CN111920375B - Vascular Endothelial Growth Factor (VEGF) resistance curative effect prediction device and method - Google Patents
Vascular Endothelial Growth Factor (VEGF) resistance curative effect prediction device and method Download PDFInfo
- Publication number
- CN111920375B CN111920375B CN202011061099.8A CN202011061099A CN111920375B CN 111920375 B CN111920375 B CN 111920375B CN 202011061099 A CN202011061099 A CN 202011061099A CN 111920375 B CN111920375 B CN 111920375B
- Authority
- CN
- China
- Prior art keywords
- oct
- image
- oct images
- vegf
- images
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B3/00—Apparatus for testing the eyes; Instruments for examining the eyes
- A61B3/10—Objective types, i.e. instruments for examining the eyes independent of the patients' perceptions or reactions
- A61B3/102—Objective types, i.e. instruments for examining the eyes independent of the patients' perceptions or reactions for optical coherence tomography [OCT]
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B3/00—Apparatus for testing the eyes; Instruments for examining the eyes
- A61B3/0016—Operational features thereof
- A61B3/0025—Operational features thereof characterised by electronic signal processing, e.g. eye models
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B3/00—Apparatus for testing the eyes; Instruments for examining the eyes
- A61B3/10—Objective types, i.e. instruments for examining the eyes independent of the patients' perceptions or reactions
- A61B3/12—Objective types, i.e. instruments for examining the eyes independent of the patients' perceptions or reactions for looking at the eye fundus, e.g. ophthalmoscopes
- A61B3/1225—Objective types, i.e. instruments for examining the eyes independent of the patients' perceptions or reactions for looking at the eye fundus, e.g. ophthalmoscopes using coherent radiation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/253—Fusion techniques of extracted features
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H30/00—ICT specially adapted for the handling or processing of medical images
- G16H30/20—ICT specially adapted for the handling or processing of medical images for handling medical images, e.g. DICOM, HL7 or PACS
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
- G06T2207/10101—Optical tomography; Optical coherence tomography [OCT]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30101—Blood vessel; Artery; Vein; Vascular
Abstract
The application relates to the field of medical science and technology, and particularly discloses a Vascular Endothelial Growth Factor (VEGF) resistance curative effect prediction device and method. The device includes: the system comprises an acquisition module, a processing module and a processing module, wherein the acquisition module is used for acquiring a plurality of Optical Coherence Tomography (OCT) images acquired aiming at a macular region; the characteristic extraction module is used for extracting the characteristics of the plurality of OCT images to obtain a plurality of image characteristics corresponding to the plurality of OCT images, wherein the image characteristic corresponding to each OCT image in the plurality of OCT images comprises a multi-scale characteristic in each OCT image; the spatial information fusion module is used for carrying out spatial information fusion on a plurality of image characteristics corresponding to the plurality of OCT images to obtain characteristics corresponding to the plurality of OCT images; and the determining module is used for determining the prediction result of the anti-VEGF curative effect according to the corresponding characteristics of the plurality of OCT images. The embodiment of the application is beneficial to improving the accuracy of the prediction of the anti-VEGF curative effect.
Description
Technical Field
The application relates to the field of medical science and technology, in particular to a Vascular Endothelial Growth Factor (VEGF) resistance curative effect prediction device and method.
Background
Wet age-related macular degeneration (AMD) is a major blinding eye disease. Intraocular injection of anti-Vascular Endothelial Growth Factor (VEGF) is an effective treatment for wet AMD, but the cost of anti-VEGF injection therapy is high and has strict indications, which vary from patient to patient. Due to the lack of effective prediction of anti-VEGF efficacy, physicians often resort to uniform injection of VEGF into patients, resulting in anti-VEGF injections also in some patients who are not suitable. Therefore, prediction of effective anti-VEGF therapeutic effects is an urgent need of physicians.
Optical Coherence Tomography (OCT) is a device commonly used at present for diagnosing ophthalmic diseases, and provides a reference in an image for detection, treatment, and the like of ophthalmic diseases using a light reflection technique like ultrasonic imaging. In the currently commonly used anti-VEGF therapeutic effect prediction method, a focus area (such as effusion, high-reflection points and the like) is segmented through a segmentation network, and then anti-VEGF therapeutic effect prediction is performed after segmentation. However, in the training process of the segmentation network based on deep learning, a large number of labels of doctors are required, and the accuracy of the labels and the accuracy of segmentation of the segmentation network affect the curative effect prediction result. In addition, in the features extracted based on the segmented network, a large amount of retina tissue change information which can improve the accuracy of the prediction of the anti-VEGF curative effect is lost, so that the accuracy of the prediction of the anti-VEGF curative effect is low.
Disclosure of Invention
The application provides a device and a method for predicting the curative effect of anti-vascular endothelial growth factor VEGF, which are beneficial to improving the accuracy of predicting the curative effect of anti-VEGF.
The first aspect of the present application provides a device for predicting the therapeutic effect of VEGF, comprising: the system comprises an acquisition module, a processing module and a processing module, wherein the acquisition module is used for acquiring a plurality of Optical Coherence Tomography (OCT) images acquired aiming at a macular region; the characteristic extraction module is used for extracting the characteristics of the plurality of OCT images to obtain a plurality of image characteristics corresponding to the plurality of OCT images, wherein the image characteristic corresponding to each OCT image in the plurality of OCT images comprises a multi-scale characteristic in each OCT image; the spatial information fusion module is used for carrying out spatial information fusion on a plurality of image characteristics corresponding to the plurality of OCT images to obtain characteristics corresponding to the plurality of OCT images; and the determining module is used for determining the prediction result of the anti-VEGF curative effect according to the corresponding characteristics of the plurality of OCT images.
In a second aspect, the present application provides a method for predicting the therapeutic effect of VEGF, comprising: acquiring a plurality of Optical Coherence Tomography (OCT) images collected aiming at a macular region; performing feature extraction on the plurality of OCT images to obtain a plurality of image features corresponding to the plurality of OCT images, wherein the image feature corresponding to each OCT image in the plurality of OCT images comprises a multi-scale feature in each OCT image; performing spatial information fusion on a plurality of image characteristics corresponding to the plurality of OCT images to obtain characteristics corresponding to the plurality of OCT images; and determining the prediction result of the anti-VEGF curative effect according to the corresponding characteristics of the plurality of OCT images.
