# US 2022/0101527 A1 · Visualization method for evaluating brain addiction traits, apparatus, and medium

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2022/0101527 A1
- Document ID: US20220101527A1
- Status: Published application
- Kind: A1
- Application number: 17/549,258
- Filed: 2021-12-13
- Published: 2022-03-31
- Earliest claimed: 2020-03-11
- Assignees: Shenzhen Institutes of Advanced Technology
- Applicants: Shenzhen Institutes of Advanced Technology
- Inventors: Shuqiang WANG; Wen YU; Chenchen XIAO; Shengye HU
- Web page: https://patentgrep.com/patent/US20220101527A1

## CPC

- G06N 3/088: Non-supervised learning, e.g. competitive learning
- G06N 3/045: Combinations of networks
- G06N 3/0464: Convolutional networks [CNN, ConvNet]
- G06N 3/047: Probabilistic or stochastic networks
- G06N 3/0475: Generative networks
- G06N 3/048: Activation functions
- G06N 3/0495: Quantised networks; Sparse networks; Compressed networks
- G06N 3/0895: Weakly supervised learning, e.g. semi-supervised or self-supervised learning
- G06N 3/09: Supervised learning
- G06N 3/094: Adversarial learning
- G06T 7/0012: {Biomedical image inspection}
- G06V 10/267: {by performing operations on regions, e.g. growing, shrinking or watersheds}
- G06V 10/32: Normalisation of the pattern dimensions
- G06V 10/34: Smoothing or thinning of the pattern; Morphological operations; Skeletonisation
- G06V 10/764: using classification, e.g. of video objects
- G06V 10/774: Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting
- G06V 10/82: using neural networks
- G06T 2207/10088: Magnetic resonance imaging [MRI]
- G06T 2207/20081: Training; Learning
- G06T 2207/20084: Artificial neural networks [ANN]
- G06T 2207/30016: Brain
- G06V 2201/03: Recognition of patterns in medical or anatomical images
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING

## Abstract

A visualization method for evaluating brain addiction traits, an apparatus, and a computer-readable storage medium are provided. The method includes the following. A visualization processing request is received from a client, where the visual processing request contains an image to-be-processed. The image to-be-processed is masked to obtain a perturbation image masked. The perturbation image is classified with a visualization processing model to obtain a classification result, and the classification result is calculated to obtain an evaluation value of the perturbation image, where the evaluation value of the perturbation image is less than an evaluation value of the image to-be-processed without masking. The visualization evaluation result is determined according to the evaluation value of the perturbation image. The visualization evaluation result is sent to the client.

## Claims

### Claim 1

A visualization method for evaluating brain addiction traits, comprising:
receiving a visualization processing request from a client, wherein the visual processing request contains an image to-be-processed and is used for requesting a visualization evaluation result of the image to-be-processed;
masking the image to-be-processed to obtain a perturbation image masked;
classifying the perturbation image with a visualization processing model to obtain a classification result, and calculating the classification result to obtain an evaluation value of the perturbation image, wherein the evaluation value of the perturbation image is less than an evaluation value of the image to-be-processed without masking;
determining the visualization evaluation result according to the evaluation value of the perturbation image; and
sending the visualization evaluation result to the client.

### Claim 2 (depends on 1)

The method of claim 1, wherein the masking comprises blurring, and masking the image to-be-processed to obtain the perturbation image masked comprises:
determining a masking region set, wherein the masking region set comprises at least one masking region; and
in the image to-be-processed, blurring respective masking regions in the masking region set to obtain the perturbation image, wherein the image to-be-processed comprises the respective masking regions in the masking region set.

### Claim 3 (depends on 1)

The method of claim 1, further comprising: before classifying the perturbation image with the visualization processing model,
obtaining a noise vector and a vector annotation, and processing the noise vector and the vector annotation with a deconvolution network to obtain a target image-annotation pair, the target image-annotation pair comprising a target generated image and a target generated image annotation;
obtaining a first sample image and a second sample image-annotation pair, the second sample image-annotation pair comprising a second sample image and a sample image annotation;
training the visualization processing model according to the target image-annotation pair, the first sample image, and the second sample image-annotation pair to obtain a model loss function; and
constructing the visualization processing model according to the model loss function.

### Claim 4 (depends on 3)

The method of claim 3, wherein the model loss function comprises a generation loss function, wherein the generation loss function is a loss function of a generator, and training the visualization processing model according to the target image-annotation pair, the first sample image, and the second sample image-annotation pair to obtain the model loss function comprises:
discriminating the target image-annotation pair to generate a first discrimination result;
determining a reconstruction loss according to the target generated image, the first sample image, and the second sample image; and
determining the generation loss function of the generator according to the first discrimination result and the reconstruction loss.

