Applegate Matthew B, Kose Kivanc, Ghimire Sandesh, Rajadhyaksha Milind, Dy Jennifer
Northeastern University, Department of Electrical and Computer Engineering, Boston, Massachusetts, United States.
Dermatology Service at Memorial Sloan Kettering Cancer Center, New York, United States.
J Med Imaging (Bellingham). 2023 Mar;10(2):024005. doi: 10.1117/1.JMI.10.2.024005. Epub 2023 Mar 27.
Deep learning has demonstrated excellent performance enhancing noisy or degraded biomedical images. However, many of these models require access to a noise-free version of the images to provide supervision during training, which limits their utility. Here, we develop an algorithm (noise2Nyquist) that leverages the fact that Nyquist sampling provides guarantees about the maximum difference between adjacent slices in a volumetric image, which allows denoising to be performed without access to clean images. We aim to show that our method is more broadly applicable and more effective than other self-supervised denoising algorithms on real biomedical images, and provides comparable performance to algorithms that need clean images during training.
We first provide a theoretical analysis of noise2Nyquist and an upper bound for denoising error based on sampling rate. We go on to demonstrate its effectiveness in denoising in a simulated example as well as real fluorescence confocal microscopy, computed tomography, and optical coherence tomography images.
We find that our method has better denoising performance than existing self-supervised methods and is applicable to datasets where clean versions are not available. Our method resulted in peak signal to noise ratio (PSNR) within 1 dB and structural similarity (SSIM) index within 0.02 of supervised methods. On medical images, it outperforms existing self-supervised methods by an average of 3 dB in PSNR and 0.1 in SSIM.
noise2Nyquist can be used to denoise any volumetric dataset sampled at at least the Nyquist rate making it useful for a wide variety of existing datasets.
深度学习在增强有噪声或退化的生物医学图像方面表现出色。然而，这些模型中的许多需要访问图像的无噪声版本以在训练期间提供监督，这限制了它们的实用性。在此，我们开发了一种算法（noise2Nyquist），该算法利用了奈奎斯特采样可保证体积图像中相邻切片之间的最大差异这一事实，从而允许在无需访问干净图像的情况下进行去噪。我们旨在表明，我们的方法比其他自监督去噪算法在真实生物医学图像上更具广泛适用性且更有效，并且在训练期间提供与需要干净图像的算法相当的性能。
我们首先对noise2Nyquist进行理论分析，并基于采样率给出去噪误差的上限。接着，我们在模拟示例以及真实荧光共聚焦显微镜、计算机断层扫描和光学相干断层扫描图像中证明其去噪效果。
我们发现我们的方法比现有的自监督方法具有更好的去噪性能，并且适用于没有干净版本的数据集。我们的方法在峰值信噪比（PSNR）方面与监督方法相差在1dB以内，在结构相似性（SSIM）指数方面相差在0.02以内。在医学图像上，它在PSNR方面比现有的自监督方法平均高出3dB，在SSIM方面高出0.1。
noise2Nyquist可用于对至少以奈奎斯特速率采样的任何体积数据集进行去噪，使其对各种现有数据集都有用。
