6533b873fe1ef96bd12d4535
RESEARCH PRODUCT
Multi-Path U-Net Architecture for Cell and Colony-Forming Unit Image Segmentation
Alexey LihachevVilen JumutcDmitrijs Bliznukssubject
Layer Normalizationneural networkChemical technologyStem CellsTP1-1185U-NetBiochemistryencoder–decoderAtomic and Molecular Physics and OpticsAnalytical Chemistryskip-connectionsImage Processing Computer-AssistedNeural Networks ComputerU-Net; skip-connections; neural network; encoder–decoder; Layer NormalizationElectrical and Electronic EngineeringInstrumentationdescription
U-Net is the most cited and widely-used deep learning model for biomedical image segmentation. In this paper, we propose a new enhanced version of a ubiquitous U-Net architecture, which improves upon the original one in terms of generalization capabilities, while addressing several immanent shortcomings, such as constrained resolution and non-resilient receptive fields of the main pathway. Our novel multi-path architecture introduces a notion of an individual receptive field pathway, which is merged with other pathways at the bottom-most layer by concatenation and subsequent application of Layer Normalization and Spatial Dropout, which can improve generalization performance for small datasets. In general, our experiments show that the proposed multi-path architecture outperforms other state-of-the-art approaches that embark on similar ideas of pyramid structures, skip-connections, and encoder–decoder pathways. A significant improvement of the Dice similarity coefficient is attained at our proprietary colony-forming unit dataset, where a score of 0.809 was achieved for the foreground class.
year | journal | country | edition | language |
---|---|---|---|---|
2022-01-27 | Sensors; Volume 22; Issue 3; Pages: 990 |