A lesion area segmentation method, device, electronic equipment and storage medium

Abstract

The present invention provides a lesion area segmentation method, device, electronic equipment and storage medium. The method includes: acquiring a near-infrared image of the lesion, and using a preset hue value range to determine the lesion area and the nucleus area of ​​the lesion in the near-infrared image of the lesion; determining the lesion The center of gravity of the nuclear area, and determine the shortest distance between the center of gravity and the edge of the lesion area and the distance between the centers of gravity; use the center of gravity and the shortest distance to establish a circular diffusion area, and use the shortest distance and the distance between the lines to calculate the distance between the circular diffusion area The coordinates of the intersection points between them; use the coordinates of the intersection points to determine the dividing line between the circular diffusion areas, and use the dividing line to segment the lesion area. Based on the feature that the diffusion of the lesion conforms to the diffusion equation, the present invention approximates the diffusion area of ​​the lesion nucleus as a circular area, and then determines the dividing line by determining multiple circular diffusion areas and solving the intersection points between the circular diffusion areas. Finally, the dividing line can be quickly determined and the lesion area can be effectively divided.

Description

technical field [0001] The present invention relates to the field of medicine, in particular to a lesion area segmentation method, device, electronic equipment and storage medium. Background technique [0002] Lesion refers to the part of the body where the disease occurs. Since multiple lesions usually appear in a certain area of ​​the body and overlap each other, in order to reduce the complexity of analyzing the lesions by medical personnel, it is usually necessary to segment the lesion area. [0003] In related technologies, the estimation method, convolutional neural network or regional competition method is generally used for lesion segmentation. However, the calculation accuracy of the estimation method is low. The convolutional neural network requires a large number of lesion samples and the calculation is complicated. The regional competition method has boundaries that are difficult to locate. The problem. In other words, the above method has problems of poor calc...

AI Technical Summary

Problems solved by technology

[0003] In related technologies, the estimation method, convolutional neural network or regional competition method is generally used for lesion segmentation. However, the calculation accuracy of the estimation method is low. The convolutional neural network requires a large number of lesion samples and the calculation is complicated. The regional competition method has boundaries that are difficult to locate. The problem

In other words, the above method has problems of poor calculation accuracy and flexibility, which in turn reduces the reliability and flexibility of lesion region segmentation

Images