Medical image generation, localizaton, registration system
AssigneeCaide Systems, Inc.
InventorsJung Hwan Cho, Manohar Karki, Kye Wook Lee
Life of this patent
Estimating the term…
Abstract
A method for generating a synthesized medical image by receiving a normal image includes generating first data based on a random selection, generating second data, and, based at least in part on the first and second data, modifying the normal image to form the synthesized medical image. Modifying the normal image comprises combining the first data and the second data. The first data characterizes an image that represents a lesion and the second data characterizes a transformation of that image as well as a location of the lesion.
Claims15
A method for generating a synthesized medical image, said method comprising receiving a normal image, generating first data based on a random selection, generating second data, and modifying said normal image to form said synthesized medical image, wherein modifying said normal image comprises combining said first data and said second data, wherein said first data characterizes an image that represents a lesion, wherein said second data characterizes a transformation of said image, and wherein said second data characterizes at least a location of said lesion.
A method for delineating a lesion by anatomical normalization as seen on a medical image, said method comprising:
- receiving an image obtained from a patient, said image showing an abnormal lesion,
- receiving a demographic-specific template, said demographic-specific template having been created by averaging normalized images obtained from a population of healthy subjects that are within a designated demographic group,
- receiving a mean and a standard deviation image volume based on an anatomical normalization of said images, and
- generating a delineation of said lesion using a statistical voxel comparison between said image showing said abnormal lesion and said anatomical normalization.
Dependencies are inferred from the claim text.