Tongue-image-based diagnostic system and diagnostic method

1. A tongue-image-based diagnostic system, comprising:

a parameter collector configured to acquire parameter information corresponding to image training data in real time, wherein the parameter collector comprises: a temperature collector configured to collect temperature information, a brightness collector configured to collect brightness information, and a physiological parameter collector configured to collect physiological parameter information, wherein the parameter collector is further configured to acquire at least one of drug administration information or dietary parameter information of a user;

a model establishment circuitry configured to perform a training process using a neural network algorithm and using the image training data and the parameter information to establish an estimation model, wherein the image training data comprises a first tongue image for training, each first tongue image is associated with the temperature information and the brightness information, the physiological parameter information and at least one of the drug administration information or the dietary parameter information of the user which are acquired in real time when acquiring the first tongue image; and

an analysis circuitry configured to analyze acquired image information using the estimation model, and to generate at least an analysis result about a probability that the user related to the acquired image information suffers from a cancer, wherein the acquired image information comprises a second tongue image of the user to be diagnosed.

2. The diagnostic system according to claim 1 , wherein the physiological parameter information comprises at least one of body temperature information, respiratory rate information or heart rate information of a user, and

the model establishment circuitry is further configured to perform the training process using the image training data, the parameter information corresponding to the image training data and the physiological parameter information, and establish the estimation model.

3. The diagnostic system according to claim 1 , wherein the model establishment circuitry is further configured to validate the estimation model, and

the analysis circuitry is further configured to, if the model establishment circuitry successfully validates the estimation model, analyze the second tongue image using the validated estimation model and generate the analysis result corresponding to the second tongue image.

4. The diagnostic system according to claim 1 , further comprising:

an image collection assembly comprising an image collector and configured to acquire at least one of the first tongue image and the second tongue image with the image collector, and transmit the first tongue image to the model establishment circuitry and transmit the second tongue image to the analysis circuitry.

5. The diagnostic system according to claim 4 , wherein the image collection assembly further comprises:

a light source configured to generate light;

a light brightness sensor configured to collect a brightness value of the environment; and

a brightness adjustment circuitry configured to:

receive the brightness value from the light brightness sensor,

compare the received brightness value with a predetermined brightness value,

if the received brightness value is different from the predetermined brightness value, adjust the brightness value of the light generated by the light source until the brightness value collected by the light brightness sensor is the same as the predetermined brightness value, and

if the brightness value collected by the light brightness sensor is the same as the predetermined brightness value, control the image collector to collect image information.

6. The diagnostic system according to claim 1 , wherein the image training data further comprises health condition data corresponding to the first tongue image, and the health condition data comprises information about the user suffering from the cancer or information about a probability that the user suffers from the cancer.

7. The diagnostic system according to claim 4 , wherein the image collector is further configured to acquire an oral image of the user when acquiring the first tongue image, and the model establishment circuitry is further configured to analyze the oral image.

8. The diagnostic system according to claim 3 , wherein the model establishment circuitry is further configured to:

validate the estimation model using a plurality of groups of validation data, wherein each group of validation data comprises image information and a health estimation result corresponding to the image information; and

based on coherence of validation results of the plurality of groups of validation data with the health estimation result being greater than a predetermined probability, determine that the estimation model is validated successfully.

9. A tongue-image-based diagnostic method, comprising:

acquiring parameter information corresponding to image training data in real time, wherein the parameter information comprises temperature information, brightness information, physiological parameter information and at least one of drug administration information or dietary parameter information of a user;

performing a training process using a neural network algorithm and using the image training data and the parameter information, and establishing an estimation model, wherein the image training data comprises a first tongue image for training, each first tongue image is associated with the temperature information and the brightness information, the physiological parameter information and at least one of the drug administration information or the dietary parameter information of the user which are acquired in real time when acquiring the first tongue image; and

analyzing acquired image information using the estimation model, and generating an analysis result about a probability that the user corresponding to the acquired image information suffers from a cancer, wherein the acquired image information comprises a second tongue image of the user to be diagnosed.

10. The diagnostic method according to claim 9 , further comprising:

validating the estimation model;

if the estimation model is validated successfully, analyzing the second tongue image using the validated estimation model; and

generating the analysis result corresponding to the second tongue image.

11. The diagnostic method according to claim 10 , wherein the image training data further comprises health condition data of the user corresponding to the first tongue image, and the health condition data comprises information about the user corresponding to the first tongue image suffering from a cancer or information about a probability that the user suffers from the cancer.

12. A tongue-image-based diagnostic system, comprising:

a processor, and

a memory storing computer programs executable by the processor,

wherein the computer programs, when executed by the processor, cause the processor to perform the diagnostic method according to claim 9 .

13. A tongue-image-based diagnostic system, comprising:

a parameter collector configured to acquire parameter information corresponding to image training data in real time, wherein the parameter collector comprises: a temperature collector configured to collect temperature information, a brightness collector configured to collect brightness information, and a physiological parameter collector configured to collect physiological parameter information, wherein the parameter collector is further configured to acquire at least one of drug administration information or dietary parameter information of a user;

a model establishment circuitry configured to perform a training process using a neural network algorithm and using the image training data and the parameter information to establish an estimation model, wherein the image training data comprises first tongue image for training, and each first tongue image is associated with the temperature information the brightness information, the physiological parameter information and at least one of the drug administration information or the dietary parameter information of the user which are acquired in real time when acquiring the first tongue image;

an image collection assembly configured to acquire second tongue image;

a user management circuitry configured to store the second tongue image with information about the user, and transmit the second tongue image to an analysis circuitry; and

the analysis circuitry configured to analyze the second tongue image using the estimation model, and to generate at least an analysis result about a probability that the user related to the acquired image information suffers from a cancer, wherein the second tongue image comprises tongue image of the user to be diagnosed;

wherein the user management circuitry is further configured to receive the analysis result from the analysis circuitry and store the analysis result with the information about the user, to establish a personal profile database for each user.

14. The diagnostic method according to claim 9 , wherein:

after performing the training process using image training data and the parameter information, and establishing the estimation model, the diagnostic method further comprises:

acquiring the second tongue image; and

storing the second tongue image with the information about the user; and

after the analyzing acquired image information using the estimation model, and generating an analysis result about a probability that a user corresponding to the acquired image information suffers from a cancer, the diagnostic method further comprises:

storing the analysis result with the information about the user, to establish a personal profile database for each user.

15. The diagnostic system according to claim 1 , further comprising:

a data generation circuitry configured to perform an image conversion operation on the first tongue image information, and generate the image training data based one at least one of the first tongue image information or the converted first tongue image information.

16. The diagnostic system according to claim 15 , wherein the image conversion operation comprises at least one of an image translation operation or an image rotation operation, or at least one of an image segmentation operation or an image annotation operation.

17. The diagnostic method according to claim 9 , wherein before performing a training process using received image training data and the parameter information, and establishing an estimation model, the diagnostic method further comprises:

acquiring the first tongue image information;

performing an image conversion operation on the acquired first tongue image information; and

generating the image training data using at least one of the first tongue image information or the converted first tongue image information.

18. The diagnostic method according to claim 17 , wherein the performing an image conversion operation on the acquired first tongue image information comprises:

performing at least one of an image segmentation operation or an image annotation operation on the first tongue image information, or

performing at least one of an image translation operation or an image rotation operation on the first tongue image information.