A third aspect of the present application provides an electronic device comprising a processor, a memory, a communication interface, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the processor, the programs comprising instructions for performing the steps of the method of any of the second aspects of the present application.
A fourth aspect of the present application provides a computer readable storage medium having a computer program stored thereon for execution by a processor to perform some or all of the steps described in any of the methods of the second aspect of the present application.
It can be seen that, according to the prediction device and method for the therapeutic effect of the anti-vascular endothelial growth factor VEGF provided by the application, firstly, a plurality of Optical Coherence Tomography (OCT) images collected for a macular region are acquired. Secondly, performing feature extraction on the plurality of OCT images to obtain a plurality of image features corresponding to the plurality of OCT images, wherein the image feature corresponding to each OCT image in the plurality of OCT images comprises a multi-scale feature in each OCT image. Secondly, spatial information fusion is carried out on a plurality of image characteristics corresponding to the plurality of OCT images, and characteristics corresponding to the plurality of OCT images are obtained. And finally, determining the prediction result of the anti-VEGF curative effect according to the corresponding characteristics of the plurality of OCT images. Therefore, after a plurality of OCT images are acquired, segmentation is not needed, the labeling cost in a segmentation network is reduced, and meanwhile accuracy errors of prediction of the anti-VEGF curative effect caused by inaccurate segmentation and the like are avoided. In addition, the extracted image characteristics corresponding to each OCT image comprise multi-scale characteristics, and the richness and comprehensiveness of the extracted characteristics are increased, so that the accuracy of the prediction of the anti-VEGF curative effect is improved. In addition, spatial information fusion is carried out on the characteristics of a plurality of images, the spatial characteristics of a plurality of OCT images are effectively utilized, the spatial information of the characteristics is enriched, and the accuracy of prediction of the anti-VEGF curative effect is improved.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings without creative efforts.
Fig. 1 is a schematic diagram of a net structure for predicting the therapeutic effect of VEGF, which is an anti-VEGF agent according to an embodiment of the present disclosure.
Fig. 2 is a schematic flowchart of a method for predicting therapeutic effect of VEGF, which is an embodiment of the present disclosure.
Fig. 3 is a schematic diagram of a feature extraction network according to an embodiment of the present application.
FIG. 4 is a schematic flow chart of another method for predicting the therapeutic effect of VEGF on VEGF.
Fig. 5 is a schematic diagram of an apparatus for predicting VEGF therapeutic effect according to an embodiment of the present disclosure.
Fig. 6 is a schematic structural diagram of an electronic device in a hardware operating environment according to an embodiment of the present application.
Detailed Description
The device and the method for predicting the anti-Vascular Endothelial Growth Factor (VEGF) curative effect are beneficial to improving the accuracy of predicting the anti-VEGF curative effect.
In order to make the technical solutions better understood by those skilled in the art, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only partial embodiments of the present application, but not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
The term "at least one" as used in the embodiments of the present application means one or more, and the "plurality" means two or more. "and/or" describes the association relationship of the associated objects, meaning that there may be three relationships, e.g., a and/or B, which may mean: a exists alone, A and B exist simultaneously, and B exists alone, wherein A and B can be singular or plural. The character "/" generally indicates that the former and latter associated objects are in an "or" relationship. "at least one of the following" or similar expressions refer to any combination of these items, including any combination of the singular or plural items.
The terms "first," "second," "third," "fourth," and the like in the description and claims of this application and in the above-described drawings are used for distinguishing between different objects and not necessarily for describing a particular sequential or chronological order, priority, or degree of importance. For example, the first information and the second information are only for distinguishing different information, and do not indicate a difference in content, priority, transmission order, or importance of the two information. Furthermore, the terms "include" and "have," as well as any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, system, article, or apparatus that comprises a list of steps or elements is not limited to only those steps or elements listed, but may alternatively include other steps or elements not listed, or inherent to such process, method, article, or apparatus.
To facilitate understanding of the present application, the concepts related to the present application will be explained first.
Wet age-related macular degeneration (AMD): AMD is a major blinding eye disease, an aging change in the structure of the macular region. The main manifestation is that the phagocytic and digestive ability of retinal pigment epithelial cells to the outer segment disc membrane of the eye cells is reduced, and as a result, the residual bodies of the disc membrane which are not completely digested are retained in the primary pulp of basal cells and are discharged out of the cells to be deposited on the Bruch membrane to form drusen.
Vascular Endothelial Growth Factor (VEGF): VEGF, also known as Vascular Permeability Factor (VPF), is a highly specific vascular endothelial cell growth factor and has the effects of promoting vascular permeability, extracellular matrix degeneration, vascular endothelial cell migration, proliferation, and angiogenesis. Intraocular injection of anti-VEGF is an effective treatment for wet AMD, but anti-VEGF injection is expensive and has strict indications that vary from patient to patient.
Optical Coherence Tomography (OCT) is a device commonly used at present for diagnosing ophthalmic diseases, and provides a reference in an image for detection, treatment, and the like of ophthalmic diseases using a light reflection technique like ultrasonic imaging.
As described above in the background of the present application, the technical features of the embodiments of the present application are described below.
Referring first to fig. 1, fig. 1 is a schematic diagram of a net structure for predicting the therapeutic effect of VEGF, an anti-VEGF according to an embodiment of the present invention. As shown in fig. 1, the network structure in the embodiment of the present application includes a feature extraction network for extracting features of OCT images and a long-term and short-term memory artificial neural network for fusing spatial information of multiple OCT images, so as to finally realize prediction of an anti-VEGF therapeutic effect.
First, a plurality of optical coherence tomography OCT images acquired for the macular region are acquired. In one possible embodiment, the plurality of OCT images may be twelve line OCT images, i.e., 12 OCT images.