### Claim 5 (depends on 3)

The method of claim 3, wherein the model loss function comprises a classification loss function, wherein the classification loss function is a loss function of a classifier, and training the visualization processing model according to the target image-annotation pair, the first sample image, and the second sample image-annotation pair to obtain the model loss function comprises:
classifying the first sample image to obtain a predicted annotation of the first sample image, and discriminating the first sample image and the predicted annotation to obtain a second discrimination result;
performing supervised training on the target image-annotation pair and the second sample image-annotation pair to obtain a first cross entropy of the target image-annotation pair and a second cross entropy of the second sample image-annotation pair; and
determining the classification loss function of the classifier according to the second discrimination result, the first cross entropy, and the second cross entropy.

### Claim 6 (depends on 3)

The method of claim 3, wherein the model loss function comprises a discrimination loss function, where the discrimination loss function is a loss function of a discriminator, and training the visualization processing model according to the target image-annotation pair, the first sample image, and the second sample image-annotation pair to obtain the model loss function comprises:
discriminating the target image-annotation pair to generate a third discrimination result;
discriminating the first sample image and the predicted annotation to generate a fourth discrimination result;
discriminating the second sample image-annotation pair to generate a fifth discrimination result; and
determining the discrimination loss function of the discriminator according to the third discrimination result, the fourth discrimination result, and the fifth discrimination result.

### Claim 7 (depends on 3)

The method of claim 3, wherein the model loss function comprises a generation loss function, a classification loss function, and a discrimination loss function, and constructing the visualization processing model according to the model loss function comprises:
updating a parameter of a generator according to the generation loss function using a back propagation algorithm;
updating a parameter of a classifier according to the classification loss function using a back propagation algorithm;
updating a parameter of a discriminator according to the discrimination loss function using a back propagation algorithm; and
constructing the visualization processing model according to the parameter of the generator, the parameter of the classifier, and the parameter of the discriminator.

### Claim 8

A visualization processing apparatus for evaluating brain addiction traits, comprising a processor, a memory, and a communication interface connected with each other, wherein the memory is configured to store a computer program comprising program instructions, and the processor is configured to invoke the program instructions to:
receive a visualization processing request from a client, wherein the visual processing request contains an image to-be-processed and is used for requesting a visualization evaluation result of the image to-be-processed;
mask the image to-be-processed to obtain a perturbation image masked;
classify the perturbation image with a visualization processing model to obtain a classification result, and calculating the classification result to obtain an evaluation value of the perturbation image, wherein the evaluation value of the perturbation image is less than an evaluation value of the image to-be-processed without masking;
determine the visualization evaluation result according to the evaluation value of the perturbation image; and
send the visualization evaluation result to the client.

### Claim 9 (depends on 8)

The visualization processing apparatus of claim 8, wherein the masking comprises blurring, and the processor configured to invoke the program instructions to mask the image to-be-processed to obtain a perturbation image masked is configured to invoke the program instructions to:
determine a masking region set, wherein the masking region set comprises at least one masking region; and
in the image to-be-processed, blur respective masking regions in the masking region set to obtain the perturbation image, wherein the image to-be-processed comprises the respective masking regions in the masking region set.

### Claim 10 (depends on 8)

The visualization processing apparatus of claim 8, the processor is further configured to invoke the program instructions to:
obtain a noise vector and a vector annotation, and process the noise vector and the vector annotation with a deconvolution network to obtain a target image-annotation pair, the target image-annotation pair comprising a target generated image and a target generated image annotation;
obtain a first sample image and a second sample image-annotation pair, the second sample image-annotation pair comprising a second sample image and a sample image annotation;
train the visualization processing model according to the target image-annotation pair, the first sample image, and the second sample image-annotation pair to obtain a model loss function; and
construct the visualization processing model according to the model loss function.

### Claim 11 (depends on 10)

The visualization processing apparatus of claim 10, wherein the model loss function comprises a generation loss function, wherein the generation loss function is a loss function of a generator, and the processor configured to invoke the program instructions to train the visualization processing model according to the target image-annotation pair, the first sample image, and the second sample image-annotation pair to obtain the model loss function is configured to invoke the program instructions to:
discriminate the target image-annotation pair to generate a first discrimination result;
determine a reconstruction loss according to the target generated image, the first sample image, and the second sample image; and
determine the generation loss function of the generator according to the first discrimination result and the reconstruction loss.

### Claim 12 (depends on 10)

The visualization processing apparatus of claim 10, wherein the model loss function comprises a classification loss function, wherein the classification loss function is a loss function of a classifier, and the processor configured to invoke the program instructions to train the visualization processing model according to the target image-annotation pair, the first sample image, and the second sample image-annotation pair to obtain the model loss function is configured to invoke the program instructions to:
classify the first sample image to obtain a predicted annotation of the first sample image, and discriminating the first sample image and the predicted annotation to obtain a second discrimination result;
perform supervised training on the target image-annotation pair and the second sample image-annotation pair to obtain a first cross entropy of the target image-annotation pair and a second cross entropy of the second sample image-annotation pair; and
determine the classification loss function of the classifier according to the second discrimination result, the first cross entropy, and the second cross entropy.

