Abstract
A coherent anti-Stokes Raman scattering (CARS) rigid endoscope was developed to visualize peripheral nerves without labeling for nerve-sparing endoscopic surgery. The developed CARS endoscope had a problem with low imaging speed, i.e. low imaging rate. In this study, we demonstrate that noise reduction with deep learning boosts the nerve imaging speed with CARS endoscopy. We employ fine-tuning and ensemble learning and compare deep learning models with three different architectures. In the fine-tuning strategy, deep learning models are pre-trained with CARS microscopy nerve images and retrained with CARS endoscopy nerve images to compensate for the small dataset of CARS endoscopy images. We propose using the equivalent imaging rate (EIR) as a new evaluation metric for quantitatively and directly assessing the imaging rate improvement by deep learning models. The highest EIR of the deep learning model was 7.0 images/min, which was 5 times higher than that of the raw endoscopic image of 1.4 images/min. We believe that the improvement of the nerve imaging speed will open up the possibility of reducing postoperative dysfunction by intraoperative nerve identification.
Highlights
Visualization of nerves is one of the problems to be solved in nerve-sparing surgery
We propose a new evaluation metric called the equivalent imaging rate (EIR), and use the EIR to quantitatively evaluate the improvement of the imaging rate
Since the high numerical aperture (NA) of a coherent anti-Stokes Raman scattering (CARS) microscopy system increases the imaging rate and facilitates nerve exploration, CARS microscopy can provide a large dataset of nerve images compared with CARS endoscopy
Summary
Visualization of nerves is one of the problems to be solved in nerve-sparing surgery. One alternative approach to label-free imaging is Raman microscopy; nerve imaging with conventional spontaneous Raman scattering requires a long exposure time because of the considerably low efficiency of Raman scattering. A label-free, high-imaging-rate method is desired for intraoperative nerve visualization. We have developed a coherent anti-Stokes Raman scattering (CARS) rigid endoscope for visualizing myelinated nerves without labeling[8,9]. We achieved the visualization of rat sciatic nerves and rabbit periprostatic nerves with the developed CARS endoscope. The low NA (0.26) of our CARS endoscope, which was used to increase the field of view to 650 μm , limited the imaging rate. We apply deep learning to denoising in order to improve the nerve imaging rate with the developed CARS rigid endoscope. We propose a new evaluation metric called the equivalent imaging rate (EIR), and use the EIR to quantitatively evaluate the improvement of the imaging rate
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
