Abstract: Systems or techniques that facilitate explainable visual attention for deep learning are provided. In various embodiments, a system can access a medical image generated by a medical imaging scanner. In various aspects, the system can perform, via execution of a deep learning neural network, an inferencing task on the medical image. In various instances, the deep learning neural network can receive as input the medical image and can produce as output both an inferencing task result and an attention map indicating on which pixels or voxels of the medical image the deep learning neural network focused in generating the inferencing task result.

Abstract: Systems/techniques that facilitate deep learning multi-planar reformatting of medical images are provided. In various embodiments, a system can access a three-dimensional medical image. In various aspects, the system can localize, via execution of a machine learning model, a set of landmarks depicted in the three-dimensional medical image, a set of principal anatomical planes depicted in the three-dimensional medical image, and a set of organs depicted in the three-dimensional medical image. In various instances, the system can determine an anatomical orientation exhibited by the three-dimensional medical image, based on the set of landmarks, the set of principal anatomical planes, or the set of organs. In various cases, the system can rotate the three-dimensional medical image, such that the anatomical orientation now matches a predetermined anatomical orientation.

Abstract: Various methods and systems are provided for transforming images from one energy level to another. In an example, a method includes obtaining an image at a first energy level acquired with a single-energy computed tomography (CT) imaging system, identifying a contrast phase of the image, entering the image as input into an energy transformation model trained to output a transformed image at a second energy level, different than the first energy level, the energy transformation model selected from among a plurality of energy transformation models based on the contrast phase, and displaying a final transformed image and/or saving the final transformed image in memory, wherein the final transformed image is the transformed image or is generated based on the transformed image.

Abstract: Methods and systems are provided for transforming images from one energy level to another. In an example, a method includes obtaining an image at a first energy level, identifying a contrast phase of the image, entering the image as input to a segmentation model trained to output an anatomy mask that identifies each tissue type in the image, generating a guide image from the image and the anatomy mask using a regression model, entering the image and the guide image as input into an energy transformation model trained to output a transformed image at a different, second energy level, the energy transformation model selected from among a plurality of energy transformation models based on the contrast phase, and displaying a final transformed image and/or saving the final transformed image in memory, wherein the final transformed image is the transformed image or is generated based on the transformed image.

Abstract: Systems/techniques that facilitate interpretable task-specific dimensionality-reduction are provided. In various embodiments, a system can access a three-dimensional medical image. In various aspects, the system can generate, via execution of a first deep learning neural network, a voxel-wise weight map corresponding to the three-dimensional medical image and a set of projection vectors corresponding to the three-dimensional medical image. In various instances, the system can generate a set of two-dimensional projection images of the three-dimensional medical image, based on the voxel-wise weight map and the set of projection vectors. In various cases, the first deep learning neural network can be trained in a serial pipeline with a second deep learning neural network that is configured to perform an inferencing task on two-dimensional inputs. This can cause the set of two-dimensional projection images to be tailored to the inferencing task.

Abstract: The current disclosure provides systems and methods for automatic image alignment of three-dimensional (3D) medical image volumes. The method includes pre-processing the 3D medical image volume by selecting a sub-volume of interest, detecting anatomical landmarks in the sub-volume using a deep neural network, estimating transformation parameters based on the anatomical landmarks to adjust rotation angles and translation of the sub-volume, adjusting the rotation angles and translation to produce a first aligned sub-volume, determining confidence in the transformation parameters based on the first aligned sub-volume, and iteratively refining the transformation parameters if the confidence is below a predetermined threshold. The disclosed approach for automated image alignment reduces the need for manual alignment and, increases a probability of the 3D image volume converging to a desired orientation compared to conventional approaches.

Abstract: Various methods and systems are provided for computed tomography imaging. In one embodiment, a method includes acquiring, with an x-ray detector and an x-ray source coupled to a gantry, a three-dimensional image volume of a subject while the subject moves through a bore of the gantry and the gantry rotates the x-ray detector and the x-ray source around the subject, inputting the three-dimensional image volume to a trained deep neural network to generate a corrected three-dimensional image volume with a reduction in aliasing artifacts present in the three-dimensional image volume, and outputting the corrected three-dimensional image volume. In this way, aliasing artifacts caused by sub-sampling may be removed from computed tomography images while preserving details, texture, and sharpness in the computed tomography images.

