CN-120747045-B - Fundus image processing method, fundus image processing system, electronic device and storage medium

CN120747045BCN 120747045 BCN120747045 BCN 120747045BCN-120747045-B

Abstract

The present disclosure discloses a fundus image processing method and system. The method comprises the steps of obtaining a three-dimensional fundus OCT image and at least one fundus OCTA projection image of a target eye, inputting the three-dimensional fundus OCT image into a focus detection model to obtain a first detection result of a fundus focus of the target eye, inputting the at least one fundus OCTA projection image into a blood flow anomaly detection model to obtain a second detection result of the fundus focus of the target eye based on the first detection result of the fundus focus and the at least one fundus blood flow anomaly detection result.

Inventors

WANG XIAO
HAN LIMING
Yang Zhuozhen

Assignees

图湃(北京)医疗科技有限公司

Dates

Publication Date: 20260505
Application Date: 20250819

Claims (9)

1. A method of processing fundus images, comprising: Acquiring a three-dimensional fundus OCT image of a target eye; Selecting at least one fundus OCTA projection map, wherein the at least one fundus OCTA projection map corresponds to at least one retina structure layer one by one; inputting the three-dimensional fundus OCT image into a focus detection model to obtain a first detection result of the fundus focus of the target eye, wherein the first detection result of the fundus focus is an OCT detection result; Inputting the at least one fundus OCTA projection image into a blood flow anomaly detection model to obtain at least one fundus blood flow anomaly detection result of the target eye, wherein the at least one fundus blood flow anomaly detection result is an OCTA detection result, and Projecting the first detection result of the fundus focus on a retina structural layer corresponding to the at least one fundus OCTA projection chart to obtain a fundus focus detection result on the corresponding at least one retina structural layer; matching the fundus focus detection result on the at least one retina structure layer with the corresponding at least one fundus blood flow abnormality detection result to obtain a corresponding at least one matching result; Obtaining a second detection result of the fundus focus of the target eye based on the at least one matching result; Wherein the first detection result of the fundus focus, the at least one fundus blood flow abnormality detection result and the second detection result of the fundus focus are used as intelligent auxiliary diagnosis results for clinical diagnosis of doctors.
2. The method of claim 1, further comprising: Acquiring fundus quantization index of the target eye, and And inputting the second detection result of the fundus focus and the fundus quantitative index into a disease detection model to obtain a fundus disease detection result of the target eye.
3. The method of claim 2, wherein obtaining fundus quantification indicators of the target eye comprises at least one of: based on a retina layering model, performing retina layering on the three-dimensional fundus OCT image to obtain a retina layering result, and obtaining a retina thickness value and/or a choroid thickness value based on the retina layering result; And based on the image segmentation model, carrying out image segmentation on the OCTA projection image of the retina surface layer to obtain a target segmentation region, and obtaining a corresponding retina quantization index based on the target segmentation region.
4. The method of claim 3, wherein the target segmented region comprises at least one of a foveal avascular region, a fundus avascular region.
5. The method of claim 1, wherein the at least one fundus OCTA projection comprises at least one of a retinal surface layer OCTA projection, a retinal deep layer OCTA projection, a retinal avascular layer OCTA projection, a retinal OCTA projection.
6. The method of claim 1, wherein matching fundus focus detection results on the at least one retinal structural layer with corresponding fundus blood flow anomaly detection results comprises at least one of: Matching the fundus focus detection result on the at least one retina structural layer with the corresponding fundus blood flow abnormality detection result in position; And performing morphological feature matching on the fundus focus detection result on the at least one retina structural layer and the corresponding fundus blood flow abnormality detection result.
7. A fundus image processing system, comprising: The image acquisition module is configured to acquire a three-dimensional fundus OCT image of a target eye, and select at least one fundus OCTA projection image, wherein the at least one fundus OCTA projection image corresponds to at least one retina structure layer one by one; the OCT image processing module is configured to input the three-dimensional fundus OCT image into a focus detection model to obtain a first detection result of a fundus focus of the target eye, wherein the first detection result of the fundus focus is an OCT detection result; An OCTA projection map processing module configured to input the at least one fundus OCTA projection map into a blood flow anomaly detection model to obtain at least one fundus blood flow anomaly detection result of the target eye, wherein the at least one fundus blood flow anomaly detection result is an OCTA detection result, and The processing result acquisition module comprises a projection unit, a matching unit and a processing result acquisition unit, wherein the projection unit is configured to project the first detection result of the fundus focus on a retina structure layer corresponding to the at least one fundus OCTA projection chart to obtain a fundus focus detection result on the corresponding at least one retina structure layer; the matching unit is configured to match the fundus focus detection result on the at least one retina structural layer with the corresponding at least one fundus blood flow abnormality detection result to obtain a corresponding at least one matching result; the processing result acquisition unit is configured to obtain a second detection result of the fundus focus of the target eye based on the at least one matching result; Wherein the first detection result of the fundus focus, the at least one fundus blood flow abnormality detection result and the second detection result of the fundus focus are used as intelligent auxiliary diagnosis results for clinical diagnosis of doctors.
8. An electronic device includes a processor, and a memory communicatively coupled to the processor; the memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the fundus image processing method according to any one of claims 1 to 6.
9. A computer-readable storage medium having stored therein computer-executable instructions which, when executed by a processor, are for implementing the fundus image processing method of any of claims 1 to 6.

