Generation of Modified Medical Images and Detection of Abnormal Structures

Simple SummaryContent extracted from patent full text and abstract with AI.

This patent describes a method for generating modified medical images by removing abnormal structures through a trained inpainting function, and then using patches of these abnormalities to simulate them in other images. It also covers methods for improving detection of abnormal structures in medical images using these modified images, alongside systems and software for implementing the methods.

Use CasesContent extracted from patent full text and abstract with AI.

• Training artificial intelligence systems to better detect tumors or lesions in medical images
• Simulating rare medical conditions by embedding abnormality patches into normal medical images for research or education
• Improving diagnostic support software for radiologists and clinicians
• Developing robust algorithms for automated computer-aided detection in various imaging modalities (e.g., MRI, CT, X-ray)
• Testing and validating new image analysis software under controlled conditions

BenefitsContent extracted from patent full text and abstract with AI.

• Enables creation of diverse and controlled medical image datasets for training machine learning models
• Improves detection accuracy of abnormal structures, potentially leading to earlier disease diagnosis
• Facilitates simulation of rare or subtle abnormalities, enhancing research and educational resources
• Reduces the need for large datasets of real abnormal cases by generating synthetic examples
• Supports development of more robust and generalizable medical imaging analysis tools

Technical Classifications (CPCs)

Inventors & Applicants

Patent Abstract

A method is for generating modified medical images. An embodiment of the method includes receiving a first medical image displaying an abnormal structure within a patient, and applying a trained inpainting function to the first medical image to generate a modified first medical image, the trained inpainting function being trained to inpaint abnormal structures within a medical image. The method includes determining an abnormality patch based on the first medical image and the modified first medical image; receiving a second medical image of the same type as the first medical image; and including the abnormality patch into the second medical image to generate a modified second medical image. A method is for detecting abnormal structures using a trained detection function trained based on modified second medical images. Systems, computer programs and computer-readable media related to those methods are also disclosed.