A method for monitoring the state of thyroid eye disease, and a system for performing this monitoring.

Abstract

A method for monitoring the entire treatment of thyroid eye disease is proposed, and the method includes: a step of requesting a user device to take a facial image and input a questionnaire survey content according to a treatment monitoring cycle during a treatment period in which a therapeutic drug proven to be effective in the treatment of thyroid eye disease through clinical trials is prescribed and administered to the user; a step of acquiring the facial image and the questionnaire survey content from the user device; a step of acquiring personalized estimates of the user corresponding to exophthalmos information, CAS (Clinical Activity Score) information, and diplopia information among the indicators proven in the approval stage of the therapeutic drug using the facial image and the questionnaire survey content acquired from the user device; and a step of storing the personalized estimates of the user in the user device in chronological order from the time the user started administering the therapeutic drug. the displaying step including visualizing on the user device by comparing the user's personalized estimate corresponding to the exophthalmos information with the exophthalmos information at the end of the administration of the therapeutic agent; and the displaying step including visualizing on the user device by comparing the user's personalized estimate corresponding to the exophthalmos information with the exophthalmos information at the end of the administration of the therapeutic agent; and the displaying step including visualizing on the user device a message suggesting a visit to a hospital based on a difference between the user's personalized estimate corresponding to the exophthalmos information and the exophthalmos information at the end of the administration of the therapeutic agent.

Claims

1. A method for estimating the exophthalmos value of a target, which is performed by one or more processors: The communication unit receives a frontal face image representing at least both eyes, the nose, and the eyebrows of the subject. Here, the frontal face image includes a plurality of pixels, each of which includes at least one of the pixel values, where the pixel value is not a depth value indicating calculated distance information; The step of preprocessing the frontal face image to obtain input data, wherein the preprocessing includes at least segmenting the region relating to at least one eye of the subject; and The step of estimating the exophthalmos value for the subject by applying the input data to a pre-trained exophthalmos estimation model, wherein the pre-trained exophthalmos estimation model is trained using a set of training input data and label data, wherein the training input data and label data are acquired by medical staff in at least one clinical setting, wherein the training input data includes a clinical frontal face image acquired by capturing an image of at least one patient, wherein the label data is acquired by measuring the eyes of at least one patient. A method for providing this.

2. The method according to claim 1, wherein the pre-trained exophthalmos estimation model includes at least one of the following: a linear regression model, a polynomial regression model, a ridge regression model, a lasso regression model, a support vector machine (SVM) model, a decision tree regression model, a random forest regression model, a k-nearest neighbors (KNN) model, a feedforward neural network model, a convolutional neural network (CNN) model, a recurrent neural network (RNN) model, a long-term memory (LSTM) network model, a gated recurrent unit (GRU) model, a gradient boosting model, a LightGBM model, a CatBoost model, and an AdaBoost model.

3. The method according to claim 1, wherein the preprocessing of the frontal face image includes aligning the horizontal orientation of the frontal face image.

4. Aligning the horizontal orientation of the aforementioned frontal facial image is: Determining the central position of the pupils of both eyes of the subject in the frontal face image; and Adjusting the orientation of the frontal face image so that the line connecting the central positions of the pupils of both eyes is horizontal. The method according to claim 3, including the method described in claim 3.

5. The method according to claim 1, wherein the preprocessing of the frontal face image includes obtaining a plurality of Radial MPLD (Mid-Pupil Lid Distance) values ​​for at least one of the eyes of the subject in the frontal face image, wherein the plurality of Radial MPLD values ​​include a first Radial MPLD value corresponding to a first angle and a second Radial MPLD value corresponding to a second angle.

6. The method according to claim 5, wherein the input data includes the sum of the first Radial MPLD value and the second Radial MPLD value.

7. The input data further includes the horizontal length of at least one of the two eyes, Here, the first angle corresponds to 0 degrees, Here, the second angle corresponds to 180 degrees, Here, 0 degrees is defined as the direction from the center of the pupil of at least one of the eyes in the frontal face image to the lacrimal caruncle. The method according to claim 6.

8. The method according to claim 7, wherein the input data further includes the central position of the pupil of at least one of the eyes, and the length between a straight line connecting the left and right ends of at least one of the eyes.

9. The input data further includes the vertical length of at least one of the two eyes, Here, the first angle corresponds to 90 degrees, Here, the second angle corresponds to 270 degrees, Here, 0 degrees is defined as the direction from the center of the pupil of at least one of the eyes in the frontal face image to the lacrimal caruncle. The method according to claim 6.

10. The method according to claim 9, wherein the input data further includes the central position of the pupil of at least one of the eyes, and the length between a straight line connecting the upper and lower ends of at least one of the eyes.

11. The method according to claim 1, wherein the input data includes a segmentation image representing the region relating to both of the eyes of the subject.

12. The method according to claim 11, wherein the region relating to both of the eyes of the subject includes the ocular region and the pupil region of both of the eyes of the subject.

13. The method according to any one of claims 1 to 12, wherein the preprocessing of the frontal face image includes obtaining a plurality of 3D face landmark coordinate values, wherein the plurality of 3D face landmark coordinate values ​​include a first landmark coordinate value and a second landmark coordinate value.

14. The method according to claim 13, wherein the input data includes the z-axis length value between the first landmark coordinate value and the second landmark coordinate value.

15. The first landmark coordinate value represents the location of the outer edge of the pupil of at least one of the two eyes, The second landmark coordinate value represents the location of the corner on the outer circumference of at least one of the two eyes. The method according to claim 14.

Images