Secondly, inputting the plurality of OCT images into a feature extraction network to obtain a plurality of image features corresponding to the plurality of OCT images, wherein the feature extraction network is of a residual error network structure, and the convolution layer in the feature extraction network is used for extracting the multi-scale features in each OCT image.
And then inputting the time sequence of the image characteristics corresponding to the plurality of OCT images into a long-short term memory artificial neural network to obtain the characteristics corresponding to the plurality of OCT images, wherein the long-short term memory artificial neural network is used for fusing the spatial information of the plurality of OCT images.
And finally, predicting the anti-VEGF curative effect according to the corresponding characteristics of the multiple OCT images to obtain an anti-VEGF curative effect prediction result, wherein the anti-VEGF curative effect prediction result comprises vision improvement or vision deterioration. Specifically, the prediction result of the anti-VEGF curative effect can provide effective reference for the treatment scheme of a doctor. When the anti-VEGF curative effect is predicted, the anti-VEGF curative effect prediction probability is determined according to the corresponding characteristics of a plurality of OCT images. And when the anti-VEGF curative effect prediction probability is smaller than a preset probability threshold value, determining that the anti-VEGF curative effect prediction result is vision deterioration, and indicating that anti-VEGF injection is not recommended. And when the anti-VEGF curative effect prediction probability is larger than a preset probability threshold value, determining that the anti-VEGF curative effect prediction result is vision improvement, and indicating that anti-VEGF injection is recommended. And when the anti-VEGF curative effect prediction probability is equal to a preset probability threshold value, determining that the anti-VEGF curative effect prediction result is deteriorated vision or improved vision, and specifically determining whether the deteriorated vision or the improved vision is required.
Referring to fig. 2, fig. 2 is a schematic flow chart of a method for predicting the therapeutic effect of VEGF according to an embodiment of the present invention. As shown in fig. 2, a method for predicting an anti-VEGF therapeutic effect provided in an embodiment of the present application may include the following steps.
201. Acquiring a plurality of Optical Coherence Tomography (OCT) images acquired aiming at a macular region.
Optionally, a plurality of initial OCT images obtained by scanning the eye of the patient with the OCT may be acquired, and the region acquired for the macular region in the plurality of initial OCT images may be extracted to obtain a plurality of OCT images acquired for the macular region.
Specifically, when predicting the anti-VEGF therapeutic effect of a patient, firstly, the OCT apparatus is used to scan the eye of the patient, and a plurality of initial OCT images can be obtained, wherein each OCT image in the plurality of initial OCT images includes the retinal tissue information of the patient. And after the plurality of initial OCT images are obtained, extracting image parts scanned aiming at a macular region in the plurality of initial OCT images, wherein the macular region is positioned in the center of the retina, and the macular region is the most concentrated part of the central visual cells of the retina of the human eye. By extracting the scanned area for the macular area, the peripheral area without effective information can be removed, thereby saving the calculation time in the subsequent calculation process.
Optionally, after acquiring a plurality of OCT images collected for the macular region, the method may further perform preprocessing on the plurality of OCT images, which specifically includes: and (4) carrying out image correction on each OCT image, and carrying out contrast enhancement processing after correction. Wherein the image correction comprises an image tilt correction and/or an image brightness correction.
Specifically, when the eyes of the patient are scanned, due to the influence of external factors, such as light, angle change of the eyes of the patient when the eyes of the patient are scanned, etc., too bright or too dark partial images may appear in the obtained multiple OCT images, or image tilt may appear, which is not favorable for further processing the OCT images in the following. Therefore, before feature extraction is performed on the plurality of OCT images, preprocessing is performed on the plurality of OCT images, including image correction processing and contrast enhancement processing, so as to correct an image that is too bright or too dark and correct an image that is tilted, and at the same time, to improve the contrast of the image, so as to improve the visual effect of the image.
In one possible embodiment, the plurality of OCT images may be twelve line OCT images, i.e., 12 OCT images.
202. And performing feature extraction on the plurality of OCT images to obtain a plurality of image features corresponding to the plurality of OCT images, wherein the image feature corresponding to each OCT image in the plurality of OCT images comprises a multi-scale feature in each OCT image.
Optionally, the plurality of OCT images are input to a feature extraction network, so as to obtain a plurality of image features corresponding to the plurality of OCT images. Referring to fig. 3, fig. 3 is a schematic diagram of a feature extraction network according to an embodiment of the present application. As shown in fig. 3, the feature extraction network is a residual network structure in which an input is directly transmitted to an output as an initial result by way of shortcut connection. Wherein the convolution layer in the feature extraction network is used for extracting the multi-scale features in each OCT image.
Specifically, before determining the prediction result of the anti-VEGF therapeutic effect, a plurality of image features corresponding to the plurality of OCT images are extracted through a feature extraction network. In the characteristic extraction process, the plurality of OCT images are input into a characteristic extraction network, the characteristic extraction network is of a residual error network structure, the residual error network is one of convolutional neural networks, optimization is easy, accuracy can be improved by increasing depth, residual error blocks in the residual error network are connected in a jumping mode, and the gradient disappearance problem caused by increasing depth in a deep neural network is relieved. The convolution layer in the feature extraction network is used for extracting multi-scale features in each OCT image, wherein the features of different scales reflect different image features, the features of lighter scales reflect image features of lighter levels such as edges and the like, and the features of deeper scales reflect image features of deeper levels such as object outlines and the like.
The characteristic extraction network with the combination of the residual error network structure and the multi-scale characteristic extraction function is used for extracting the characteristics of the plurality of OCT images, on one hand, the characteristic extraction network with the residual error network structure is used for extracting the characteristics of the plurality of OCT images, the calculation cost can be reduced, and the problem of gradient disappearance caused by depth increase in the deep neural network can be solved. On the other hand, the characteristic extraction network with the multi-scale characteristic extraction function is used for extracting the characteristics of a plurality of OCT images, so that the multi-scale characteristics in the OCT images can be effectively obtained, the richness and the comprehensiveness of the extracted characteristics are increased, and the prediction result is more accurate.
