W2S: Microscopy Data with Joint Denoising and Super-Resolution for Widefield to SIM Mapping

Zhou, Ruofan; Helou, Majed El; Sage, Daniel; Laroche, Thierry; Seitz, Arne; Süsstrunk, Sabine

doi:10.1007/978-3-030-66415-2_31

Cited by 18 publications

(39 citation statements)

References 50 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To validate the trained Noise2Noise plugin ML model on images that differ from the FMD dataset, we evaluate denoising plugin performance on 360 widefield fluorescence microscopy images of varying PSNR from the W2S dataset [27]. We also investigate whether the FMD trained ML plugin model applied to the W2S dataset introduces performance degradation or artifacts compared to three of the best performing self-supervised (no external training) ML methods: N2V [20], VST+BM3D [13] and BM3D [9].…”

Section: Model Validationmentioning

confidence: 99%

Section: Denoising Performance On the W2s Datasetmentioning

confidence: 99%

“…To address the fundamental problem of denoising microscopy images containing MPG noise sources, we train and evaluate well-known CNN models on MPG noise microscopy data from the FMD dataset [23] in Sections 2 and 3. In Section 4, the trained ML model is validated by evaluating performance in a diverse set of microscopy images, including the W2S fluorescence microscopy dataset [24], with out-of-training-distribution PSNR and contrast. The method consistently produces high-quality denoised images and is packaged as an ImageJ plugin to reduce the barrier-to-entry for applying denoising CNNs to microscopy data.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Real-time Image Denoising of Mixed Poisson-Gaussian Noise in Fluorescence Microscopy Images using ImageJ

Mannam

Zhang

Zhu

et al. 2021

Preprint

View full text Add to dashboard Cite

Fluorescence microscopy imaging speed is fundamentally limited by the measurement signal-to-noise ratio (SNR). To improve image SNR for a given image acquisition rate, computational denoising techniques can be used to suppress noise. However, analytical techniques to estimate a denoised image from a single frame are either computationally expensive or rely on simple noise statistical models. These models assume Poisson or Gaussian noise statistics, which are not appropriate for many fluorescence microscopy applications that contain quantum shot noise and electronic Johnson–Nyquist noise, and therefore a mixture of Poisson and Gaussian noise. In this paper, we show convolutional neural networks (CNNs) trained on mixed Poisson and Gaussian noise images to overcome the limitations of existing image denoising methods. The trained CNN is presented as an open-source ImageJ plugin that performs instant image denoising (within tens of milliseconds) with superior performance (∼8.1 dB SNR improvement) compared to the conventional fluorescence microscopy denoising methods. The method is validated on external datasets with out-of-distribution noise and contrast from the training data and consistently achieves high performance (> 8dB) denoising in less time than other fluorescence microscopy denoising methods.

show abstract

Section: Model Validationmentioning

confidence: 99%

Section: Denoising Performance On the W2s Datasetmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Real-time Image Denoising of Mixed Poisson-Gaussian Noise in Fluorescence Microscopy Images using ImageJ

Mannam

Zhang

Zhu

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…There remains scope for improvements in the performance of our deep-learning framework, most notably through the collection of larger training datasets from different biomedical applications and the development of more advanced neural network architectures . The collection of a large data set of paired low SNR, LR and high SNR, HR hyperspectral Raman images would enable the training of a joint denoising and super-resolution neural network, which we anticipate would produce improved performance in line with existing studies on multitask neural networks . Performance could potentially be further improved by implementing a generative adversarial network (GAN) architecture .…”

Section: Resultsmentioning

confidence: 95%

High-Throughput Molecular Imaging via Deep-Learning-Enabled Raman Spectroscopy

et al. 2021

View full text Add to dashboard Cite

Raman spectroscopy enables non-destructive, label-free imaging with unprecedented molecular contrast but is limited by slow data acquisition, largely preventing high-throughput imaging applications. Here, we present a comprehensive framework for higher-throughput molecular imaging via deep learning enabled Raman spectroscopy, termed DeepeR, trained on a large dataset of hyperspectral Raman images, with over 1.5 million spectra (400 hours of acquisition) in total. We firstly perform denoising and reconstruction of low signal-tonoise ratio Raman molecular signatures via deep learning, with a 9 improvement in mean squared error over state-of-the-art Raman filtering methods. Next, we develop a neural network for robust 2-4 super-resolution of hyperspectral Raman images that preserves molecular cellular information. Combining these approaches, we achieve Raman imaging speed-ups of up to 160, enabling high resolution, high signal-to-noise ratio cellular imaging in under one minute. Finally, transfer learning is applied to extend DeepeR from cell to tissuescale imaging. DeepeR provides a foundation that will enable a host of higher-throughput Raman spectroscopy and molecular imaging applications across biomedicine.

show abstract

“…After their initial breakthrough on AWGN removal, deep learning based denoising solutions were developed to improve their blind and universal aspects [8], their applicability to real images [9], or their joint application along with demosaicking [10,11], or super-resolution [12]. Comparably less progress was made with respect to the performance on the AWGN removal problem, and the understanding of the performance of a given network.…”

Section: Introductionmentioning

confidence: 99%

Deep Gaussian Denoiser Epistemic Uncertainty and Decoupled Dual-Attention Fusion

Lin

Helou

et al. 2021

2021 IEEE International Conference on Image Processing (ICIP)

View full text Add to dashboard Cite

Following the performance breakthrough of denoising networks, improvements have come chiefly through novel architecture designs and increased depth. While novel denoising networks were designed for real images coming from different distributions, or for specific applications, comparatively small improvement was achieved on Gaussian denoising. The denoising solutions suffer from epistemic uncertainty that can limit further advancements. This uncertainty is traditionally mitigated through different ensemble approaches. However, such ensembles are prohibitively costly with deep networks, which are already large in size.Our work focuses on pushing the performance limits of state-of-the-art methods on Gaussian denoising. We propose a model-agnostic approach for reducing epistemic uncertainty while using only a single pretrained network. We achieve this by tapping into the epistemic uncertainty through augmented and frequency-manipulated images to obtain denoised images with varying error. We propose an ensemble method with two decoupled attention paths, over the pixel domain and over that of our different manipulations, to learn the final fusion. Our results significantly improve over the state-of-the-art baselines and across varying noise levels.

show abstract

W2S: Microscopy Data with Joint Denoising and Super-Resolution for Widefield to SIM Mapping

Cited by 18 publications

References 50 publications

Real-time Image Denoising of Mixed Poisson-Gaussian Noise in Fluorescence Microscopy Images using ImageJ

Real-time Image Denoising of Mixed Poisson-Gaussian Noise in Fluorescence Microscopy Images using ImageJ

High-Throughput Molecular Imaging via Deep-Learning-Enabled Raman Spectroscopy

Deep Gaussian Denoiser Epistemic Uncertainty and Decoupled Dual-Attention Fusion

Contact Info

Product

Resources

About