DeepLSR: a deep learning approach for laser speckle reduction

Abstract

Speckle artifacts degrade image quality in virtually all modalities that utilize coherent energy, including optical coherence tomography, reflectance confocal microscopy, ultrasound, and widefield imaging with laser illumination. We present an adversarial deep learning framework for laser speckle reduction, called DeepLSR (https://durr.jhu.edu/DeepLSR), that transforms images from a source domain of coherent illumination to a target domain of speckle-free, incoherent illumination. We apply this method to widefield images of objects and tissues illuminated with a multi-wavelength laser, using light emitting diode-illuminated images as ground truth. In images of gastrointestinal tissues, DeepLSR reduces laser speckle noise by 6.4 dB, compared to a 2.9 dB reduction from optimized non-local means processing, a 3.0 dB reduction from BM3D, and a 3.7 dB reduction from an optical speckle reducer utilizing an oscillating diffuser. Further, DeepLSR can be combined with optical speckle reduction to reduce speckle noise by 9.4 dB. This dramatic reduction in speckle noise may enable the use of coherent light sources in applications that require small illumination sources and high-quality imaging, including medical endoscopy.

Scattering oblique plane microscopy for in-vivo blood cell imaging

“…Additionally, it is clear from our in vitro blood cell phantom results that speckle noise is degrading the signal. Thus, the use of a higher power broadband source or other means of speckle reduction such as a commercial laser speckle reducer, rotating diffuser, or despeckling algorithms [33] are worth investigating. Finally, volumetric tissue imaging can be achieved by incorporating scanning and descanning galvanometric mirrors into the sOPM design, similar to how SCAPE is used for fluorescence imaging [34].…”

Development of a 3-D Physical Dynamics Monitoring System Using OCM with DVC for Quantification of Sprouting Endothelial Cells Interacting with a Collagen Matrix

“…Speckles, a feature of coherence imaging, are accompanied by noise and actual structural information of the media. Many techniques, such as image filtering, [57,58], artificial intelligence [59], or spectral compounding [60,61], have been reported to achieve visual improvements. Additionally, hardware-based methods such as optical chopper [62] or angular compounding [63] have been suggested for diminishing the effect of speckles.…”

Deep Learning in Biomedical Optics

“…Brightness can also be increased via laser‐illumination, which allows greater coupling efficiency than incoherent sources, but results in laser speckle noise in the image from coherent interference. Conditional GANs have been applied to predict speckle‐free images from laser‐illumination endoscopy images by training on image pairs acquired of the same tissue with both coherent and incoherent illumination sources [140].…”