US 12,086,982 B2
Method for confirming cup-disc ratio based on neural network, apparatus, device, and computer readable storage medium
Ge Li, Guangdong (CN); Guanju Cheng, Guangdong (CN); Chan Zeng, Guangdong (CN); Peng Gao, Guangdong (CN); and Guotong Xie, Guangdong (CN)
Assigned to PING AN TECHNOLOGY (SHENZHEN) CO., LTD., Shenzhen (CN)
Appl. No. 17/612,566
Filed by PING AN TECHNOLOGY (SHENZHEN) CO., LTD., Guangdong (CN)
PCT Filed Oct. 30, 2020, PCT No. PCT/CN2020/125009
§ 371(c)(1), (2) Date Nov. 18, 2021,
PCT Pub. No. WO2021/189849, PCT Pub. Date Sep. 30, 2021.
Claims priority of application No. 202010992346.X (CN), filed on Sep. 21, 2020.
Prior Publication US 2022/0309654 A1, Sep. 29, 2022
Int. Cl. G06T 7/00 (2017.01); G06T 7/11 (2017.01); G06T 7/12 (2017.01); G06T 7/62 (2017.01)
CPC G06T 7/0012 (2013.01) [G06T 7/11 (2017.01); G06T 7/12 (2017.01); G06T 7/62 (2017.01); G06T 2207/20084 (2013.01); G06T 2207/30041 (2013.01)] 16 Claims
OG exemplary drawing
 
1. A method for confirming a cup-disc ratio based on a neural network; the method comprising following steps:
obtaining a retinal image, and detecting an optical disc region in the retinal image to obtain the optical disc region;
inputting the optical disc region into a pre-trained neural network stored in a storage, by a processor, and setting a result outputted by the pre-trained neural network as a prediction cup-disc ratio and segment images of an optical cup and an optical disc;
computing a cup-disc ratio based on the segment images of the optical cup and the optical disc; and
confirming a practical cup-disc ratio based on the prediction cup-disc ratio and the computed cup-disc ratio;
wherein the neural network comprises a feature extraction layer, a convolution layer, a connection layer, an image segment layer, and a regression forecasting layer; the step of inputting the optical disc region into a pre-trained neural network to obtain a prediction cup-disc ratio and segment images of an optical cup and an optical disc further comprising:
extracting a feature of the optical disc region by the feature extraction layer to obtain a first feature image;
executing a convolution of the first feature image by the convolution layer to obtain a second feature image representing a segment image of the cup-disc ratio and a third feature image representing a segment image of the optical disc;
connecting the second feature image and the third feature image by the connection layer to obtain a fourth feature image; and
inputting the fourth feature image into the regression forecasting layer to obtain the prediction cup-disc ratio and the segments images of the optical cup and the optical disc.