203. And carrying out spatial information fusion on a plurality of image characteristics corresponding to the plurality of OCT images to obtain the characteristics corresponding to the plurality of OCT images.
Optionally, acquiring a time corresponding to each of the plurality of OCT images; determining a time sequence of a plurality of image characteristics corresponding to the plurality of OCT images according to the time corresponding to each OCT image; and inputting the time sequence of the image characteristics corresponding to the plurality of OCT images into a long-short term memory artificial neural network to obtain the characteristics corresponding to the plurality of OCT images.
Specifically, after a plurality of image features corresponding to a plurality of OCT images are obtained through a feature extraction network, spatial information fusion is performed on the plurality of image features, so that spatial information of the features is enriched. When spatial information fusion is performed, firstly, the time corresponding to each OCT image is acquired, and the image features corresponding to each OCT image are sequenced according to the time corresponding to each OCT image and the time sequence, so that the time sequence of the image features corresponding to the plurality of OCT images is obtained. And finally, inputting the time series of the image characteristics corresponding to the plurality of OCT images into a long-short-term memory artificial neural network (LSTM) to obtain the characteristics corresponding to the plurality of OCT images. The long-short term memory artificial neural network is a time cycle neural network and comprises a forgetting gate, an input gate, an output gate and the like. The image features extracted from the spatial sequence are input into the long-short term memory artificial neural network in a time sequence mode, so that a plurality of image features corresponding to a plurality of OCT images of the same patient are subjected to spatial information fusion, the spatial information of the features is enriched, the prediction mode of the network is closer to the process of reading the images by a doctor, and the prediction result is more accurate.
204. And determining the prediction result of the anti-VEGF curative effect according to the corresponding characteristics of the plurality of OCT images.
Optionally, according to the corresponding features of the plurality of OCT images, performing two classifications on the anti-VEGF therapeutic effect to obtain the anti-VEGF therapeutic effect prediction result, where the anti-VEGF therapeutic effect prediction result includes visual improvement or visual deterioration.
Specifically, the prediction result of the anti-VEGF curative effect can provide effective reference for the treatment scheme of a doctor. When the anti-VEGF curative effect is predicted, the anti-VEGF curative effect prediction probability is determined according to the corresponding characteristics of a plurality of OCT images. And when the anti-VEGF curative effect prediction probability is smaller than a preset probability threshold value, determining that the anti-VEGF curative effect prediction result is vision deterioration, and indicating that anti-VEGF injection is not recommended. And when the anti-VEGF curative effect prediction probability is larger than a preset probability threshold value, determining that the anti-VEGF curative effect prediction result is vision improvement, and indicating that anti-VEGF injection is recommended. And when the anti-VEGF curative effect prediction probability is equal to a preset probability threshold value, determining that the anti-VEGF curative effect prediction result is deteriorated vision or improved vision, and specifically determining whether the deteriorated vision or the improved vision is required.
It can be seen that, by the prediction method for the therapeutic effect of the VEGF for resisting vascular endothelial growth factor, provided by the embodiment of the application, after a plurality of OCT images are acquired, segmentation is not required, the labeling cost in a segmentation network is reduced, and meanwhile, the accuracy error of prediction of the VEGF resistant therapeutic effect caused by inaccurate segmentation and the like is avoided. In addition, the extracted image characteristics corresponding to each OCT image comprise multi-scale characteristics, and the richness and comprehensiveness of the extracted characteristics are increased, so that the accuracy of the prediction of the anti-VEGF curative effect is improved. In addition, spatial information fusion is carried out on the characteristics of a plurality of images, the spatial characteristics of a plurality of OCT images are effectively utilized, the spatial information of the characteristics is enriched, and the accuracy of prediction of the anti-VEGF curative effect is improved.
In one embodiment of the present application, the solution of the present application can also be applied to the field of smart medicine. For example, a plurality of OCT images input by a doctor are received, and the prediction result of the anti-VEGF therapeutic effect is determined by the anti-VEGF therapeutic effect prediction method provided by the present application. Due to the adoption of the anti-VEGF curative effect prediction method provided by the application, the anti-VEGF curative effect prediction result can be determined more accurately, so that a more accurate judgment basis can be provided for a treatment scheme of a doctor, and the treatment efficiency and accuracy of the doctor are improved.
Referring to fig. 4, fig. 4 is a schematic flow chart of another method for predicting the therapeutic effect of VEGF according to the embodiment of the present application. As shown in fig. 4, another method for predicting the therapeutic effect of VEGF provided in the embodiment of the present invention may include the following steps.
401. Acquiring a plurality of Optical Coherence Tomography (OCT) images acquired aiming at a macular region.
Optionally, a plurality of initial OCT images obtained by scanning the eye of the patient with the OCT may be acquired, and the region acquired for the macular region in the plurality of initial OCT images may be extracted to obtain a plurality of OCT images acquired for the macular region.
Specifically, when predicting the anti-VEGF therapeutic effect of a patient, firstly, the OCT apparatus is used to scan the eye of the patient, and a plurality of initial OCT images can be obtained, wherein each OCT image in the plurality of initial OCT images includes the retinal tissue information of the patient. And after the plurality of initial OCT images are obtained, extracting image parts scanned aiming at a macular region in the plurality of initial OCT images, wherein the macular region is positioned in the center of the retina, and the macular region is the most concentrated part of the central visual cells of the retina of the human eye. By extracting the scanned area for the macular area, the peripheral area without effective information can be removed, thereby saving the calculation time in the subsequent calculation process.
In one possible embodiment, the plurality of OCT images may be twelve line OCT images, i.e., 12 OCT images.
402. The plurality of OCT images are preprocessed.
Preprocessing the plurality of OCT images includes: and (4) carrying out image correction on each OCT image, and carrying out contrast enhancement processing after correction. Wherein the image correction comprises an image tilt correction and/or an image brightness correction.
