Segmentation of Intracellular Structures in Fluorescence Microscopy Images by Fusing Low-Level Features

Guo, Yuanhao; Huang, Jiaxing; Zhou, Yanfeng; Luo, Yaoru; Li, Wenjing; Yang, Ge

doi:10.1007/978-3-030-88010-1_32

Cited by 4 publications

(8 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This indicates that models optimized for segmentation accuracy may have poor robustness, consistent with findings of previous studies on image classification [64][65][66]. Second, UNet 3 and Sim UNet, have slightly better accuracy than UNet, as reported in [16], but show poor robustness against noise or blurring. This indicates that models should not be optimized solely for accuracy.…”

Section: Discussionsupporting

confidence: 84%

Benchmarking Robustness of Deep Neural Networks in Semantic Segmentation of Fluorescence Microscopy Images

Zhong,

Li,

Yang

2024

Preprint

Self Cite

View full text Add to dashboard Cite

Fluorescence microscopy (FM) is an important and widely adopted biological imaging technique. Segmentation is generally the first step in quantitative analysis of FM images. Although deep neural networks (DNNs) have become the state-of-the-art tools for image segmentation, their performance on natural images may collapse under certain image corruptions or adversarial attacks. This poses real risks to their deployment in real-world applications. Although the robustness of DNN models in segmenting natural images has been studied extensively, their robustness in segmenting FM images remains poorly understood. To address this deficiency, we have developed an assay that benchmarks robustness of DNN segmentation models using both realistic synthetic and real FM images with precisely controlled corruptions or adversarial attacks. Specifically, we have benchmarked robustness of ten representative models such as DeepLab and Vision Transformer in semantic segmentation of FM images. We find that models with good robustness on natural images may perform poorly on FM images. We also find new robustness properties of DNN models and new connections between their corruption robustness and adversarial robustness. Based on comprehensive benchmarking of robustness of the selected models, we make specific recommendations on selection and design of robust models for segmentation of FM images.

show abstract

Section: Discussionsupporting

confidence: 84%

Benchmarking Robustness of Deep Neural Networks in Semantic Segmentation of Fluorescence Microscopy Images

Zhong,

Li,

Yang

2024

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…This indicates that models optimized for segmentation accuracy may have poor robustness, consistent with findings of previous studies on image classification [59][60][61]. Second, UNet_3 and Sim_UNet, have slightly better accuracy than UNet, as reported in [16], but show poor robustness against noise or blurring. This indicates that models should not be optimized solely for accuracy.…”

Section: ) Summarysupporting

confidence: 85%

“…Accuracy on clean images: Sim_UNet achieves overall the best accuracy on clean images with the highest mIoU at 81.84%, consistent with finding in [16] that the simplified models achieve higher accuracy on FM images than the original model. ICNet provides overall the worst accuracy at 62.42% on clean images.…”

Section: ) Performance Under Different Types Of Corruptionssupporting

confidence: 83%

“…Concatenated features largely preserve the information of each encoding layer, making segmentation more accurate. Recently, it has been shown that simplification of the U-Net by reducing its level of downsampling and up-sampling improves segmentation accuracy on FM images [16]. To check how such simplification may influence model robustness, we test two simplified variants of the UNet: UNet_3 retains the first three encode-decode layers of the original UNet and removes the last two layers, whereas Sim_UNet further reduces parameters of each layer of UNet_3 to obtain a more simplified architecture [16].…”

Section: B Segmentation Modelsmentioning

confidence: 99%

“…In contrast, natural images often have multiple semantic classes. Although DNN models have found great success in segmentation of FM images [15,16], most of them are developed originally for natural images. The differences between natural images and FM images raise the question of whether conclusions on robustness of DNN models in segmentation of natural image are still valid on FM images.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Characterizing Robustness of Deep Neural Networks in Semantic Segmentation of Fluorescence Microscopy Images

Liqun,

Li,

Yang

2022

Preprint

Self Cite

View full text Add to dashboard Cite

<p>Fluorescence microscopy (FM) is an imaging technique with many important applications in biomedical sciences. After FM images are acquired, segmentation is often the first step in their quantitative analysis. Although deep neural networks (DNNs) have become the state-of-the-art tools for segmentation, it is known that their performance may collapse on natural images under certain corruptions or adversarial attacks. This poses serious risks to their deployment in real-world applications. Although various assays have been developed to benchmark the robustness of different DNN models in semantic segmentation of natural images, such assays remain lacking for FM images. So far, robustness of DNN models in sematic segmentation of FM images remains to be characterized. In this study, we have developed an assay to address this deficiency. At the core of the assay is a method we have developed to synthesize realistic FM images with precisely controlled forms and levels of degradations. Using this assay, we examine the robustness of DNN models against corruptions and adversarial attacks of FM images. We find that models with good robustness on natural images may perform poorly on FM images. We also find new robustness properties of DNN models and new connections between their corruption robustness and adversarial robustness. Based on comprehensive comparison of eight representative models, we make specific recommendations on which models to choose and how to design robust models for segmentation of FM images.</p>

show abstract

Fluorescence Microscopy Images Segmentation Based on Prototypical Networks with a Few Annotations

Guo

Luo

et al. 2022

Pattern Recognition and Computer Vision

View full text Add to dashboard Cite

Segmentation of Intracellular Structures in Fluorescence Microscopy Images by Fusing Low-Level Features

Cited by 4 publications

References 21 publications

Benchmarking Robustness of Deep Neural Networks in Semantic Segmentation of Fluorescence Microscopy Images

Benchmarking Robustness of Deep Neural Networks in Semantic Segmentation of Fluorescence Microscopy Images

Characterizing Robustness of Deep Neural Networks in Semantic Segmentation of Fluorescence Microscopy Images

Fluorescence Microscopy Images Segmentation Based on Prototypical Networks with a Few Annotations

Contact Info

Product

Resources

About