Method for medical image conversion in frequency domain using artificial intelligence, and device thereof
AssigneePolestar Healthcare Co., Ltd.
InventorsYeo Dong YOON, Eun Chong LEE
Life of this patent
Estimating the term…
Abstract
One mode of the present invention relates to a medical image conversion method, and, more specifically, to a medical image conversion method using a generative adversarial network (GAN), wherein an embodiment of the present invention has the effect of providing a medical image conversion method that enables a medical team to perform medical examination, diagnosis, and treatment using accurate information through conversion between medical images on the basis of a machine learning model, and enables a patient to receive suitable medical services.
Claims10
A method of converting a medical image using a generative adversarial network (GA), the method comprising:
- a first operation of receiving a first image selected from a paired data set including the first image and a second image;
- a second operation of constructing a third image matching the first image based on the first image;
- and
- a third operation of converting the second image and the third image to frequency domain images, respectively, comparing the converted frequency domain images, and reflecting a result of the comparison to a learning model which performs medical image conversion.
A device for converting a medical image using a generative adversarial network (GAN), the device comprising:
- a first learning model configured to receive a first image selected from a paired data set including the first image and a second image and constructing a third image based on the first image;
- and
- a second learning model receiving a third frequency domain image to which the third image is converted in a frequency domain and constructing a quality-enhanced third frequency domain image based on the third frequency domain image.
A method of converting a medical image using a generative adversarial network (GAN), the method comprising:
- an image construction operation of receiving a first image and constructing a third image;
- a first frequency conversion operation of converting the third image to a third frequency domain image in a frequency domain;
- and
- an image quality enhancement operation of receiving the third frequency domain image and constructing a quality-enhanced third frequency domain image.
Dependencies are inferred from the claim text.