Specifically, when the eyes of the patient are scanned, due to the influence of external factors, such as light, angle change of the eyes of the patient when the eyes of the patient are scanned, etc., too bright or too dark partial images may appear in the obtained multiple OCT images, or image tilt may appear, which is not favorable for further processing the OCT images in the following. Therefore, before feature extraction is performed on the plurality of OCT images, preprocessing is performed on the plurality of OCT images, including image correction processing and contrast enhancement processing, so as to correct an image that is too bright or too dark and correct an image that is tilted, and at the same time, to improve the contrast of the image, so as to improve the visual effect of the image.
In one possible embodiment, the method for performing image brightness correction on the plurality of OCT images may be performing Gamma transformation on the plurality of OCT images, wherein the formula for performing Gamma transformation on each OCT image is
Where V represents R, G, B three channels for each OCT image, respectively.
In one possible embodiment, the formula for performing contrast enhancement processing on the plurality of OCT images is
. Where V represents R, G, B three channels for each OCT image, respectively. That is, for R, G, B channel data of each OCT image, it is subtracted by the minimum value of all data of the channel, then divided by the maximum value of all data of the channel, and finally multiplied by 255 to restore it to [0, 255%]The operation of the value range of (1). Therefore, R, G, B three-channel data of each OCT image are distributed more uniformly on the range of 0-255, the contrast of the image is improved, and the aims of improving the subjective visual effect of the image and enhancing the details of the image are fulfilled.
403. And inputting the plurality of OCT images into a feature extraction network to obtain a plurality of image features corresponding to the plurality of OCT images.
Specifically, the feature extraction network is a residual error network structure, and the convolution layer in the feature extraction network is used for extracting the multi-scale features in each OCT image.
Before determining the prediction result of the anti-VEGF curative effect, firstly extracting a plurality of image features corresponding to the plurality of OCT images through a feature extraction network. In the characteristic extraction process, the plurality of OCT images are input into a characteristic extraction network, the characteristic extraction network is of a residual error network structure, the residual error network is one of convolutional neural networks, optimization is easy, accuracy can be improved by increasing depth, residual error blocks in the residual error network are connected in a jumping mode, and the gradient disappearance problem caused by increasing depth in a deep neural network is relieved. The convolution layer in the feature extraction network is used for extracting multi-scale features in each OCT image, wherein the features of different scales reflect different image features, the features of lighter scales reflect image features of lighter levels such as edges and the like, and the features of deeper scales reflect image features of deeper levels such as object outlines and the like.
In one possible embodiment, the plurality of OCT images are twelve line OCT images (i.e., 12 OCT images). Inputting 12 OCT images into a feature extraction network to obtain 12 image features corresponding to the 12 OCT images.
The characteristic extraction network with the combination of the residual error network structure and the multi-scale characteristic extraction function is used for extracting the characteristics of the plurality of OCT images, on one hand, the characteristic extraction network with the residual error network structure is used for extracting the characteristics of the plurality of OCT images, the calculation cost can be reduced, and the problem of gradient disappearance caused by depth increase in the deep neural network can be solved. On the other hand, the characteristic extraction network with the multi-scale characteristic extraction function is used for extracting the characteristics of a plurality of OCT images, so that the multi-scale characteristics in the OCT images can be effectively obtained, the richness and the comprehensiveness of the extracted characteristics are increased, and the prediction result is more accurate.
404. And determining a time sequence of a plurality of image characteristics corresponding to the plurality of OCT images.
Specifically, when spatial information fusion is performed, time corresponding to each OCT image is first acquired, and image features corresponding to each OCT image are sequenced according to the time corresponding to each OCT image and the time sequence, so that a time sequence of a plurality of image features corresponding to the plurality of OCT images is obtained.
405. And inputting the time series of the image characteristics corresponding to the plurality of OCT images into the long-short term memory artificial neural network to obtain the characteristics corresponding to the plurality of OCT images.
Specifically, after a plurality of image features corresponding to a plurality of OCT images are obtained through a feature extraction network, spatial information fusion is performed on the plurality of image features, so that spatial information of the features is enriched. After obtaining the time series of the image features corresponding to the plurality of OCT images, the time series of the image features corresponding to the plurality of OCT images is input to a long-short-term-memory artificial neural network (LSTM) to obtain the features corresponding to the plurality of OCT images. The long-short term memory artificial neural network is a time cycle neural network and comprises a forgetting gate, an input gate, an output gate and the like. The image features extracted from the spatial sequence are input into the long-short term memory artificial neural network in a time sequence mode, so that a plurality of image features corresponding to a plurality of OCT images of the same patient are subjected to spatial information fusion, the spatial information of the features is enriched, the prediction mode of the network is closer to the process of reading the images by a doctor, and the prediction result is more accurate.
406. And determining the prediction probability of the anti-VEGF curative effect according to the corresponding characteristics of the plurality of OCT images.
Specifically, the anti-VEGF therapeutic effect prediction probability is used to determine an anti-VEGF therapeutic effect prediction result. The anti-VEGF curative effect prediction result comprises vision improvement or vision deterioration, and the anti-VEGF curative effect prediction result can provide effective reference for a treatment scheme of a doctor.
407. And judging whether the anti-VEGF curative effect prediction probability is smaller than a preset probability threshold value.
408. And when the anti-VEGF curative effect prediction probability is smaller than a preset probability threshold value, determining that the anti-VEGF curative effect prediction result is vision deterioration.
When the prediction result of the anti-VEGF curative effect is visual deterioration, the anti-VEGF injection is not recommended.
409. And when the anti-VEGF curative effect prediction probability is not smaller than a preset probability threshold, determining that the anti-VEGF curative effect prediction result is vision improvement.
When the prediction result of the anti-VEGF curative effect is vision improvement, the anti-VEGF injection is recommended.
