Abstract: A method and system for classification of endoscopic images is disclosed. An initial trained deep network classifier is used to classify endoscopic images and determine confidence scores for the endoscopic images. The confidence score for each endoscopic image classified by the initial trained deep network classifier is compared to a learned confidence threshold. For endoscopic images with confidence scores higher than the learned threshold value, the classification result from the initial trained deep network classifier is output. Endoscopic images with confidence scores lower than the learned confidence threshold are classified using a first specialized network classifier built on a feature space of the initial trained deep network classifier.

Abstract: A method and system for determining fractional flow reserve (FFR) for a coronary artery stenosis of a patient is disclosed. In one embodiment, medical image data of the patient including the stenosis is received, a set of features for the stenosis is extracted from the medical image data of the patient, and an FFR value for the stenosis is determined based on the extracted set of features using a trained machine-learning based mapping. In another embodiment, a medical image of the patient including the stenosis of interest is received, image patches corresponding to the stenosis of interest and a coronary tree of the patient are detected, an FFR value for the stenosis of interest is determined using a trained deep neural network regressor applied directly to the detected image patches.

Abstract: A method and a system for automatically aligning a positionable X-ray source of an X-ray system in alignment with a mobile X-ray detector is disclosed where the X-ray system detects the position of the mobile X-ray detector using a 3D camera and then driving the positionable X-ray source to a position in alignment with the mobile X-ray detector.

Abstract: A method for controlling a scanner comprises: sensing an outer surface of a body of a subject to collect body surface data, using machine learning to predict a surface of an internal organ of the subject based on the body surface data, and controlling the scanner based on the predicted surface of the internal organ.

Abstract: A method and system for determining fractional flow reserve (FFR) for a coronary artery stenosis of a patient is disclosed. In one embodiment, medical image data of the patient including the stenosis is received, a set of features for the stenosis is extracted from the medical image data of the patient, and an FFR value for the stenosis is determined based on the extracted set of features using a trained machine-learning based mapping. In another embodiment, a medical image of the patient including the stenosis of interest is received, image patches corresponding to the stenosis of interest and a coronary tree of the patient are detected, an FFR value for the stenosis of interest is determined using a trained deep neural network regressor applied directly to the detected image patches.

Abstract: A method and system for anatomical landmark detection in medical images using deep neural networks is disclosed. For each of a plurality of image patches centered at a respective one of a plurality of voxels in the medical image, a subset of voxels within the image patch is input to a trained deep neural network based on a predetermined sampling pattern. A location of a target landmark in the medical image is detected using the trained deep neural network based on the subset of voxels input to the trained deep neural network from each of the plurality of image patches.

Abstract: A method and apparatus for X-ray tube scanner automation using a 3D camera is disclosed. An RGBD image of a patient on a patient table is received from a 3D camera mounted on an X-ray tube. A transformation between a coordinate system of the 3D camera and a coordinate system of the patient table is calculated. A patient model is estimated from the RGBD image of the patient. The X-ray tube is automatically controlled to acquire an X-ray image of a region of interest of the patient based on the patient model.

Abstract: For human body representation, bone length or other size characteristic that varies within the population is incorporated into the geometric model of the skeleton. The geometric model may be normalized for shape or tissue modeling, allowing modeling of the shape without dedicating aspects of the data-driven shape model to the length or other size characteristic. Given the same number or extent of components of the data-driven shape model, greater or finer details of the shape may be modeled since components are not committed to the size characteristic.

Abstract: Robust calcification tracking is provided in fluoroscopic imagery. A patient with an inserted catheter is scanned over time. A processor detects the catheter in the patient from the scanned image data. The processor tracks the movement of the catheter. The processor also detects a structure represented in the data. The structure is detected as a function of movement with a catheter. The processor tracks the movement of the structure using sampling based on a previous location of the structure in the patient. The processor may output an image of the structure.

Abstract: Methods and systems for image registration using an intelligent artificial agent are disclosed. In an intelligent artificial agent based registration method, a current state observation of an artificial agent is determined based on the medical images to be registered and current transformation parameters. Action-values are calculated for a plurality of actions available to the artificial agent based on the current state observation using a machine learning based model, such as a trained deep neural network (DNN). The actions correspond to predetermined adjustments of the transformation parameters. An action having a highest action-value is selected from the plurality of actions and the transformation parameters are adjusted by the predetermined adjustment corresponding to the selected action. The determining, calculating, and selecting steps are repeated for a plurality of iterations, and the medical images are registered using final transformation parameters resulting from the plurality of iterations.