Abstract: Systems/techniques that facilitate explainable confidence estimation for landmark localization are provided. In various embodiments, a system can access a three-dimensional voxel array captured by a medical imaging scanner and can localize, via execution of a first deep learning neural network, a set of anatomical landmarks depicted in the three-dimensional voxel array. In various aspects, the system can generate a multi-tiered confidence score collection based on the set of anatomical landmarks and based on a training dataset on which the first deep learning neural network was trained. In various instances, the system can, in response to one or more confidence scores from the multi-tiered confidence score collection failing to satisfy a threshold, generate, via execution of a second deep learning neural network, a classification label that indicates an explanatory factor for why the one or more confidence scores failed to satisfy the threshold.

Abstract: Systems/techniques that facilitate learning-based domain transformation for medical images are provided. In various embodiments, a system can access a medical image. In various aspects, the medical image can depict an anatomical structure according to a first medical scanning domain. In various instances, the system can generate, via execution of a machine learning model, a predicted image based on the medical image. In various aspects, the predicted image can depict the anatomical structure according to a second medical scanning domain that is different from the first medical scanning domain. In some cases, the first and second medical scanning domains can be first and second energy levels of a computed tomography (CT) scanning modality. In other cases, the first and second medical scanning domains can be first and second contrast phases of the CT scanning modality.

Abstract: Various methods and systems are provided for computed tomography imaging. In one embodiment, a method includes acquiring, with an x-ray detector and an x-ray source coupled to a gantry, a three-dimensional image volume of a subject while the subject moves through a bore of the gantry and the gantry rotates the x-ray detector and the x-ray source around the subject, inputting the three-dimensional image volume to a trained deep neural network to generate a corrected three-dimensional image volume with a reduction in aliasing artifacts present in the three-dimensional image volume, and outputting the corrected three-dimensional image volume. In this way, aliasing artifacts caused by sub-sampling may be removed from computed tomography images while preserving details, texture, and sharpness in the computed tomography images.

Abstract: Systems/techniques that facilitate machine learning generation of low-noise and high structural conspicuity images are provided. In various embodiments, a system can access an image and can apply at least one of image denoising or image resolution enhancement to the image, thereby yielding a first intermediary image. In various instances, the system can generate, via execution of a plurality of machine learning models, a plurality of second intermediary images based on the first intermediary image, wherein a given machine learning model in the plurality of machine learning models receives as input the first intermediary image, wherein the given machine learning model produces as output a given second intermediary image in the plurality of second intermediary images, and wherein the given second intermediary image represents a kernel-transformed version of the first intermediary image. In various cases, the system can generate a blended image based on the plurality of second intermediary images.

Abstract: Systems/techniques that facilitate self-supervised deblurring are provided. In various embodiments, a system can access an input image generated by an imaging device. In various aspects, the system can train, in a self-supervised manner based on a point spread function of the imaging device, a machine learning model to deblur the input image. More specifically, the system can append to the model one or more non-trainable convolution layers having a blur kernel that is based on the point spread function of the imaging device. In various aspects, the system can feed the input image to the model, the model can generate a first output image based on the input image, the one or more non-trainable convolution layers can generate a second output image by convolving the first output image with the blur kernel, and the system can update parameters of the model based on a difference between the input image and the second output image.

Abstract: Methods and systems are provided for interpolating missing views in dual-energy computed tomography data. In one example, a method includes obtaining a first sinogram missing a plurality of views and a second sinogram missing a different plurality of views, the first sinogram acquired with a first X-ray source energy during a scan and the second sinogram acquired with a second, different X-ray source energy during the scan; initializing each of the first sinogram and the second sinogram to form a first initialized sinogram and a second initialized sinogram; entering the first initialized sinogram and the second initialized sinogram into the same or different interpolation models trained to output a first filled sinogram based on the first initialized sinogram and output a second filled sinogram based on the second initialized sinogram; and reconstructing one or more images from the first filled sinogram and the second filled sinogram.