Description

Fundus image processing method, fundus image processing system, electronic device and storage medium Technical Field The present disclosure relates to the field of AI medical treatment, and in particular, to a fundus image processing method, a fundus image processing system, an electronic apparatus, and a storage medium. Background Both optical coherence tomography (Optical Coherence Tomography angiography, OCT) and optical coherence tomography (Optical Coherence Tomography angiography, OCTA) are non-invasive imaging techniques. OCT images can provide anatomical information of the retina and choroid. The OCTA images can provide not only anatomical information of the retina and choroid, but also microvascular blood flow information of the retina and choroid. Nevertheless, there are still some limitations in the clinical diagnostic process of doctors based on OCT images and OCTA images. For example, the amount of OCTA data on the fundus macula of an individual patient is large, and a physician needs to spend a great deal of time reading to locate the lesion and give a final diagnosis, which is time consuming and severely dependent on the physician's experience. Disclosure of Invention The present disclosure aims to solve, at least to some extent, one of the technical problems in the related art. To achieve the above object, a first aspect of the present disclosure provides a method for processing a fundus image, including: acquiring a three-dimensional fundus OCT image and at least one fundus OCTA projection image of a target eye; inputting the three-dimensional fundus OCT image into a focus detection model to obtain a first detection result of the fundus focus of the target eye; Inputting the at least one fundus OCTA projection map into a blood flow anomaly detection model to obtain at least one fundus blood flow anomaly detection result of the target eye, and And obtaining a second detection result of the fundus focus of the target eye based on the first detection result of the fundus focus and the at least one fundus blood flow abnormality detection result. To achieve the above object, a second aspect of the present disclosure proposes a fundus image processing system including: An image acquisition module configured to acquire a three-dimensional fundus OCT image of a target eye and at least one fundus oculi projection map; An OCT image processing module configured to input the three-dimensional fundus OCT image into a lesion detection model, obtaining a fundus lesion first detection result of the target eye; An OCTA projection map processing module configured to input the at least one fundus OCTA projection map into a blood flow anomaly detection model to obtain at least one fundus blood flow anomaly detection result of the target eye, and A processing result acquisition module configured to obtain a fundus focus second detection result of the target eye based on the fundus focus first detection result and the at least one fundus blood flow abnormality detection result. To achieve the above object, a third aspect of the present disclosure proposes an electronic device, including a processor, and a memory communicatively connected to the processor; the memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the fundus image processing method described in the foregoing first aspect. To achieve the above object, a fourth aspect of the present disclosure proposes a computer-readable storage medium having stored therein computer-executable instructions which, when executed by a processor, are for implementing the fundus image processing method described in the foregoing first aspect. To achieve the above object, a fifth aspect of the present disclosure proposes a computer program product comprising a computer program which, when executed by a processor, implements the method of the first aspect. Drawings The advantages of the various embodiments of the present disclosure will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings, in which: Fig. 1 is a flowchart of a method for processing fundus images according to an embodiment of the present disclosure; fig. 2a is a schematic illustration showing a three-dimensional fundus OCT image provided according to an embodiment of the present disclosure; fig. 2b is a schematic illustration showing a three-dimensional fundus oculi OCTA image provided in accordance with embodiments of the present disclosure; FIG. 3a is a projection of the retinal surface layer OCTA provided in accordance with an embodiment of the present disclosure; FIG. 3b is a projection view of a deep OCTA of the retina provided in accordance with an embodiment of the present disclosure; FIG. 3c is a projection view of retinal avascular layer OCTA provided in accordance with embodiments of the present disclosure; FIG. 3d is a pr