It can be seen that, according to the VEGF-resistant efficacy prediction method provided by the embodiment of the present application, the feature extraction network combining the residual network structure and the multi-scale feature extraction function is used to perform feature extraction on a plurality of OCT images, and on one hand, the feature extraction network of the residual network structure is used to perform feature extraction on a plurality of OCT images, so that the calculation cost can be reduced, and the problem of gradient disappearance caused by depth increase in the deep neural network can be alleviated. On the other hand, the characteristic extraction network with the multi-scale characteristic extraction function is used for extracting the characteristics of a plurality of OCT images, so that the multi-scale characteristics in the OCT images can be effectively obtained, the richness and the comprehensiveness of the extracted characteristics are increased, and the prediction result is more accurate. The image features extracted from the spatial sequence are input into the long-short term memory artificial neural network in a time sequence mode, so that a plurality of image features corresponding to a plurality of OCT images of the same patient are subjected to spatial information fusion, the spatial information of the features is enriched, the prediction mode of the network is closer to the process of reading the images by a doctor, and the prediction result is more accurate.
Referring to fig. 5, fig. 5 is a schematic diagram of an apparatus for predicting VEGF therapeutic effect according to an embodiment of the present disclosure. As shown in fig. 5, an apparatus for predicting VEGF therapeutic effect provided by an embodiment of the present invention may include the following modules.
An acquiring module 501, configured to acquire multiple OCT images of the optical coherence tomography acquired for the macular region.
A feature extraction module 502, configured to perform feature extraction on the multiple OCT images to obtain multiple image features corresponding to the multiple OCT images, where an image feature corresponding to each OCT image in the multiple OCT images includes a multi-scale feature in each OCT image.
The spatial information fusion module 503 is configured to perform spatial information fusion on the multiple image features corresponding to the multiple OCT images to obtain features corresponding to the multiple OCT images.
A determining module 504, configured to determine a prediction result of an anti-VEGF therapeutic effect according to features corresponding to the multiple OCT images.
In a possible implementation manner, the obtaining module 501 is specifically configured to: acquiring a plurality of initial OCT images obtained by scanning eyes through OCT (optical coherence tomography); extracting regions collected for a macular region in the plurality of initial OCT images to obtain the plurality of OCT images collected for the macular region.
In a possible implementation, the apparatus further includes a processing module configured to: performing image correction processing on the plurality of OCT images to obtain the corrected plurality of OCT images, wherein the image correction processing comprises image inclination correction and/or image brightness correction; and carrying out contrast enhancement processing on the corrected multiple OCT images to obtain the multiple OCT images with enhanced contrast.
In a possible implementation, the feature extraction module 502 is specifically configured to: inputting the plurality of OCT images into a feature extraction network to obtain a plurality of image features corresponding to the plurality of OCT images, wherein the feature extraction network is of a residual error network structure, and a convolution layer in the feature extraction network is used for extracting the multi-scale features in each OCT image.
In a possible implementation manner, the spatial information fusion module 503 is specifically configured to: acquiring time corresponding to each OCT image in the plurality of OCT images; determining a time sequence of a plurality of image characteristics corresponding to the plurality of OCT images according to the time corresponding to each OCT image; and inputting the time sequence of the image characteristics corresponding to the plurality of OCT images into a long-short term memory artificial neural network to obtain the characteristics corresponding to the plurality of OCT images.
In a possible implementation manner, the determining module 504 is specifically configured to: and performing two classifications on the anti-VEGF curative effect according to the corresponding characteristics of the plurality of OCT images to obtain the anti-VEGF curative effect prediction result, wherein the anti-VEGF curative effect prediction result comprises vision improvement or vision deterioration.
In a possible implementation manner, the determining module 504 is specifically configured to: determining the prediction probability of the anti-VEGF curative effect according to the corresponding characteristics of the plurality of OCT images; when the anti-VEGF curative effect prediction probability is smaller than a preset probability threshold value, determining that the anti-VEGF curative effect prediction result is deteriorated vision; when the anti-VEGF curative effect prediction probability is larger than the preset probability threshold, determining that the anti-VEGF curative effect prediction result is vision improvement; and when the anti-VEGF curative effect prediction probability is equal to the preset probability threshold value, determining that the anti-VEGF curative effect prediction result is vision deterioration or vision improvement.
For specific implementation of the device for predicting the therapeutic effect of VEGF, reference may be made to the above embodiments of the method for predicting the therapeutic effect of VEGF, which are not described herein again.
Referring to fig. 6, fig. 6 is a schematic structural diagram of an electronic device in a hardware operating environment according to an embodiment of the present application. As shown in fig. 6, an electronic device of a hardware operating environment according to an embodiment of the present application may include: a processor 601, such as a CPU. The memory 602 may alternatively be a high speed RAM memory or a stable memory such as a disk memory. A communication interface 603 for implementing connection communication between the processor 601 and the memory 602.
Those skilled in the art will appreciate that the configuration of the electronic device shown in fig. 6 does not constitute a limitation of the electronic device and may include more or fewer components than those shown, or some components may be combined, or a different arrangement of components.
As shown in fig. 6, the memory 602 may include an operating system, a network communication module, and an anti-VEGF therapeutic prediction program. The operating system is a program for managing and controlling hardware and software resources of the electronic equipment and supports the running of the anti-vascular endothelial growth factor VEGF curative effect prediction program and other software or programs. The network communication module is used for communication among the components in the memory 602, and with other hardware and software in the electronic device.
In the electronic device shown in fig. 6, the processor 601 is configured to execute the anti-VEGF therapy prediction program stored in the memory 602, and implement the following steps: acquiring a plurality of Optical Coherence Tomography (OCT) images collected aiming at a macular region; performing feature extraction on the plurality of OCT images to obtain a plurality of image features corresponding to the plurality of OCT images, wherein the image feature corresponding to each OCT image in the plurality of OCT images comprises a multi-scale feature in each OCT image; performing spatial information fusion on a plurality of image characteristics corresponding to the plurality of OCT images to obtain characteristics corresponding to the plurality of OCT images; and determining the prediction result of the anti-VEGF curative effect according to the corresponding characteristics of the plurality of OCT images.
For specific implementation of the electronic device in the embodiment of the present application, reference may be made to the above embodiments of the method for predicting the therapeutic effect of VEGF, which are not described herein again.