### Claim 13 (depends on 10)

The visualization processing apparatus of claim 10, wherein the model loss function comprises a discrimination loss function, where the discrimination loss function is a loss function of a discriminator, and the processor configured to invoke the program instructions to train the visualization processing model according to the target image-annotation pair, the first sample image, and the second sample image-annotation pair to obtain the model loss function is configured to invoke the program instructions to:
discriminate the target image-annotation pair to generate a third discrimination result;
discriminate the first sample image and the predicted annotation to generate a fourth discrimination result;
discriminate the second sample image-annotation pair to generate a fifth discrimination result; and
determine the discrimination loss function of the discriminator according to the third discrimination result, the fourth discrimination result, and the fifth discrimination result.

### Claim 14 (depends on 10)

The visualization processing apparatus of claim 10, wherein the model loss function comprises a generation loss function, a classification loss function, and a discrimination loss function, and the processor configured to invoke the program instructions to construct the visualization processing model according to the model loss function is configured to invoke the program instructions to:
update a parameter of a generator according to the generation loss function using a back propagation algorithm;
update a parameter of a classifier according to the classification loss function using a back propagation algorithm;
update a parameter of a discriminator according to the discrimination loss function using a back propagation algorithm; and
construct the visualization processing model according to the parameter of the generator, the parameter of the classifier, and the parameter of the discriminator.

### Claim 15

A computer-readable storage medium storing one or more instructions, wherein the one or more instructions are adapted to be loaded by a processor to:
receive a visualization processing request from a client, wherein the visual processing request contains an image to-be-processed and is used for requesting a visualization evaluation result of the image to-be-processed;
mask the image to-be-processed to obtain a perturbation image masked;
classify the perturbation image with a visualization processing model to obtain a classification result, and calculating the classification result to obtain an evaluation value of the perturbation image, wherein the evaluation value of the perturbation image is less than an evaluation value of the image to-be-processed without masking;
determine the visualization evaluation result according to the evaluation value of the perturbation image; and
send the visualization evaluation result to the client.

### Claim 16 (depends on 15)

The computer-readable storage medium of claim 15, wherein the masking comprises blurring, and the one or more instructions adapted to be loaded by the processor to mask the image to-be-processed to obtain a perturbation image masked are adapted to be loaded by the processor to:
determine a masking region set, wherein the masking region set comprises at least one masking region; and
in the image to-be-processed, blur respective masking regions in the masking region set to obtain the perturbation image, wherein the image to-be-processed comprises the respective masking regions in the masking region set.

### Claim 17 (depends on 15)

The computer-readable storage medium of claim 15, wherein the one or more instructions are further adapted to be loaded by the processor to:
obtain a noise vector and a vector annotation, and process the noise vector and the vector annotation with a deconvolution network to obtain a target image-annotation pair, the target image-annotation pair comprising a target generated image and a target generated image annotation;
obtain a first sample image and a second sample image-annotation pair, the second sample image-annotation pair comprising a second sample image and a sample image annotation;
train the visualization processing model according to the target image-annotation pair, the first sample image, and the second sample image-annotation pair to obtain a model loss function; and
construct the visualization processing model according to the model loss function.

### Claim 18 (depends on 17)

The computer-readable storage medium of claim 17, wherein the model loss function comprises a generation loss function, wherein the generation loss function is a loss function of a generator, and the one or more instructions adapted to be loaded by the processor to train the visualization processing model according to the target image-annotation pair, the first sample image, and the second sample image-annotation pair to obtain the model loss function are adapted to be loaded by the processor to:
discriminate the target image-annotation pair to generate a first discrimination result;
determine a reconstruction loss according to the target generated image, the first sample image, and the second sample image; and
determine the generation loss function of the generator according to the first discrimination result and the reconstruction loss.

### Claim 19 (depends on 17)

The computer-readable storage medium of claim 17, wherein the model loss function comprises a classification loss function, wherein the classification loss function is a loss function of a classifier, and the one or more instructions adapted to be loaded by the processor to train the visualization processing model according to the target image-annotation pair, the first sample image, and the second sample image-annotation pair to obtain the model loss function are adapted to be loaded by the processor to:
classify the first sample image to obtain a predicted annotation of the first sample image, and discriminating the first sample image and the predicted annotation to obtain a second discrimination result;
perform supervised training on the target image-annotation pair and the second sample image-annotation pair to obtain a first cross entropy of the target image-annotation pair and a second cross entropy of the second sample image-annotation pair; and
determine the classification loss function of the classifier according to the second discrimination result, the first cross entropy, and the second cross entropy.

### Claim 20 (depends on 17)

The computer-readable storage medium of claim 17, wherein the model loss function comprises a discrimination loss function, where the discrimination loss function is a loss function of a discriminator, and the one or more instructions adapted to be loaded by the processor to train the visualization processing model according to the target image-annotation pair, the first sample image, and the second sample image-annotation pair to obtain the model loss function are adapted to be loaded by the processor to:
discriminate the target image-annotation pair to generate a third discrimination result;
discriminate the first sample image and the predicted annotation to generate a fourth discrimination result;
discriminate the second sample image-annotation pair to generate a fifth discrimination result; and
determine the discrimination loss function of the discriminator according to the third discrimination result, the fourth discrimination result, and the fifth discrimination result.
