当前位置: X-MOL 学术Int. J. CARS › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Learning from irregularly sampled data for endomicroscopy super-resolution: a comparative study of sparse and dense approaches.
International Journal of Computer Assisted Radiology and Surgery ( IF 2.3 ) Pub Date : 2020-05-15 , DOI: 10.1007/s11548-020-02170-7
Agnieszka Barbara Szczotka 1 , Dzhoshkun Ismail Shakir 2 , Daniele Ravì 3 , Matthew J Clarkson 1 , Stephen P Pereira 4 , Tom Vercauteren 2
Affiliation  

PURPOSE Probe-based confocal laser endomicroscopy (pCLE) enables performing an optical biopsy via a probe. pCLE probes consist of multiple optical fibres arranged in a bundle, which taken together generate signals in an irregularly sampled pattern. Current pCLE reconstruction is based on interpolating irregular signals onto an over-sampled Cartesian grid, using a naive linear interpolation. It was shown that convolutional neural networks (CNNs) could improve pCLE image quality. Yet classical CNNs may be suboptimal in regard to irregular data. METHODS We compare pCLE reconstruction and super-resolution (SR) methods taking irregularly sampled or reconstructed pCLE images as input. We also propose to embed a Nadaraya-Watson (NW) kernel regression into the CNN framework as a novel trainable CNN layer. We design deep learning architectures allowing for reconstructing high-quality pCLE images directly from the irregularly sampled input data. We created synthetic sparse pCLE images to evaluate our methodology. RESULTS The results were validated through an image quality assessment based on a combination of the following metrics: peak signal-to-noise ratio and the structural similarity index. Our analysis indicates that both dense and sparse CNNs outperform the reconstruction method currently used in the clinic. CONCLUSION The main contributions of our study are a comparison of sparse and dense approach in pCLE image reconstruction. We also implement trainable generalised NW kernel regression as a novel sparse approach. We also generated synthetic data for training pCLE SR.

中文翻译:

从不规则采样数据中学习内镜超分辨率:稀疏和密集方法的比较研究。

目的 基于探针的共聚焦激光内窥镜 (pCLE) 能够通过探针进行光学活检。pCLE 探头由成束排列的多根光纤组成,它们一起以不规则采样模式生成信号。当前的 pCLE 重建基于使用朴素线性插值将不规则信号插值到过采样笛卡尔网格上。结果表明,卷积神经网络 (CNN) 可以提高 pCLE 图像质量。然而,就不规则数据而言,经典 CNN 可能不是最理想的。方法 我们比较了以不规则采样或重建的 pCLE 图像作为输入的 pCLE 重建和超分辨率 (SR) 方法。我们还建议将 Nadaraya-Watson (NW) 内核回归嵌入到 CNN 框架中,作为新的可训练 CNN 层。我们设计了深度学习架构,允许直接从不规则采样的输入数据中重建高质量的 pCLE 图像。我们创建了合成稀疏 pCLE 图像来评估我们的方法。结果 通过基于以下指标组合的图像质量评估来验证结果:峰值信噪比和结构相似性指数。我们的分析表明,密集和稀疏的 CNN 都优于目前临床上使用的重建方法。结论 我们研究的主要贡献是比较了 pCLE 图像重建中的稀疏和密集方法。我们还实现了可训练的广义 NW 内核回归作为一种新颖的稀疏方法。我们还生成了用于训练 pCLE SR 的合成数据。
更新日期:2020-05-15
down
wechat
bug