Another embodiment of the present application provides a computer-readable storage medium storing a computer program for execution by a processor to perform the steps of: acquiring a plurality of Optical Coherence Tomography (OCT) images collected aiming at a macular region; performing feature extraction on the plurality of OCT images to obtain a plurality of image features corresponding to the plurality of OCT images, wherein the image feature corresponding to each OCT image in the plurality of OCT images comprises a multi-scale feature in each OCT image; performing spatial information fusion on a plurality of image characteristics corresponding to the plurality of OCT images to obtain characteristics corresponding to the plurality of OCT images; and determining the prediction result of the anti-VEGF curative effect according to the corresponding characteristics of the plurality of OCT images.
For specific implementation of the computer-readable storage medium in the embodiments of the present application, reference may be made to the above embodiments of the method for predicting therapeutic effect of VEGF, which are not described herein again.
It is also noted that while for simplicity of explanation, the foregoing method embodiments have been described as a series of acts or combination of acts, it will be appreciated by those skilled in the art that the present application is not limited by the order of acts, as some steps may, in accordance with the present application, occur in other orders and concurrently. Further, those skilled in the art should also appreciate that the embodiments described in the specification are preferred embodiments and that the acts and modules referred to are not necessarily required in this application. In the foregoing embodiments, the descriptions of the respective embodiments have respective emphasis, and for parts that are not described in detail in a certain embodiment, reference may be made to related descriptions of other embodiments.
The above embodiments are only used for illustrating the technical solutions of the present application, and not for limiting the same; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and the modifications or the substitutions do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the embodiments of the present application.
Claims (7)
1. An anti-VEGF therapeutic effect prediction device, comprising:
the acquisition module is used for acquiring a plurality of Optical Coherence Tomography (OCT) images acquired aiming at a macular region, and is specifically used for: acquiring a plurality of initial OCT images obtained by scanning eyes through OCT (optical coherence tomography); extracting areas collected for a macular region in the plurality of initial OCT images to obtain a plurality of OCT images collected for the macular region;
a feature extraction module, configured to perform feature extraction on the multiple OCT images to obtain multiple image features corresponding to the multiple OCT images, where an image feature corresponding to each OCT image in the multiple OCT images includes a multi-scale feature in each OCT image, and specifically configured to: inputting the plurality of OCT images into a feature extraction network to obtain a plurality of image features corresponding to the plurality of OCT images, wherein the feature extraction network is of a residual error network structure, a convolution layer in the feature extraction network is used for extracting multi-scale features in each OCT image, the multi-scale features in each OCT image comprise a light image feature and a deep image feature of the macular region, the light image feature is edge information of the macular region in each OCT image, and the deep image feature is profile information of the macular region in each OCT image;
the spatial information fusion module is used for carrying out spatial information fusion on a plurality of image characteristics corresponding to the plurality of OCT images to obtain characteristics corresponding to the plurality of OCT images;
and the determining module is used for determining the prediction result of the anti-VEGF curative effect according to the corresponding characteristics of the plurality of OCT images.
2. The apparatus of claim 1, further comprising a processing module to:
performing image correction processing on the plurality of OCT images to obtain the corrected plurality of OCT images, wherein the image correction processing comprises image inclination correction and/or image brightness correction;
and carrying out contrast enhancement processing on the corrected multiple OCT images to obtain the multiple OCT images with enhanced contrast.
3. The apparatus according to claim 1 or 2, wherein the spatial information fusion module is specifically configured to:
acquiring time corresponding to each OCT image in the plurality of OCT images;
determining a time sequence of a plurality of image characteristics corresponding to the plurality of OCT images according to the time corresponding to each OCT image;
and inputting the time sequence of the image characteristics corresponding to the plurality of OCT images into a long-short term memory artificial neural network to obtain the characteristics corresponding to the plurality of OCT images.
4. The apparatus of claims to 3, wherein the determining module is specifically configured to:
and performing two classifications on the anti-VEGF curative effect according to the corresponding characteristics of the plurality of OCT images to obtain the anti-VEGF curative effect prediction result, wherein the anti-VEGF curative effect prediction result comprises vision improvement or vision deterioration.
5. The apparatus of claim 4, wherein the determining module is specifically configured to:
determining the prediction probability of the anti-VEGF curative effect according to the corresponding characteristics of the plurality of OCT images;
when the anti-VEGF curative effect prediction probability is smaller than a preset probability threshold value, determining that the anti-VEGF curative effect prediction result is deteriorated vision;
when the anti-VEGF curative effect prediction probability is larger than the preset probability threshold, determining that the anti-VEGF curative effect prediction result is vision improvement;
and when the anti-VEGF curative effect prediction probability is equal to the preset probability threshold value, determining that the anti-VEGF curative effect prediction result is vision deterioration or vision improvement.
6. An electronic device comprising a processor, a memory, a communication interface, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the processor to perform the steps of:
acquiring a plurality of Optical Coherence Tomography (OCT) images collected aiming at a macular region, specifically comprising: acquiring a plurality of initial OCT images obtained by scanning eyes through OCT (optical coherence tomography); extracting areas collected for a macular region in the plurality of initial OCT images to obtain a plurality of OCT images collected for the macular region;
performing feature extraction on the plurality of OCT images to obtain a plurality of image features corresponding to the plurality of OCT images, where the image feature corresponding to each OCT image in the plurality of OCT images includes a multi-scale feature in each OCT image, and specifically includes: inputting the plurality of OCT images into a feature extraction network to obtain a plurality of image features corresponding to the plurality of OCT images, wherein the feature extraction network is of a residual error network structure, a convolution layer in the feature extraction network is used for extracting multi-scale features in each OCT image, the multi-scale features in each OCT image comprise a light image feature and a deep image feature of the macular region, the light image feature is edge information of the macular region in each OCT image, and the deep image feature is profile information of the macular region in each OCT image;
performing spatial information fusion on a plurality of image characteristics corresponding to the plurality of OCT images to obtain characteristics corresponding to the plurality of OCT images;
and determining the prediction result of the anti-VEGF curative effect according to the corresponding characteristics of the plurality of OCT images.