Abstract: For human body representation, bone length or other size characteristic that varies within the population is incorporated into the geometric model of the skeleton. The geometric model may be normalized for shape or tissue modeling, allowing modeling of the shape without dedicating aspects of the data-driven shape model to the length or other size characteristic. Given the same number or extent of components of the data-driven shape model, greater or finer details of the shape may be modeled since components are not committed to the size characteristic.

Abstract: A method and system for determining fractional flow reserve (FFR) for a coronary artery stenosis of a patient is disclosed. In one embodiment, medical image data of the patient including the stenosis is received, a set of features for the stenosis is extracted from the medical image data of the patient, and an FFR value for the stenosis is determined based on the extracted set of features using a trained machine-learning based mapping. In another embodiment, a medical image of the patient including the stenosis of interest is received, image patches corresponding to the stenosis of interest and a coronary tree of the patient are detected, an FFR value for the stenosis of interest is determined using a trained deep neural network regressor applied directly to the detected image patches.

Abstract: Object detection uses a deep or multiple layer network to learn features for detecting the object in the image. Multiple features from different layers are aggregated to train a classifier for the object. In addition or as an alternative to feature aggregation from different layers, an initial layer may have separate learnt nodes for different regions of the image to reduce the number of free parameters. The object detection is learned or a learned object detector is applied.

Abstract: A method and system for anatomical object detection using marginal space deep neural networks is disclosed. The pose parameter space for an anatomical object is divided into a series of marginal search spaces with increasing dimensionality. A respective sparse deep neural network is trained for each of the marginal search spaces, resulting in a series of trained sparse deep neural networks. Each of the trained sparse deep neural networks is trained by injecting sparsity into a deep neural network by removing filter weights of the deep neural network.

Abstract: A method and system for determining fractional flow reserve (FFR) for a coronary artery stenosis of a patient is disclosed. In one embodiment, medical image data of the patient including the stenosis is received, a set of features for the stenosis is extracted from the medical image data of the patient, and an FFR value for the stenosis is determined based on the extracted set of features using a trained machine-learning based mapping. In another embodiment, a medical image of the patient including the stenosis of interest is received, image patches corresponding to the stenosis of interest and a coronary tree of the patient are detected, an FFR value for the stenosis of interest is determined using a trained deep neural network regressor applied directly to the detected image patches.

Abstract: A method and system for anatomical object detection using marginal space deep neural networks is disclosed. The pose parameter space for an anatomical object is divided into a series of marginal search spaces with increasing dimensionality. A respective deep neural network is trained for each of the marginal search spaces, resulting in a series of trained deep neural networks. Each of the trained deep neural networks can evaluate hypotheses in a current parameter space using discriminative classification or a regression function. An anatomical object is detected in a medical image by sequentially applying the series of trained deep neural networks to the medical image.

Abstract: A computer-implemented method for providing a see-through visualization of a patient includes receiving an image dataset representative of anatomical features of the patient acquired using a medical image scanner and acquiring a body surface model of the patient using an RGB-D sensor. The body surface model is aligned with the image dataset in a canonical/common coordinate system to yield an aligned body surface model. A relative pose of a mobile device is determined with respect to the RGB-D sensor and a pose dependent visualization of the patient is created by rendering the image dataset at a viewpoint corresponding to the relative pose of the mobile device. Then, the pose dependent visualization of the patient may be presented on the mobile device.

Abstract: A method and system for approximating a deep neural network for anatomical object detection is discloses. A deep neural network is trained to detect an anatomical object in medical images. An approximation of the trained deep neural network is calculated that reduces the computational complexity of the trained deep neural network. The anatomical object is detected in an input medical image of a patient using the approximation of the trained deep neural network.

Abstract: A computer-implemented method for automatically calibrating an RGB-D sensor and an imaging device using a transformation matrix includes using a medical image scanner to acquire a first dataset representative of an apparatus attached to a downward facing surface of a patient table, wherein corners of the apparatus are located at a plurality of corner locations. The plurality of corner locations are identified based on the first dataset and the RGB-D sensor is used to acquire a second dataset representative of a plurality of calibration markers displayed on an upward facing surface of the patient table at the corner locations. A plurality of calibration marker locations are identified based on the second dataset and the transformation matrix is generated by aligning the first dataset and the second dataset using the plurality of corner locations and the plurality of calibration marker locations.

Abstract: A method and apparatus for X-ray tube scanner automation using a 3D camera is disclosed. An RGBD image of a patient on a patient table is received from a 3D camera mounted on an X-ray tube. A transformation between a coordinate system of the 3D camera and a coordinate system of the patient table is calculated. A patient model is estimated from the RGBD image of the patient. The X-ray tube is automatically controlled to acquire an X-ray image of a region of interest of the patient based on the patient model.