Abstract: For COPD classification in a medical imaging system, machine learning is used to learn to classify whether a patient has COPD. An image-to-image network deep learns spatial features indicative of various or any type of COPD. The pulmonary function test may be used as the ground truth in training the features and classification from the spatial features. Due to the high availability of pulmonary function test results and corresponding CT scans, there are many training samples. Values from learned features of the image-to-image network are then used to create a spatial distribution of level of COPD, providing information useful for distinguishing between types of COPD without requiring ground truth annotation of spatial distribution of COPD in the training.

Abstract: Methods and systems for artificial intelligence based medical image segmentation are disclosed. In a method for autonomous artificial intelligence based medical image segmentation, a medical image of a patient is received. A current segmentation context is automatically determined based on the medical image and at least one segmentation algorithm is automatically selected from a plurality of segmentation algorithms based on the current segmentation context. A target anatomical structure is segmented in the medical image using the selected at least one segmentation algorithm.

Abstract: Systems and methods for generating synthesized images are provided. An input medical image patch, a segmentation mask, a vector of appearance related parameters, and manipulable properties are received. A synthesized medical image patch including a synthesized nodule is generated based on the input medical image patch, the segmentation mask, the vector of appearance related parameters, and the manipulable properties using a trained object synthesis network. The synthesized nodule is synthesized according to the manipulable properties. The synthesized medical image patch is output.

Abstract: Medical image data may be applied to a machine-learned network learned on training image data and associated image segmentations, landmarks, and view classifications to classify a view of the medical image data, detect a location of one or more landmarks in the medical image data, and segment a region in the medical image data based on the application of the medical image data to the machine-learned network. The classified view, the segmented region, or the location of the one or more landmarks may be output.

Abstract: Methods and systems for artificial intelligence based medical image segmentation are disclosed. In a method for autonomous artificial intelligence based medical image segmentation, a medical image of a patient is received. A current segmentation context is automatically determined based on the medical image and at least one segmentation algorithm is automatically selected from a plurality of segmentation algorithms based on the current segmentation context. A target anatomical structure is segmented in the medical image using the selected at least one segmentation algorithm.

Abstract: Systems and methods are provided for generating a synthesized medical image patch of a nodule. An initial medical image patch and a class label associated with a nodule to be synthesized are received. The initial medical image patch has a masked portion and an unmasked portion. A synthesized medical image patch is generated using a trained generative adversarial network. The synthesized medical image patch includes the unmasked portion of the initial medical image patch and a synthesized nodule replacing the masked portion of the initial medical image patch. The synthesized nodule is synthesized according to the class label. The synthesized medical image patch is output.

Abstract: Embodiments of the invention relate to a method of processing a medical image to remove one or more portions of the image corresponding to bone structures, the method comprising: receiving first image data representing a first, three-dimensional, medical image; processing the first image data to generate second image data representing a plurality of two-dimensional image channels each corresponding to a different slice of the first medical image; receiving the second image data at a neural network system; applying an attention mechanism at the neural network system to the second image data to generate an attention map representing one or more regions of interest; and determining, at least partly on the basis of the attention map, that one or more portions of the second image data represent a bone structure.

Abstract: Systems and methods for determining whether a bone of a patient is injured are provided. A medical image of a bone of a patient is received. A synthesized bone image is generated over the bone in the medical image to provide a reconstructed image. The synthesized bone image represents uninjured bone. The medical image is compared with the reconstructed image to evaluate an injury to the bone of the patient.

Abstract: For COPD classification in a medical imaging system, machine learning is used to learn to classify whether a patient has COPD. An image-to-image network deep learns spatial features indicative of various or any type of COPD. The pulmonary function test may be used as the ground truth in training the features and classification from the spatial features. Due to the high availability of pulmonary function test results and corresponding CT scans, there are many training samples. Values from learned features of the image-to-image network are then used to create a spatial distribution of level of COPD, providing information useful for distinguishing between types of COPD without requiring ground truth annotation of spatial distribution of COPD in the training.

Abstract: Methods and apparatus for automated medical image analysis using deep learning networks are disclosed. In a method of automatically performing a medical image analysis task on a medical image of a patient, a medical image of a patient is received. The medical image is input to a trained deep neural network. An output model that provides a result of a target medical image analysis task on the input medical image is automatically estimated using the trained deep neural network. The trained deep neural network is trained in one of a discriminative adversarial network or a deep image-to-image dual inverse network.

