A system and method for using gaze information to extract visual attention information combined with computer derived local saliency information from medical images to (1) infer object and background cues from a region of interest indicated by the eye-tracking and (2) perform a medical image segmentation process. Moreover, an embodiment is configured to notify a medical professional of overlooked regions on medical images and/or train the medical professional to review regions that he/she often overlooks.

A multi-label generating system is configured to: store a first plurality of medical scans with corresponding global labels and a second plurality of medical scans with corresponding region labels, wherein the global labels each correspond to one of a set of abnormality classes and wherein each of the region labels correspond to one of the set of abnormality classes; generate a computer vision model by training on the first plurality of medical scans with the corresponding global labels and the second plurality of medical scans with the corresponding region labels; receive a new medical scan; generate global probability data based on the computer vision model, wherein the global probability data indicates a set of global probability values corresponding to the set of abnormality classes, and wherein each of the set of global probability values indicates a probability that a corresponding one of the set of abnormality classes is present in the new medical scan; and transmit the global probability data to a client device for display via a display device.

The present disclosure relates to a method for training a classifier. The method includes: acquiring an original image; determining a candidate target by segmenting the original image based on at least two segmentation models; determining a universal set of features by extracting features from the candidate target; determining a reference subset of features by selecting features from the universal set of features; and determining a classifier by performing classifier training based on the reference subset of features.

A method for implementing a dynamic three-dimensional lung map view for navigating a probe inside a patient's lungs includes loading a navigation plan into a navigation system, the navigation plan including a planned pathway shown in a 3D model generated from a plurality of CT images, inserting the probe into a patient's airways, registering a sensed location of the probe with the planned pathway, selecting a target in the navigation plan, presenting a view of the 3D model showing the planned pathway and indicating the sensed location of the probe, navigating the probe through the airways of the patient's lungs toward the target, iteratively adjusting the presented view of the 3D model showing the planned pathway based on the sensed location of the probe, and updating the presented view by removing at least a part of an object forming part of the 3D model.

The invention relates to a system for assessing a pulmonary image which allows for an improved assessment with respect to lung nodules detectability. The pulmonary image is smoothed for providing different pulmonary images (20, 21, 22) with different degrees of smoothing, wherein signal values and noise values, which are indicative of the lung vessel detectability and the noise in these images, are determined and used for determining an image quality being indicative of the usability of the pulmonary image to be assessed for detecting lung nodules. Since a pulmonary image shows lung vessels with many different vessel sizes and with many different image values, which cover the respective ranges of potential lung nodules generally very well, the image quality determination based on the different pulmonary images with different degrees of smoothing allows for a reliable assessment of the pulmonary image's usability for detecting lung nodules. The image quality is used to determine a radiation dose level to be applied for generating a next pulmonary image.

In one embodiment, a computer-implemented method is for providing lymph node information. The method includes receiving medical imaging data; receiving atlas data spatially relating lymph node stations to at least one non-lymphatic anatomical structure; determining a lymph node position in the medical imaging data; generating the lymph node information, the lymph node information being indicative of a lymph node station, to which the lymph node position is anatomically allocated, by applying an algorithm onto the medical imaging data, the atlas data and the lymph node position; and providing the lymph node information.

A medical image aided diagnosis method and system combining image recognition and report editing. The medical image aided diagnosis method comprises the following steps S1, establishing an image semantic expression knowledge graph of medical images, S2, obtaining a medical image of a patient, determining a region of interest on a two-dimensional image, and providing a candidate focus option of the patient according to the image semantic expression knowledge graph and the region of interest; and S3, determining a focus type according to the region of interest and the candidate focus option; performing division to obtain a lesion region according to the focus type, and generating a structured report related to a region-of-interest of the medical image of the patient, and adding the lesion region and corresponding expression content of image semantics into a corresponding focus image library. In the method, medical image recognition is performed by combining an image semantic expression knowledge graph and a variety of machine leaning, sample images can be deeply accumulated, the image semantic expression knowledge graph can be continuously improved, and aided diagnosis capabilities of medical images can be enhanced.

A Computer Aided Diagnosis, CADx, system (200) is described that comprises: at least one input (210, 212, 214) configured to provide at least one input medical image; and a CADx processing engine (220) configured to receive and process the at least one input medical image and produce at least one CADx score. A CADx score mapping circuit is operably coupled to the CADx processing engine (220) and configured to: map the at least one CADx score to a risk adjusted virtual score; and generate an output (235) of at least the risk adjusted virtual score associated with the processed at least one input medical image. The at least one CADx score and the risk adjusted virtual score correspond to an equivalent risk of condition or disease associated with a patient.

A local extension method for segmentation of anatomical treelike structures includes receiving an initial segmentation of 3D image data including an initial treelike structure. A target point in the 3D image data is defined, and a region of interest based on the target point is extracted to create a sub-image. Highly tubular voxels are detected in the sub-image, and a spillage-constrained region growing is performed using the highly tubular voxels as seed points. Connected components are extracted from the results of the region growing. The extracted components are pruned to discard components not likely to be connected to the initial treelike structure, keeping only candidate components likely to be a valid sub-tree of the initial treelike structure. The candidate components are connected to the initial treelike structure, thereby extending the initial segmentation in the region of interest.

A content-based medical image retrieval method and a retrieval system using the same include: obtaining m (2mn) number of unit images from a three-dimensional (3D) medical image including n (n2) number of unit images and extracting features per unit image from each of the m (2mn) number of unit images through a feature extraction unit, wherein the 3D medical image is voxel data including a plurality of slices and each of the plurality of slices is defined as a unit image; inputting features of each unit image extracted from the m (2mn) number of unit images to a recurrent neural network to generate an output value; and performing medical image retrieval using the output value through an input processing unit, wherein a plurality of 3D medical images to be compared with the output value include a 3D medical image having p (p2, pn) number of unit images.