Image registration device, image registration method, and ultrasonic diagnosis apparatus having image registration device
There is provided an image registration device and an image registration method. The device includes: a feature extractor configured to extract, from a first image, a first feature group and to extract, from a second image, a second feature group; a feature converter configured to convert, using a converted neural network in which a correlation between features is learned, the extracted second feature group to correspond to the extracted first feature group, to obtain a converted group; and a register configured to register the first image and the second image based on the converted group and the extracted first feature group.
1. An image registration device, comprising:
at least one processor configured to execute:
an artificial neural network in which a correlation between features is learned;
a feature extractor configured to extract, using the artificial neural network, a first feature group from a first medical image and a second feature group from a second medical image;
a feature converter configured to convert the extracted second feature group to correspond to the extracted first feature group to obtain a converted group;
a register configured to register the first medical image and the second medical image based on the converted group and the extracted first feature group; and
a learning device configured to perform an unsupervised learning on the artificial neural network.
2. The image registration device according to claim 1 , wherein the artificial neural network comprises a first artificial neural network and a second artificial neural network which is different from the first artificial neural network, and the feature extractor is configured to extract, using the first artificial neural network, the first feature group from the first medical image and to extract, using the second artificial neural network, the second feature group from the second medical image.
3. The image registration device according to claim 1 , wherein the artificial neural network has a multilayer perceptron structure including at least one layer.
4. The image registration device according to claim 3 , wherein the at least one layer includes at least one unit, and among the at least one unit, only units included in layers adjacent to each other are connected according to a Boltzmann machine technique.
5. The image registration device according to claim 4 , wherein the learning device is configured to perform the unsupervised learning on the artificial neural network by updating a connection strength between the units based on input data.
6. The image registration device according to claim 5 , wherein the connection strength between the units is determined by the unsupervised learning.
7. The image registration device according to claim 5 , wherein the learning device is configured to perform the unsupervised learning on the artificial neural network by expanding or reducing the artificial neural network.
8. The image registration device according to claim 1 , wherein the first medical image and the second medical image are obtained by medical devices having different modalities.
9. An image registration method, comprising:
generating at least one artificial neural network in which a correlation between features is learned;
extracting a first feature group from a first medical image and a second feature group from a second medical image by using the at least one artificial neural network;
converting the extracted second feature group to correspond to the extracted first feature group to obtain a converted group;
registering the first medical image and the second medical image based on the converted group and the extracted first feature group; and
performing an unsupervised learning on the at least one artificial neural network.
10. The image registration method according to claim 9 , wherein the at least one artificial neural network has a multilayer perceptron structure including at least one layer.
11. The image registration method according to claim 10 , wherein the performing of the unsupervised learning on the at least one artificial neural network comprises performing an unsupervised learning on the at least one artificial neural network by updating a connection strength between units based on input data.
12. The image registration method according to claim 9 , wherein the performing of the unsupervised learning on the at least one artificial neural network comprises performing an unsupervised learning on the at least one artificial neural network by expanding or reducing the at least one artificial neural network.