Medical Image Segmentation Review: The Success of U-Net

Automatic medical image segmentation is a crucial topic in the medical domain and successively a critical counterpart in the computer-aided diagnosis paradigm. U-Net is the most widespread image segmentation architecture due to its flexibility, optimized modular design, and success in all medical im...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:IEEE transactions on pattern analysis and machine intelligence Ročník 46; číslo 12; s. 10076 - 10095
Hlavní autoři: Azad, Reza, Aghdam, Ehsan Khodapanah, Rauland, Amelie, Jia, Yiwei, Avval, Atlas Haddadi, Bozorgpour, Afshin, Karimijafarbigloo, Sanaz, Cohen, Joseph Paul, Adeli, Ehsan, Merhof, Dorit
Médium: Journal Article
Jazyk:angličtina
Vydáno: United States IEEE 01.12.2024
Témata:
Biomedical imaging
Computer architecture
Convolutional neural network
deep learning
Feature extraction
Humans
Image Interpretation, Computer-Assisted - methods
Image Processing, Computer-Assisted - methods
Image segmentation
Machine Learning
medical image segmentation
Neural Networks, Computer
Task analysis
Taxonomy
transformer
Transformers
U-Net
ISSN:0162-8828, 1939-3539, 2160-9292, 1939-3539
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:Automatic medical image segmentation is a crucial topic in the medical domain and successively a critical counterpart in the computer-aided diagnosis paradigm. U-Net is the most widespread image segmentation architecture due to its flexibility, optimized modular design, and success in all medical image modalities. Over the years, the U-Net model has received tremendous attention from academic and industrial researchers who have extended it to address the scale and complexity created by medical tasks. These extensions are commonly related to enhancing the U-Net's backbone, bottleneck, or skip connections, or including representation learning, or combining it with a Transformer architecture, or even addressing probabilistic prediction of the segmentation map. Having a compendium of different previously proposed U-Net variants makes it easier for machine learning researchers to identify relevant research questions and understand the challenges of the biological tasks that challenge the model. In this work, we discuss the practical aspects of the U-Net model and organize each variant model into a taxonomy. Moreover, to measure the performance of these strategies in a clinical application, we propose fair evaluations of some unique and famous designs on well-known datasets. Furthermore, we provide a comprehensive implementation library with trained models. In addition, for ease of future studies, we created an online list of U-Net papers with their possible official implementation.
Bibliografie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
ObjectType-Review-3
content type line 23
ISSN:0162-8828
1939-3539
2160-9292
1939-3539
DOI:10.1109/TPAMI.2024.3435571