A platform is provided for generating 3D models of a tumor segmented from a series of 2D medical images and for identifying from these 3D models, radiomic features that may be used for diagnostic, prognostic, and treatment response assessment of the tumor. The radiomic features may be shape-based features, intensity-based features, textural features, and filter-based features. The radiomic features are compared to remove sufficiently redundant features, thereby producing a reduced set of radiomic features, which is then compared to separate genomic data and/or outcome data to identify clinically and biologically significant radiomic features for diagnostic, prognostic, and treatment response assessment, other applications.

Embodiments of this application disclose an image processing method performed by a computer device, and a computer-readable storage medium. The method includes: obtaining a to-be-detected image, and performing down-sampling abnormality classification processing on the to-be-detected image, to obtain a predicted abnormality category label and a target feature image; performing preliminary abnormality positioning processing based on the predicted abnormality category label and the target feature image, to obtain an initial positioning image corresponding to the to-be-detected image; performing up-sampling abnormality positioning processing on the initial positioning image, to obtain a target positioning image corresponding to the to-be-detected image; and outputting the predicted abnormality category label and the target positioning image, the initial positioning image and the target positioning image being configured for reflecting attribute information of a target region associated with the predicted abnormality category label within the to-be-detected image.

The invention concerns automated methods for assessing tissue morphometry in digital images of tissue sections derived from small intestine biopsy samples from patients submitted for evaluation of celiac disease. The methods generally involve digital image analysis of tissue section images, and specifically involve post-processing each image to produce a binary mask capturing the tissue area footprint on the glass slide. Virtual stereology probes are placed on each image and assessed to estimate the ratio of the surface area to volume of the tissue specimen. The surface area to volume ratio is used to diagnose celiac disease and make inferences about the severity of celiac disease in those individuals with a positive diagnosis of celiac disease.

A method for evaluating a gastrointestinal tract may include characterizing one or more disease parameters using objective measures obtained from imaging data of a gastrointestinal tract. The one or more disease parameters reflect a measure of at least one of lesions, ulcers, bleeding, stenosis, and vasculature. The method may also include using the one or more characterized disease parameters to classify a disease state.

A method is provided for evaluating agents for the treatment of different intestinal motility disorders, using distinct methodological parts related to musculature and nerves of the GI tract which communicate with the brain. In particular, the present invention provides a method for the selection of an agent effective for the treatment of an intestinal motility disorder, wherein said method comprises: a) a step of spatiotemporal (ST) mapping carried out on a gastrointestinal segment to analyse the effect of said agent on gastrointestinal motility; and b) a step of ex vivo nerve bundle recording carried out on a gastrointestinal segment to analyse the effect of said agent on mesenteric afferent nerve firing. Bacterial strains selected by the methods of the invention and the use of said bacterial strains in the treatment of intestinal motility disorders are also provided.

A landmark estimating method estimates a position of a landmark that is a hole existing in an object and is a site through which an insertion portion penetrates, in an endoscope image obtained by picking up an image of the object by an endoscope with the insertion portion bent. The landmark estimating method includes estimating an axis of the insertion portion, estimating a boundary of the insertion portion and the object, and estimating the position of the landmark based on the axis and the boundary that are estimated.

A method for diagnosing a gastric lesion from endoscopic images is provided. The method comprises: acquiring a plurality of gastric lesion images; generating a dataset by linking the plurality of gastric lesion images with patient information; preprocessing the dataset in a way that is applicable to a deep learning algorithm; and building an artificial neural network by training the artificial neural network by using the preprocessed dataset as input and gastric lesion classification results as output.

RGB image signals are inputted. A B/G ratio is calculated from the B and G image signals. A G/R ratio is calculated from the G and R image signals. First, second, and third areas are located in a feature space formed by the B/G and G/R ratios. An equal angular magnification process is performed on an angle in a region R1x including a first reference line passing through the second area. An angle expansion process or an angle compression process is performed on an angle in a region R1y located outside the region R1x. An equal radial-coordinate magnification process is performed on a radial coordinate in a region R2x, which includes a second reference line passing through the second area and intersecting the first reference line. A radial-coordinate expansion process or a radial-coordinate compression process is performed on a radial coordinate in a region R2y located outside the region R2x.

Obtaining a surface image captured by an endoscope inserted in a tubular organ associated with a surrounding blood vessel and representing an inner surface of a wall of the organ, generating, from a three-dimensional image representing a three-dimensional area including the organ, an adjacent blood vessel image depicting a portion of the blood vessel adjacent to the wall from a viewpoint in the three-dimensional image corresponding the viewpoint of the surface image, generating, from a three-dimensional image representing a three-dimensional area including a surrounding area of the organ, a surrounding blood vessel image depicting the blood vessel from a viewpoint in the three-dimensional image corresponding the viewpoint of the surface image, and causing the surface image, adjacent blood vessel image, and surrounding blood vessel image to be displayed in this order on a display unit.

Methods, devices, systems and apparatus for medical imaging, e.g., Magnetic Resonance (MR) imaging or Computed Tomography (CT) imaging, using neural networks are provided. In one aspect, an imaging method includes: determining a first neural network and a second neural network corresponding to a target imaging task, the first neural network including a first neural network parameter and a first neural network model, the second neural network including a second neural network parameter and a second neural network model, obtaining a reconstructed image by performing reconstruction for down-sampling data of a tissue under test using the first neural network, the target imaging task corresponding to the tissue under test, and obtaining an image output by a second neural network as a target image of the tissue under test by performing an image processing operation corresponding to the target imaging task for the reconstructed image using the second neural network.