Described herein is a novel method and system for segmentation of the spine using 3D volumetric data. In embodiments, a method includes an extracting step, localization step, and segmentation step. The extracting step comprises detecting the spine centerline and the spine canal centerline. The localization step comprises localizing the vertebra and intervertebral disc centers. Background and foreground constraints are created for each vertebra digit. Segmentation is performed for each vertebra digit and based on the hard constraints.

According to one embodiment, a support information-generation apparatus includes processing circuitry. The processing circuitry extracts a vertebral body region from a medical image and calculates a first volume of a vertebral body targeted for treatment by percutaneous vertebroplasty. The processing circuitry estimates a second volume of the targeted vertebral body related to an estimated shape. The processing circuitry calculates, based on a difference between the first volume and the second volume, a volume of an object to be inserted so as to deform the targeted vertebral body into the estimated shape.

A method for identifying a body region in a medical image includes obtaining a medical image including a number of consecutive bio-section images, inputting the medical image into a preset machine learning model to obtain a numerical value for each of the bio-section images corresponding to the body region to which the bio-section image belongs, determining whether the numerical values of the medical image are abnormal, adjusting the numerical values when the numerical values are abnormal, determining the body region corresponding to the numerical values or the adjusted numerical values, and labeling the body region in the medical image and outputting the labeled medical image. The bio-section images are cross-sectional images of a living body.

A method for analysis of spine anatomy and stenosis is disclosed herein. The method includes preprocessing (3203) of images and then running segmentation models for each area of interest such as the foramen, disc, canal, vertebra (3206). In image post processing (3207), the system runs various heuristics to ensure accuracy (3208), computes areas (3209), and then runs a comparison model (3210). The report template produces HTML that is then converted to a PDF file of the final report (3214).

A machine-based learning method estimates a probability of bone fractures in a 3D image, more specifically vertebral fractures. The method and system utilizing such method utilize a data-driven computational model to learn 3D image features for classifying vertebra fractures. A three-dimensional medical image analysis system for predicting a presence of a vertebral fracture in a subject includes a 3D image processor for receiving and processing 3D image data of a 3D image of the subject, producing two or more sets of 3D voxels. Each of the sets of 3D voxels corresponds to an entirety of the 3D image and each of the sets of 3D voxels consists of equal 3D voxels of different dimensions. The system also includes a voxel classifier for assigning the 3D voxels one or more class probabilities each of the 3D voxels contains a fracture using a computational model, and a fracture probability estimator for estimating a probability of the presence of a vertebral fracture in the subject.

A low radiation, intra-operative method using two-dimensional imaging to register the positions of surgical implants relative to their pre-operative planned positions. Intraoperatively, a pair of two-dimensional fluoroscope images in different planes or a single three-dimensional image is acquired and compared to a set of three-dimensional pre-operative images, to allow registration of the implant region anatomy. A second set of intraoperative fluoroscope images is acquired of the surgical area with implants in place. The second set of images is compared with the first set of intraoperative images to ascertain alignment of the implants. Registration between first and second intraoperative image sets is accomplished using the implants themselves as fiducial markers, and the process repeated until an acceptable configuration of the implants is obtained. The method is particularly advantageous for spinal surgery.

Systems and methods are configured to extract images from provided source data files and to preprocess such images for content-based image analysis. An image analysis system applies one or more machine-learning based models for identifying specific features within analyzed images, and for determining one or more measurements based at least in part on the identified features. Such measurements may be embodied as absolute measurements for determining an absolute distance between features, or relative measurements for determining a relative relationship between features. The determined measurements are input into one or more machine-learning based models for determining a classification for the image.

An ultrasound diagnostic apparatus 1 includes an ultrasound probe 15, an image acquisition unit 8 that transmits an ultrasound beam from the ultrasound probe 15 to a subject to acquire an ultrasound image, a site recognition unit 9 that performs image analysis on the ultrasound image acquired by the image acquisition unit 8 to recognize an imaged site of the subject, a memory 11 that stores at least one peripheral site effective to detect a target site, and an operation guide unit 10 that, during detection of the target site, guides a user to operate the ultrasound probe 15 so as to detect the at least one peripheral site stored in the memory 11 and guides the user to operate the ultrasound probe 15 so as to detect the target site on the basis of a recognition result obtained by the site recognition unit 9.

Disclosed is a technique for quickly removing and correcting a cone-beam artifact generated in a computed tomography (CT) image in consideration of bone and soft tissue regions when using a large-area X-ray detector in order to reduce a CT imaging time for large volumes in a cone-beam CT system. An apparatus includes an input unit configured to receive a start image including a cone-beam artifact, a computation unit configured to separate a high-density material image and a low-density material image from the start image received by the input unit, generate a reproduced image in which the cone-beam artifact is reproduced using the low-density material image, execute a correction process for subtracting the reproduced image from the start image to generate a corrected image, and iterate the correction process using the corrected image as a start image; and an output unit configured to output a final corrected image generated.

A computer-implemented method for fully-autonomous level identification of anatomical structures within a three-dimensional medical imagery, includes: receiving a set of medical scan images of the anatomical structures; processing the set to perform an autonomous semantic segmentation of anatomical components and to store segmentation results; processing segmentation results by removing the false positives, and smoothing 3D surfaces of the generated anatomical components; determining morphological and spatial relationships of the anatomical components; grouping the anatomical components to form separate levels based on the morphological and spatial relationships of the anatomical components; processing the set using a convolutional neural network to autonomously assign an initial level type; assigning the determined level type to each group of anatomical components by combining the determined morphological and spatial relationships with the determined initial level type; assigning an ordinal identifier to each group of anatomical components; and storing information about the assigned levels with their ordinal identifier.