Abstract: Methods and apparatus for automated medical image analysis using deep learning networks are disclosed. In a method of automatically performing a medical image analysis task on a medical image of a patient, a medical image of a patient is received. The medical image is input to a trained deep neural network. An output model that provides a result of a target medical image analysis task on the input medical image is automatically estimated using the trained deep neural network. The trained deep neural network is trained in one of a discriminative adversarial network or a deep image-to-image dual inverse network.

Abstract: Embodiments can provide a method for atlas-based contouring, comprising constructing a relevant atlas database; selecting one or more optimal atlases from the relevant atlas database; propagating one or more atlases; fusing the one or more atlases; and assessing the quality of one or more propagated contours.

Abstract: Systems and methods are provided for automatic detection and quantification for traumatic bleeding. Image data is acquired using a full body dual energy CT scanner. A machine-learned network detects one or more bleeding areas on a bleeding map from the dual energy CT scan image data. A visualization is generated from the bleeding map. The predicted bleeding areas are quantified, and a risk value is generated. The visualization and risk value are presented to an operator.

Abstract: Systems and methods for determining whether a bone of a patient is injured are provided. A medical image of a bone of a patient is received. A synthesized bone image is generated over the bone in the medical image to provide a reconstructed image. The synthesized bone image represents uninjured bone. The medical image is compared with the reconstructed image to evaluate an injury to the bone of the patient.

Abstract: A method and apparatus for deep learning based automatic bone removal in medical images, such as computed tomography angiography (CTA) volumes, is disclosed. Bone structures are segmented in a 3D medical image of a patient by classifying voxels of the 3D medical image as bone or non-bone voxels using a deep neural network trained for bone segmentation. A 3D visualization of non-bone structures in the 3D medical image is generated by removing voxels classified as bone voxels from a 3D visualization of the 3D medical image.

Abstract: Systems and methods are provided for segmenting tissue within a computed tomography (CT) scan of a region of interest into one of a plurality of tissue classes. A plurality of atlases are registered to the CT scan to produce a plurality of registered atlases. A context model representing respective likelihoods that each voxel of the CT scan is a member of each of the plurality of tissue classes is determined from the CT scan and a set of associated training data. A proper subset of the plurality of registered atlases is selected according to the context model and the registered atlases. The selected proper subset of registered atlases are fused to produce a combined segmentation.

Abstract: Medical image data may be applied to a machine-learned network learned on training image data and associated image segmentations, landmarks, and view classifications to classify a view of the medical image data, detect a location of one or more landmarks in the medical image data, and segment a region in the medical image data based on the application of the medical image data to the machine-learned network. The classified view, the segmented region, or the location of the one or more landmarks may be output.

Abstract: A method for training a segmentation correction model includes performing an iterative model training process over a plurality of iterations. During each iteration, an initial segmentation estimate for an image is provided to a human annotators via an annotation interface. The initial segmentation estimate identifies one or more anatomical areas of interest within the image. Interactions with the annotation interface are automatically monitored to record annotation information comprising one or more of (i) segmentation corrections made to the initial segmentation estimate by the annotators via the annotation interface, and (ii) interactions with the annotation interface performed by the annotators while making the corrections. A base segmentation machine learning model is trained to automatically create a base segmentation based on the image. Additionally, a segmentation correction machine learning model is trained to automatically perform the segmentation corrections based on the image.

Abstract: A method and apparatus for automated vertebra localization and identification in a 3D computed tomography (CT) volumes is disclosed. Initial vertebra locations in a 3D CT volume of a patient are predicted for a plurality of vertebrae corresponding to a plurality of vertebra labels using a trained deep image-to-image network (DI2IN). The initial vertebra locations for the plurality of vertebrae predicted using the DI2IN are refined using a trained recurrent neural network, resulting in an updated set of vertebra locations for the plurality of vertebrae corresponding to the plurality of vertebrae labels. Final vertebra locations in the 3D CT volume for the plurality of vertebrae corresponding to the plurality of vertebra labels are determined by refining the updated set of vertebra locations using a trained shape-basis deep neural network.

Abstract: Methods and systems for artificial intelligence based medical image segmentation are disclosed. In a method for autonomous artificial intelligence based medical image segmentation, a medical image of a patient is received. A current segmentation context is automatically determined based on the medical image and at least one segmentation algorithm is automatically selected from a plurality of segmentation algorithms based on the current segmentation context. A target anatomical structure is segmented in the medical image using the selected at least one segmentation algorithm.