Method for integrated analysis of spinal disorders and electronic apparatus for performing this analysis

Abstract

A method for integratively analyzing spinal diseases using an electronic device according to one embodiment of this application may include the steps of: inputting a spinal image; applying the spinal image to a learned spinal segment detection model to detect multiple vertebral segments; measuring the height and width of multiple vertebrae and the distance between multiple vertebrae based on the vertebral segments to generate measurement indices; applying the vertebral segments to one or more learned imaging disease analysis models to generate disease information for the vertebrae; applying the measurement indices and disease information to a learned integrated analysis model to generate integrated disease information for the vertebrae; and outputting the integrated disease information. According to one embodiment of the present invention, there is an advantage in that the type of spinal disease and the spinal disease area can be determined by integratively analyzing an X-ray spinal image.

Claims

1. In a method of integrating and analyzing spinal disorders using electronic devices, The stage where spinal video is input; The step of applying the aforementioned spinal video to a trained vertebral segment detection model to detect multiple vertebral segments; A step of generating measurement indicators by measuring the height and width of multiple vertebrae and the distance between multiple vertebrae based on the segmentation of the vertebrae; A step of generating disease information for the vertebrae by applying the segments of the vertebrae to one or more learned visual disease analysis models; The step of applying the aforementioned measurement indicators and disease information to a learned integrated analysis model to generate integrated disease information of the vertebrae; and A spinal disease integrated analysis method, comprising the step of outputting the aforementioned integrated disease information.

2. The step of applying the vertebral segments to one or more learned visual disease analysis models to generate disease information for the vertebrae is: The step of applying the aforementioned vertebral segmentation to a learned disease detection model to determine the first disease area; A step of applying the vertebral segments to a learned disease classification model to determine the type of disease corresponding to the vertebra; The step of applying the vertebral segments to a learned segmentation model to determine a second disease area; and A spinal disease integrated analysis method according to claim 1, comprising the step of generating disease information of the vertebrae based on the first disease area, the type of disease, and the second disease area.

3. The step of applying the vertebral segments to a learned disease detection model to determine a first disease region in the spinal image is: The step of extracting multiple regions of interest from the aforementioned spinal image; The step of determining the disease area for each of the aforementioned multiple areas of interest: and The spinal disease integrated analysis method according to claim 2, further comprising the step of determining the region where the disease regions determined in each of the aforementioned multiple regions of interest overlap as a first disease region.

4. The step of extracting multiple regions of interest from the aforementioned spinal video is as follows: The step of extracting a lumbar spine image from the aforementioned spinal video; The step of extracting an image containing multiple lumbar vertebrae from the aforementioned spinal image; and The spinal disease integrated analysis method according to claim 3, comprising the step of extracting an image containing a single lumbar vertebra from the aforementioned spinal image.

5. The step of applying the vertebral segments to a learned disease classification model to determine the type of disease corresponding to the vertebra is: The step of extracting multiple regions of interest from the aforementioned spinal image; The step of determining the type of disease corresponding to the vertebrae in each of the aforementioned multiple areas of interest: and The spinal disease integrated analysis method according to claim 2, further comprising the step of determining that the types of diseases that overlap as determined in each of the aforementioned multiple areas of interest are the types of diseases corresponding to the vertebrae.

6. The step of applying the vertebral segments to the learned region segmentation model to determine the second disease region in the spinal image is: The step of extracting multiple regions of interest from the aforementioned spinal image; The step of determining the disease area for each of the aforementioned multiple areas of interest: and The spinal disease integrated analysis method according to claim 2, further comprising the step of determining a second disease area in which the disease areas determined in each of the aforementioned multiple areas of interest overlap.

7. The step of applying the aforementioned spinal video to a trained vertebral segment detection model to detect multiple vertebral segments is: A step of detecting each vertebra included in the aforementioned spinal image using a bounding box; and The spinal disease integrated analysis method according to claim 1, comprising the step of detecting the corners of each vertebra with the bounding box;

8. The step of generating measurement indicators by measuring the height and width of multiple vertebrae and the distance between multiple vertebrae based on the segmentation of the vertebrae is as follows: A step of measuring the height and width of the plurality of vertebrae and the distance between the plurality of vertebrae based on the angle of each of the vertebrae; and The spinal disease integrated analysis method according to claim 7, further comprising the step of determining a vertebra or intervertebral space that deviates from the normal standard based on the measurement index.

9. The step of applying the aforementioned measurement indicators and disease information to a learned integrated analysis model to generate integrated disease information for the vertebrae is: The spinal disease integrated analysis method according to claim 1, comprising: (i) determining the type of integrated disease and the integrated disease area based on the vertebrae or intervertebral spacing that deviate from the normal standard determined based on the measurement information and (ii) determining the disease area and type of disease determined based on the imaging disease analysis model.

10. A computer-readable recording medium having a program stored on it that causes a computer to perform the method described in any one of claims 1 to 9.

11. In the spinal disease integrated analysis device, A communication unit that inputs spinal video and outputs integrated spinal disease information; Processor; and Includes memory; The aforementioned processor, A spinal segment detection unit that detects multiple vertebral segments by applying the aforementioned spinal video to a trained spinal segment detection model; A vertebral measurement unit that measures the height and width of multiple vertebrae and the distance between multiple vertebrae based on the segmentation of the vertebral bones and generates measurement indicators; A visual disease analysis unit that applies the segments of the vertebrae to one or more learned visual disease analysis models to generate disease information of the vertebrae; and An electronic device comprising: an integrated analysis unit that applies the aforementioned measurement indicators and disease information to a learned integrated analysis model to generate integrated disease information of the vertebrae;

Images