Abstract: This disclosure describes systems, devices, and techniques for training neural networks to identify avascular and signal reduction areas of Optical Coherence Tomography Angiography (OCTA) images and for using trained neural networks. By identifying signal reduction areas in OCTA images, the avascular areas can be detected with high accuracy, even when the OCTA images include artifacts and other types of noise. Accordingly, various implementations described herein can accurately identify avascular areas from real-world clinical OCTA images. In various implementations, a method can include identifying images of retinas. The images may include thickness images, reflectance intensity maps, and OCTA images of the retinas. Avascular maps corresponding to the OCTA images can be identified. A neural network can be trained based on the images and the avascular maps.

Abstract: Disclosed are methods and systems correcting bulk-motion artifacts in phase-based functional OCT images. The disclosed methods and systems are based on the use of the standard deviation of the phase shift signal present in phase-based OCT imaging. When applied with functional OCT techniques such as OCT angiography, Doppler OCT, and OCT elastography, the disclosed methods provide improved image quality and decreased computational cost compared to other methods of bulk motion compensation.

Abstract: Disclosed herein are methods and systems for optical coherence tomography (OCT) angiography (OCTA). An interleaved scanning pattern is described herein for both raster and bidirectional scanning methods. The interleaved scanning pattern provides B-scans with different scanning intervals. OCTA images based on the B-scans may be combined to obtain a high dynamic range (HDR) OCTA image. Other embodiments may be described and claimed.

Abstract: Disclosed are methods and systems for signal attenuation-compensated projection-resolved (sacPR) optical coherence tomography angiography (OCTA). The sacPR OCTA may be free of segmentation and vascular contrast enhancement. In some embodiments, projection artifacts may be suppressed with signal compensation including flow and large vessel shadow compensation for projection removal and wavelet-based compensation for noise suppression.

Abstract: Methods and systems for identifying CNV membranes and vasculature in images obtained using noninvasive imaging techniques are described. An example method includes generating, by a first model based on at least one image of a retina of a subject, a membrane mask indicating a location of a CNV membrane in the retina. The method further includes generating, by a second model based on the membrane mask and the at least one image, a vasculature mask of the retina of the subject, the vasculature mask indicating CNV vascularization in the retina.

Abstract: Disclosed herein is a method for detecting retinal pathologies in three dimensions using structural and angiographic OCT. The method in accordance with the present disclosure may operate by detecting deviations in reflectance and perfusion from a depth-normalized standard retina created by merging and averaging scans from healthy subjects. In one example, the deviations from the standard retina highlight key pathologic features, while depth-normalization obviates the need to segment retinal layers. Additionally, a composite pathology index is disclosed herein that measures average deviation from the standard retina. The present method is amenable to automation and may be implemented in an integrated system and/or provided in the form of software encoded on a computer-readable medium.

Abstract: Methods and systems for identifying levels of an ophthalmic disease are described. An example method includes generating, by a convolutional neural network (CNN) and using a 3D image of a retina, a vector. The method further includes generating, by a first model and using the vector, a first likelihood that the retina exhibits a first level of an disease and generating, by a second model and using the vector, a second likelihood that the retina exhibits a second level of the disease. The method further includes determining whether the retina exhibits an absence of the ophthalmic disease, the first level of the disease, or the second level of the disease based on the first likelihood and the second likelihood. Further, an indication of whether the retina exhibits the absence of the ophthalmic disease, the first level of the ophthalmic disease, or the second level of the ophthalmic disease is output.

Abstract: Methods and systems for generating biomarker activation maps (BAMs) are described. An example method includes identifying a medical image depicting at least a portion of a subject; generating a BAM by inputting the medical image into a trained, U-shaped neural network (NN); and outputting the BAM overlaying the medical image, the BAM indicating at least one biomarker depicted in the medical image that is indicative of a disease.

Abstract: Methods for automatically identifying retinal boundaries from a reflectance image are disclosed. An example of the method includes identifying a reflectance image of the retina of a subject; generating a gradient map of the reflectance image, the gradient map representing dark-to-light or light-to-dark reflectance differentials between adjacent pixel pairs in the reflectance image; generating a guidance point array corresponding to a retinal layer boundary depicted in the reflectance image using the gradient map; generating multiple candidate paths estimating the retinal layer boundary in the reflectance image by performing a guided bidirectional graph search on the reflectance image using the guidance point array; and identifying the retinal layer boundary by merging two or more of the multiple candidate paths.

Abstract: Disclosed herein are methods and systems for self-tracking of motion artifacts (e.g., due to eye blinking and/or micro saccades motion) in optical coherence tomography (OCT) and/or OCT angiography (OCTA). A motion strength index (MSI) may be determined based on OCTA data, and the presence of excessive motion may be determined based on the MSI (e.g., if the MSI is greater than a threshold). The system may re-acquire OCT data at scan locations for which excessive motion is detected. A cross (X) scanning pattern for alignment is also described. Other embodiments may be described and claimed.

