Optical coherence tomography can provide visualizations of the eye both in diagnostic and surgical settings. However, noise limits the achievable image quality, especially in scenarios in which multi-frame averaging is not available. In this work, we present high-quality OCT image denoising using deep learning, only requiring unpaired volumetric capture scans for training. It is shown that, by exploiting neighboring B-scans, an artificial neural network for denoising OCT images can be trained based on a state-of-the-art approach which usually requires repeated scans from the exact same location. The effect of denoising is demonstrated for B-scans and volumetric renderings during and after mock cataract surgery on ex-vivo porcine eyes.
Four-dimensional microscope integrated optical coherence tomography (4D-miOCT) has been proposed as an alternative to conventional white-light microscopy during ophthalmic surgical interventions. Its real-time visualization capability of 3D data constitutes one of the most promising visualization techniques for many surgical use cases in ophthalmology. In this work, we conducted a comprehensive user study for optimal visualization with the highest performance use of the 4D-miOCT data, comparing an autostereoscopic light field tablet to a 3D TV. With the feedback collected as part of a user study, we are able to further optimize how we display 4D-miOCT data to surgeons.
As in other imaging modalities, noise decreases image quality in optical coherence tomography (OCT), which is especially problematic in real-time intra-surgical application, where multi-frame averaging is not available. In this work, we present an adapted self-supervised training approach to train a blind-spot denoising network for OCT data. With the proposed method, the stability of the method is improved, avoiding the occurrence of artifacts by increasing realism of training data. We show that using this approach, the quality of two-dimensional B-scans can be improved qualitatively and quantitatively even without paired training data. This improvement is also translated into live volumetric renderings composed of denoised two-dimensional scans, even when using only very small network complexities due to harsh time constraints.
Noise decreases image quality in optical coherence tomography (OCT) and can obscure important features in real-time visualizations. In this work, we show that a neural network can be applied to denoise volumetric OCT data for intra-surgical visualization in real-time. We adapt a self-supervised training approach, not requiring any paired data for training. Several optimizations and trade-offs in deployment are required, with which we achieved processing times of only few milliseconds. While still being limited by the real-time requirements, denoising in this scenario can enhance surface visibility, and therefore allow guidance for more precise intra-surgical maneuvers.
Increasing the FOV of OCTA images while keeping the acquisition time moderate requires high A-scan rates. Therefore, OCTA images appear to be noisier. Deep learning methods can be used for noise reduction. In OCTA volumes small vessels with an orientation perpendicular to the image plane are often removed by deep learning denoising algorithms, due to their small appearance.
To overcome this a 3-dimensional Unet was developed to utilize volumetric information. With the knowledge of also the third dimension, the algorithm is able to distinguish between noise and vessel contrast and is therefore less likely to remove vessels.
We present a flexible OCT engine for acquiring full eye-length, anterior and posterior segment B-scans, as well as 4D live volumes with an effective A-scan rate of up to 2MHz. It is enabled by a MEMS tunable VCSEL with flexible A-scan rates, broad spectral bandwidth and a long instantaneous coherence length. Our GPU based, custom reconstruction and rendering software is able to process and display live volume series at rates of up to 17 volumes per second. We show B-scans and volume series of model eyes.
Tunable laser sources with sweep-rates higher than 1MHz recently became commercially available. Today’s commercial ophthalmic OCT systems use sweep-rates in the 100-200kHz regime. These much faster laser sources can be used to either significantly reduce the imaging time or significantly increase the field of view (FOV). In this study we investigate the clinical value of OCT with MHz-rate swept source lasers. We implemented a versatile ophthalmic OCT system using a Frequency-Domain-Mode-Locked (FDML) laser with a sweep-rate of 1.7MHz, to address a variety of ophthalmic OCT imaging applications, exhibiting large imaging depth for wide field retinal OCT and OCT angiography (OCTA) with a field of view of up to 90 degrees, as well as for anterior segment imaging, and microscopic OCTA of the choriocapillaris with repetition rates of more than 1kHz.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.