Image processing system and medical information processing system
View Patent ↗In one embodiment, an image processing system includes a memory and processing circuitry. The memory is configured to store a predetermined program. The processing circuitry is configured, by executing the predetermined program, to perform processing on an input image by exploiting a neural network having an input layer, an output layer, and an intermediate layer provided between the input layer and the output layer, the input image being inputted to the input layer, and adjust an internal parameter based on data related to the input image, while performing the processing on the input image after training of the neural network, the internal parameter being at least one internal parameter of at least one node included in the intermediate layer, and the input parameter having been calculated by the training of the neural network.
1. An image processing system comprising:
a memory configured to store a predetermined program; and
processing circuitry configured, by executing the predetermined program, to
(a) in a training phase,
create a neural network by a machine learning,
wherein the neural network comprises an input layer, an output layer, and an intermediate layer provided between the input layer and the output layer, wherein the intermediate layer includes at least one node having at least one internal parameter, and the internal parameter is determined through the machine learning in the training phase;
(b) in an interference phase conducted after the training phase,
input an image into the neural network;
determine a maximum pixel value in the input image,
normalize the image by dividing a predetermined value by the maximum pixel value to generate a gain value,
set the gain value as a control signal,
apply denoise processing to the input image by adjusting the internal parameter of the neural network in accordance with the control signal.
2. The image processing system according to claim 1 ,
wherein the input image is an image that includes noise; and
the processing circuitry is configured to perform the processing on the input image in such a manner that the noise is reduced or removed from the input image.
3. The image processing system according to claim 1 ,
wherein the control signal is generated from SNR-related data that are related to magnitude of noise in the input image or signal-to-noise ratio of the input image.
4. The image processing system according to claim 1 ,
wherein the node has an activation function; and
the internal parameter is a parameter of the activation function.
5. The image processing system according to claim 4 ,
wherein the activation function is a function of removing such an input signal included in the input image that amplitude thereof is smaller than a threshold value; and
the internal parameter is the threshold value of the activation function.
6. The image processing system according to claim 4 ,
wherein the activation function is a function that outputs zero in a case of receiving an input value within a range including zero and outputs a value proportional to an input value in a case of receiving an input value outside the range.
7. The image processing system according to claim 1 ,
wherein, the image processing system is configured as at least one computer.