Abstract: Systems/techniques that facilitate interpretable task-specific dimensionality-reduction are provided. In various embodiments, a system can access a three-dimensional medical image. In various aspects, the system can generate, via execution of a first deep learning neural network, a voxel-wise weight map corresponding to the three-dimensional medical image and a set of projection vectors corresponding to the three-dimensional medical image. In various instances, the system can generate a set of two-dimensional projection images of the three-dimensional medical image, based on the voxel-wise weight map and the set of projection vectors. In various cases, the first deep learning neural network can be trained in a serial pipeline with a second deep learning neural network that is configured to perform an inferencing task on two-dimensional inputs. This can cause the set of two-dimensional projection images to be tailored to the inferencing task.

Abstract: Systems/techniques that facilitate explainable deep interpolation are provided. In various embodiments, a system can access a data candidate, wherein a set of numerical elements of the data candidate are missing. In various aspects, the system can generate, via execution of a deep learning neural network on the data candidate, a set of weight maps for the set of missing numerical elements. In various instances, the system can compute the set of missing numerical elements by respectively combining, according to the set of weight maps, available interpolation neighbors of the set of missing numerical elements.

Abstract: Methods and systems are provided for inferring thickness and volume of one or more object classes of interest in two-dimensional (2D) medical images, using deep neural networks. In an exemplary embodiment, a thickness of an object class of interest may be inferred by acquiring a 2D medical image, extracting features from the 2D medical image, mapping the features to a segmentation mask for an object class of interest using a first convolutional neural network (CNN), mapping the features to a thickness mask for the object class of interest using a second CNN, wherein the thickness mask indicates a thickness of the object class of interest at each pixel of a plurality of pixels of the 2D medical image; and determining a volume of the object class of interest based on the thickness mask and the segmentation mask.

Abstract: Systems/techniques that facilitate deep learning robustness against display field of view (DFOV) variations are provided. In various embodiments, a system can access a deep learning neural network and a medical image. In various aspects, a first DFOV, and thus a first spatial resolution, on which the deep learning neural network is trained can fail to match a second DFOV, and thus a second spatial resolution, exhibited by the medical image. In various instances, the system can execute the deep learning neural network on a resampled version of the medical image, where the resampled version of the medical image can exhibit the first DFOV and thus the first spatial resolution. In various cases, the system can generate the resampled version of the medical image by up-sampling or down-sampling the medical image until it exhibits the first DFOV and thus the first spatial resolution.

Abstract: Methods and systems are provided for inferring thickness and volume of one or more object classes of interest in two-dimensional (2D) medical images, using deep neural networks. In an exemplary embodiment, a thickness of an object class of interest may be inferred by acquiring a 2D medical image, extracting features from the 2D medical image, mapping the features to a segmentation mask for an object class of interest using a first convolutional neural network (CNN), mapping the features to a thickness mask for the object class of interest using a second CNN, wherein the thickness mask indicates a thickness of the object class of interest at each pixel of a plurality of pixels of the 2D medical image; and determining a volume of the object class of interest based on the thickness mask and the segmentation mask.

Abstract: A computer-implemented method for correcting artifacts in computed tomography data is provided. The method includes inputting a sinogram into a trained sinogram correction network, wherein the sinogram is missing a pixel value for at least one pixel. The method also includes processing the sinogram via one or more layers of the trained sinogram correction network, wherein processing the sinogram includes deriving complementary information from the sinogram and estimating the pixel value for the at least one pixel based on the complementary information. The method further includes outputting from the trained sinogram correction network a corrected sinogram having the estimated pixel value.

Abstract: Techniques are described for generating mono-modality training image data from multi-modality image data and using the mono-modality training image data to train and develop mono-modality image inferencing models. A method embodiment comprises generating, by a system comprising a processor, a synthetic 2D image from a 3D image of a first capture modality, wherein the synthetic 2D image corresponds to a 2D version of the 3D image in a second capture modality, and wherein the 3D image and the synthetic 2D image depict a same anatomical region of a same patient. The method further comprises transferring, by the system, ground truth data for the 3D image to the synthetic 2D image. In some embodiments, the method further comprises employing the synthetic 2D image to facilitate transfer of the ground truth data to a native 2D image captured of the same anatomical region of the same patient using the second capture modality.