A Two-Stage U-Net Algorithm for Segmentation of Nuclei in H&amp;E-Stained Tissues

Mahbod, Amirreza; Schaefer, Gerald; Ellinger, Isabella; Ecker, Rupert; Smedby, Örjan

doi:10.1007/978-3-030-23937-4_9

Cited by 29 publications

(20 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our method outperformed present state-of-the-art methods on the two datasets (described in section "Datasets") in the integrity of the segmentation of a single nucleus and the segmentation accuracy, and especially in the segmentation of overlapped nuclei regions. We compared our method against several deep learning based methods listed in Table 1, such as FCN-8 (Long et al, 2015), Mask R-CNN (He et al, 2015), U-Net (Ronneberger et al, 2015), CNN3 (Kumar et al, 2017), DIST (Naylor et al, 2019), SUNets, U-Net (DLA), a two-stage U-net (Mahbod et al, 2019), and two-stage learning U-Net (DLA) (Kang et al, 2019). In order to make the comparison objectively, we followed the same training and testing set split criteria suggested by Kumar et al (2017).…”

Section: Resultsmentioning

confidence: 99%

Nuclear Segmentation in Histopathological Images Using Two-Stage Stacked U-Nets With Attention Mechanism

Kong

Genchev

Wang

et al. 2020

Front. Bioeng. Biotechnol.

View full text Add to dashboard Cite

Nuclei segmentation is a fundamental but challenging task in histopathological image analysis. One of the main problems is the existence of overlapping regions which increases the difficulty of independent nuclei separation. In this study, to solve the segmentation of nuclei and overlapping regions, we introduce a nuclei segmentation method based on two-stage learning framework consisting of two connected Stacked U-Nets (SUNets). The proposed SUNets consists of four parallel backbone nets, which are merged by the attention generation model. In the first stage, a Stacked U-Net is utilized to predict pixel-wise segmentation of nuclei. The output binary map together with RGB values of the original images are concatenated as the input of the second stage of SUNets. Due to the sizable imbalance of overlapping and background regions, the first network is trained with cross-entropy loss, while the second network is trained with focal loss. We applied the method on two publicly available datasets and achieved stateof-the-art performance for nuclei segmentation-mean Aggregated Jaccard Index (AJI) results were 0.5965 and 0.6210, and F1 scores were 0.8247 and 0.8060, respectively; our method also segmented the overlapping regions between nuclei, with average AJI = 0.3254. The proposed two-stage learning framework outperforms many current segmentation methods, and the consistent good segmentation performance on images from different organs indicates the generalized adaptability of our approach.

show abstract

Section: Resultsmentioning

confidence: 99%

Nuclear Segmentation in Histopathological Images Using Two-Stage Stacked U-Nets With Attention Mechanism

Kong

Genchev

Wang

et al. 2020

Front. Bioeng. Biotechnol.

View full text Add to dashboard Cite

show abstract

“…1) U-Net: The U-Net architecture is prominently used in nuclear image segmentation [31], [9], [28]. The success of this architecture is based on the fact that accurate segmentation is possible even with small training sets.…”

Section: Data Augmentation and Artificial Image Synthesismentioning

confidence: 99%

“…The annotated dataset used is publicly available at the EMBL BioStudies database, accession number S-BSST265 [12], a detailed description is available [13]. The code used to evaluate the segmentation methods is publicly available 9 .…”

Section: F Data and Code Availabilitymentioning

confidence: 99%

“…oversegmented objects) to the number of ground truth objects. 9 https://github.com/perlfloccri/NuclearSegmentationPipeline 10 Examples of possible cases of ground truth objects and predictions are illustrated in Suppl. Fig.…”

Section: Evaluation Metricsmentioning

confidence: 99%

“…Despite the success of applying the U-Net architectures to nuclear image segmentation, nuclei in tight aggregations sometimes lead to under-segmentation. To overcome this, post-processing steps are commonly applied to a predicted initial nucleus segmentation [7], [8], [9], relying on morphological or intensity based features derived from segmented objects. These features are further used to identify under-segmented objects and to guide the separation or merging process.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Evaluation of Deep Learning Architectures for Complex Immunofluorescence Nuclear Image Segmentation

Kromp

Fischer

Bozsaky

et al. 2021

IEEE Trans. Med. Imaging

View full text Add to dashboard Cite

Separating and labeling each nuclear instance (instance-aware segmentation) is the key challenge in nuclear image segmentation. Deep Convolutional Neural Networks have been demonstrated to solve nuclear image segmentation tasks across different imaging modalities, but a systematic comparison on complex immunofluorescence images has not been performed. Deep learning based segmentation requires annotated datasets for training, but annotated fluorescence nuclear image datasets are rare and of limited size and complexity. In this work, we evaluate and compare the segmentation effectiveness of multiple deep learning architectures (U-Net, U-Net ResNet, Cellpose, Mask R-CNN, KG instance segmentation) and two conventional algorithms (Iterative h-min based watershed, Attributed relational graphs) on complex fluorescence nuclear images of various types.We propose and evaluate a novel strategy to create artificial images to extend the training set. Results show that instanceaware segmentation architectures and Cellpose outperform the U-Net architectures and conventional methods on complex images in terms of F1 scores, while the U-Net architectures achieve overall higher mean Dice scores. Training with additional artificially generated images improves recall and F1 scores for complex images, thereby leading to top F1 scores for three out of five sample preparation types. Mask R-CNN trained on artificial images achieves the overall highest F1 score on complex images of similar conditions to the training set images while Cellpose achieves the overall highest F1 score on complex images of new imaging conditions. We provide quantitative results demonstrat-Manuscript resubmitted on February 13, 2021.

show abstract

Comparison of Deep Learning Preprocessing Algorithms of Nuclei Segmentation on Fluorescence Immunohistology Images of Cancer Cells

Skakun

2022

Communications in Computer and Information Science

View full text Add to dashboard Cite

A Two-Stage U-Net Algorithm for Segmentation of Nuclei in H&E-Stained Tissues

Cited by 29 publications

References 9 publications

Nuclear Segmentation in Histopathological Images Using Two-Stage Stacked U-Nets With Attention Mechanism

Nuclear Segmentation in Histopathological Images Using Two-Stage Stacked U-Nets With Attention Mechanism

Evaluation of Deep Learning Architectures for Complex Immunofluorescence Nuclear Image Segmentation

Comparison of Deep Learning Preprocessing Algorithms of Nuclei Segmentation on Fluorescence Immunohistology Images of Cancer Cells

Contact Info

Product

Resources

About