U2-Net: A Bayesian U-Net Model With Epistemic Uncertainty Feedback For Photoreceptor Layer Segmentation In Pathological OCT Scans

@article{Orlando2019U2NetAB,
  title={U2-Net: A Bayesian U-Net Model With Epistemic Uncertainty Feedback For Photoreceptor Layer Segmentation In Pathological OCT Scans},
  author={J. Orlando and Philipp Seeb{\"o}ck and H. Bogunovi{\'c} and S. Klimscha and C. Grechenig and S. Waldstein and B. Gerendas and U. Schmidt-Erfurth},
  journal={2019 IEEE 16th International Symposium on Biomedical Imaging (ISBI 2019)},
  year={2019},
  pages={1441-1445}
}
In this paper, we introduce a Bayesian deep learning based model for segmenting the photoreceptor layer in pathological OCT scans. Our architecture provides accurate segmentations of the photoreceptor layer and produces pixel-wise epistemic uncertainty maps that highlight potential areas of pathologies or segmentation errors. We empirically evaluated this approach in two sets of pathological OCT scans of patients with age-related macular degeneration, retinal vein oclussion and diabetic macular… Expand
19 Citations
An amplified-target loss approach for photoreceptor layer segmentation in pathological OCT scans
  • 1
  • PDF
Exploiting Epistemic Uncertainty of Anatomy Segmentation for Anomaly Detection in Retinal OCT
  • 28
  • PDF
U-Net and its variants for medical image segmentation: theory and applications
  • 2
  • PDF
...
1
2
...

References

SHOWING 1-10 OF 20 REFERENCES
Pathological OCT Retinal Layer Segmentation Using Branch Residual U-Shape Networks
  • 41
  • Highly Influential
  • PDF
Bayesian SegNet: Model Uncertainty in Deep Convolutional Encoder-Decoder Architectures for Scene Understanding
  • 606
  • PDF
Exploring Uncertainty Measures in Deep Networks for Multiple Sclerosis Lesion Detection and Segmentation
  • 135
  • Highly Influential
  • PDF
Automated 3-D Intraretinal Layer Segmentation of Macular Spectral-Domain Optical Coherence Tomography Images
  • 504
  • PDF
U-Net: Convolutional Networks for Biomedical Image Segmentation
  • 21,056
  • Highly Influential
  • PDF
Artificial intelligence in retina
  • 173
Empirical Evaluation of Rectified Activations in Convolutional Network
  • 1,344
  • Highly Influential
  • PDF
...
1
2
...