Auxiliary evaluation method for renal interstitial fibrosis, computer equipment and computer readable storage medium

Abstract

The invention discloses an auxiliary evaluation method for renal interstitial fibrosis, computer equipment and a computer readable storage medium, and the method can accurately determine the area value of a renal interstitial fibrosis region, and can jointly judge the lesion of interstitial fibrosis in combination with other parameter values (such as area values provided by clinical diagnosis and algorithms). The method is based on deep learning, enables the model to have better robustness and generalization ability, can obtain better accuracy compared with an existing detection method, and can improve the detection efficiency.

Description

technical field [0001] The invention specifically relates to an auxiliary assessment method for renal interstitial fibrosis, computer equipment and a computer-readable storage medium. Background technique [0002] The renal interstitium is the connective tissue located between the nephron and the collecting duct, and is composed of interstitial cells, reticular fibers, a small amount of elastic fibers, fibroblasts, and macrophages. Renal interstitium is damaged due to inflammation, genetics or other diseases, which can involve renal tubules, renal pelvis, and calices, resulting in dysuria and disturbance of body electrolyte and acid-base balance. [0003] Fibrosis can occur in a variety of organs. The main pathological changes are the increase of fibrous connective tissue and the reduction of parenchymal cells in organ tissues. Continuous progress can lead to organ structural damage, functional decline, and even failure, which seriously threaten human health and life. [00...

AI Technical Summary

Problems solved by technology

Although traditional image processing methods can handle current image samples, they do not have good generalization ability for large-scale data sets, and need to manually set values ​​for individual image data (the values ​​refer to the values ​​in the function when processing images in traditional methods. value), which leads to an increase in the cost of sample processing time. In addition, manually setting the value cannot obtain the optimal parameter value, which also leads to a low accuracy rate.

Images