Diabetic retinopathy recognition system based on fundus image
Some embodiments of the disclosure provide a diabetic retinopathy recognition system (S) based on fundus image. According to an embodiment, the system includes an image acquisition apparatus ( 1 ) configured to collect fundus images. The fundus images include target fundus images and reference fundus images taken from a person. The system further includes an automatic recognition apparatus ( 2 ) configured to process the fundus images from the image acquisition apparatus by using a deep learning method. The automatic recognition apparatus automatically determines whether a fundus image has a lesion and outputs the diagnostic result. According to another embodiment, the diabetic retinopathy recognition system (S) utilizes a deep learning method to automatically determine the fundus images and output the diagnostic result.
1. A diabetic retinopathy recognition system, comprising:
an image acquisition apparatus configured to collect fundus images, the fundus images comprise target fundus images and reference fundus images taken from a person; and
an automatic recognition apparatus configured to process the fundus images from the image acquisition apparatus by using a deep learning method, and automatically determine whether a fundus image has a lesion and output a diagnostic result;
wherein the automatic recognition apparatus comprises:
a pre-processing module configured to separately pre-process the target fundus images and the reference fundus images;
a first neural network configured to generate a first advanced feature set from the target fundus images;
a second neural network configured to generate a second advanced feature set from the reference fundus images;
a feature combination module configured to combine the first advanced feature set and the second advanced feature set to form a feature combination set; and
a third neural network configured to generate a diagnosis result of lesions according to the feature combination set.
2. The diabetic retinopathy recognition system of claim 1 , further comprising an output apparatus that outputs an analysis report according to the diagnostic result.
3. The diabetic retinopathy recognition system of claim 1 , wherein the image acquisition apparatus is a handheld fundus camera.
4. The diabetic retinopathy recognition system of claim 1 , wherein the automatic recognition apparatus is arranged in a cloud server, and the image acquisition apparatus interacts with the automatic recognition apparatus based on network communication.
5. The diabetic retinopathy recognition system of claim 1 , wherein the target fundus images, and the reference fundus images are the same.
6. The diabetic retinopathy recognition system of claim 1 , wherein the target fundus images and the reference fundus images are fundus images of different eyes.
7. The diabetic retinopathy recognition system of claim 1 , wherein the first neural network and the second neural network are the same.
8. The diabetic retinopathy recognition system of claim 1 , wherein the pre-processing module comprises:
an area detection unit configured to detect designated fundus areas in the target fundus images and the reference fundus images;
an adjustment unit configured to clip and resize the target fundus images and the reference fundus images; and
a normalization unit, configured to normalize the target fundus images and the reference fundus images.
9. The diabetic retinopathy recognition system of claim 1 , wherein the third neural network generates the diagnosis result of lesions based on the feature combination set and patient information.