Abstract: Disclosed are systems and methods for generating wide-field optical coherence tomography angiography (OCTA) images. In embodiments, multiple OCTA scans of a sample are automatically acquired at overlapping locations. The systems and methods include functionality to adaptively control the scanning procedure such that eye blink and eye motion events are detected in real time and accounted for during 3D scan acquisition. Also disclosed are methods for detecting and correcting motion-related artifacts in OCTA datasets which allow for the longer scan times over larger fields of view required for wide-field imaging. These methods may include division of en face angiogram images into a set of motion-free parallel strips, and application of gross and fine registration methods to align overlapping strips into a motion- corrected composite image. A series of overlapping motion-corrected composite images may be combined into a larger montage to enable wide-field OCTA imaging using multiple OCTA scans.

Abstract: Disclosed are systems and methods for generating wide-field optical coherence tomography angiography (OCTA) images. In embodiments, multiple OCTA scans of a sample are automatically acquired at overlapping locations. The systems and methods include functionality to adaptively control the scanning procedure such that eye blink and eye motion events are detected in real time and accounted for during 3D scan acquisition. Also disclosed are methods for detecting and correcting motion-related artifacts in OCTA datasets which allow for the longer scan times over larger fields of view required for wide-field imaging. These methods may include division of en face angiogram images into a set of motion-free parallel strips, and application of gross and fine registration methods to align overlapping strips into a motion-corrected composite image. A series of overlapping motion-corrected composite images may be combined into a larger montage to enable wide-field OCTA imaging using multiple OCTA scans.

Abstract: A train window structure and a train with the train window structure, the train window structure comprises a train window frame; train window glass arranged on the train window frame and internally provided with a hollow structure; and a display arranged in the hollow structure to display information on the train window glass, and information is displayed on the display in the hollow structure and is provided for passengers.

Abstract: Described herein is an optical coherence tomograph (OCT) angiography technique based on the comparison of OCT signal amplitude to provide flow information. The full OCT spectrum can be split into several narrower spectral bands, resulting in the OCT resolution cell in each band being isotropic and less susceptible to axial motion nose. Inter-B-scan flow values can be determined using the individual spectral bands separately and then averaged. Recombining the flow images from the spectral bands yields angiograms that use the full information in the entire OCT spectral range. Such images provide significant improvement of signal-to-noise ratio (SNR) for both flow detection and connectivity of microvascular networks compared to other techniques. Further, creation of isotropic resolution cells can be useful for quantifying flow having equal sensitivity to axial and transverse flow.

Abstract: This disclosure describes systems, devices, and techniques for training neural networks to identify avascular and signal reduction areas of Optical Coherence Tomography Angiography (OCTA) images and for using trained neural networks. By identifying signal reduction areas in OCTA images, the avascular areas can be detected with high accuracy, even when the OCTA images include artifacts and other types of noise. Accordingly, various implementations described herein can accurately identify avascular areas from real-world clinical OCTA images. In various implementations, a method can include identifying images of retinas. The images may include thickness images, reflectance intensity maps, and OCTA images of the retinas. Avascular maps corresponding to the OCTA images can be identified. A neural network can be trained based on the images and the avascular maps.

Abstract: Disclosed herein are methods and systems for automated detection of shadow artifacts in optical coherence tomography (OCT) and/or OCT angiography (OCTA). The shadow detection includes applying a machine-learning algorithm to the OCT dataset and the OCTA dataset to detect one or more shadow artifacts in the sample. The machine-learning algorithm is trained with first training data from first training samples that include manufactured shadows and no perfusion defects and second training data from second training samples that include perfusion defects and no manufactured shadows. The shadow artifacts in the OCTA dataset and/or OCT dataset may be suppressed to generate a shadow-suppressed OCTA dataset and/or a shadow-suppressed OCT dataset, respectively. Other embodiments may be described and claimed.

Abstract: An example method includes generating, using a multi-scale block of a convolutional neural network (CNN), a first output image based on an optical coherence tomography (OCT) reflectance image of a retina and an OCT angiography (OCTA) image of the retina. The method further includes generating, using an encoder of the CNN, at least one second output image based on the first output image and generating, using a decoder of the CNN, a third output image based on the at least one second output image. An avascular map is generated based on the third output image. The avascular map indicates at least one avascular area of the retina depicted in the OCTA image.

Abstract: A train window structure and a train with the train window structure, the train window structure comprising: a train window frame; train window glass disposed in the train window frame; and a fastener penetrating through an inner side of the train window frame and disposed on a train body to fasten the train window frame, the axis of the fastener being parallel to a plane where the train window glass is provided. The fastener penetrates through the train window frame from an inner side of the train window frame and is screwed into the train body; the fastener is internally installed, thus reducing a gap between the adjacent train window structures, so that the train body is more aesthetically pleasing and wind resistance is lower.

Abstract: A train window structure and a train with the train window structure, the train window structure comprises a train window frame; train window glass arranged on the train window frame and internally provided with a hollow structure; and a display arranged in the hollow structure to display information on the train window glass, and information is displayed on the display in the hollow structure and is provided for passengers.

Abstract: Methods and systems for identifying CNV membranes and vasculature in images obtained using noninvasive imaging techniques are described. An example method includes generating, by a first model based on at least one image of a retina of a subject, a membrane mask indicating a location of a CNV membrane in the retina. The method further includes generating, by a second model based on the membrane mask and the at least one image, a vasculature mask of the retina of the subject, the vasculature mask indicating CNV vascularization in the retina.