 2024,
- DOI:
- Digital Object Identifier (open in new tab) 10.1007/978-3-031-70421-5_18
- Sources of funding:
-
- Statutory activity/subsidy
- Verified by:
- Gdańsk University of Technology
seen 22 times
Recommended for you
Smart Approach for Glioma Segmentation in Magnetic Resonance Imaging using Modified Convolutional Network Architecture (U-NET)
- N. Sohail,
- S. M. Anwar,
- F. Majeed
- + 1 authors