7. A computer-readable storage medium, characterized in that the computer-readable storage medium stores a computer program, which is executed by a processor to implement the steps of:
acquiring a plurality of Optical Coherence Tomography (OCT) images collected aiming at a macular region, specifically comprising: acquiring a plurality of initial OCT images obtained by scanning eyes through OCT (optical coherence tomography); extracting areas collected for a macular region in the plurality of initial OCT images to obtain a plurality of OCT images collected for the macular region;
performing feature extraction on the plurality of OCT images to obtain a plurality of image features corresponding to the plurality of OCT images, where the image feature corresponding to each OCT image in the plurality of OCT images includes a multi-scale feature in each OCT image, and specifically includes: inputting the plurality of OCT images into a feature extraction network to obtain a plurality of image features corresponding to the plurality of OCT images, wherein the feature extraction network is of a residual error network structure, a convolution layer in the feature extraction network is used for extracting multi-scale features in each OCT image, the multi-scale features in each OCT image comprise a light image feature and a deep image feature of the macular region, the light image feature is edge information of the macular region in each OCT image, and the deep image feature is profile information of the macular region in each OCT image;
performing spatial information fusion on a plurality of image characteristics corresponding to the plurality of OCT images to obtain characteristics corresponding to the plurality of OCT images;
and determining the prediction result of the anti-VEGF curative effect according to the corresponding characteristics of the plurality of OCT images.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011061099.8A CN111920375B (en) | 2020-09-30 | 2020-09-30 | Vascular Endothelial Growth Factor (VEGF) resistance curative effect prediction device and method |
| PCT/CN2020/132473 WO2021139446A1 (en) | 2020-09-30 | 2020-11-27 | Anti-vascular endothelial growth factor (vegf) curative effect prediction apparatus and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011061099.8A CN111920375B (en) | 2020-09-30 | 2020-09-30 | Vascular Endothelial Growth Factor (VEGF) resistance curative effect prediction device and method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111920375A CN111920375A (en) | 2020-11-13 |
| CN111920375B true CN111920375B (en) | 2021-01-15 |
Family
ID=73334769
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011061099.8A Active CN111920375B (en) | 2020-09-30 | 2020-09-30 | Vascular Endothelial Growth Factor (VEGF) resistance curative effect prediction device and method |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN111920375B (en) |
| WO (1) | WO2021139446A1 (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111920375B (en) * | 2020-09-30 | 2021-01-15 | 平安科技（深圳）有限公司 | Vascular Endothelial Growth Factor (VEGF) resistance curative effect prediction device and method |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2010148298A1 (en) * | 2009-06-18 | 2010-12-23 | University Of Miami | System and method for molecular in vivo imaging and theranostics |
| CN108229580A (en) * | 2018-01-26 | 2018-06-29 | 浙江大学 | Sugared net ranking of features device in a kind of eyeground figure based on attention mechanism and Fusion Features |
| CN109376767A (en) * | 2018-09-20 | 2019-02-22 | 中国科学技术大学 | Retina OCT image classification method based on deep learning |
| CN109727259A (en) * | 2019-01-07 | 2019-05-07 | 哈尔滨理工大学 | A kind of retinal images partitioning algorithm based on residual error U-NET network |
| CN109858429A (en) * | 2019-01-28 | 2019-06-07 | 北京航空航天大学 | A kind of identification of eye fundus image lesion degree and visualization system based on convolutional neural networks |
| CN110147715A (en) * | 2019-04-01 | 2019-08-20 | 江西比格威医疗科技有限公司 | A kind of retina OCT image Bruch film angle of release automatic testing method |
| CN110472530A (en) * | 2019-07-29 | 2019-11-19 | 中山大学 | Retina OCT image classification method based on wavelet transformation and transfer learning |
| CN111184508A (en) * | 2020-01-19 | 2020-05-22 | 武汉大学 | Electrocardiosignal detection device and analysis method based on joint neural network |
| CN111329445A (en) * | 2020-02-20 | 2020-06-26 | 广东工业大学 | Atrial fibrillation identification method based on group convolution residual error network and long-term and short-term memory network |
| CN111611877A (en) * | 2020-04-30 | 2020-09-01 | 杭州电子科技大学 | Age interference resistant face recognition method based on multi-temporal-spatial information fusion |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CA2950102A1 (en) * | 2014-04-28 | 2015-11-05 | Northwestern University | Devices, methods, and systems of functional optical coherence tomography |
| US10943348B2 (en) * | 2018-07-18 | 2021-03-09 | Case Western Reserve University | Predicting response to anti-vascular endothelial growth factor therapy with computer-extracted morphology and spatial arrangement features of leakage patterns on baseline fluorescein angiography in diabetic macular edema |
| CN110309862A (en) * | 2019-06-11 | 2019-10-08 | 广东省人民医院(广东省医学科学院) | DME prognosis information forecasting system and its application method based on ensemble machine learning |
| CN111920375B (en) * | 2020-09-30 | 2021-01-15 | 平安科技（深圳）有限公司 | Vascular Endothelial Growth Factor (VEGF) resistance curative effect prediction device and method |
-
2020
- 2020-09-30 CN CN202011061099.8A patent/CN111920375B/en active Active
- 2020-11-27 WO PCT/CN2020/132473 patent/WO2021139446A1/en active Application Filing
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2010148298A1 (en) * | 2009-06-18 | 2010-12-23 | University Of Miami | System and method for molecular in vivo imaging and theranostics |
| CN108229580A (en) * | 2018-01-26 | 2018-06-29 | 浙江大学 | Sugared net ranking of features device in a kind of eyeground figure based on attention mechanism and Fusion Features |
| CN109376767A (en) * | 2018-09-20 | 2019-02-22 | 中国科学技术大学 | Retina OCT image classification method based on deep learning |
| CN109727259A (en) * | 2019-01-07 | 2019-05-07 | 哈尔滨理工大学 | A kind of retinal images partitioning algorithm based on residual error U-NET network |
| CN109858429A (en) * | 2019-01-28 | 2019-06-07 | 北京航空航天大学 | A kind of identification of eye fundus image lesion degree and visualization system based on convolutional neural networks |
| CN110147715A (en) * | 2019-04-01 | 2019-08-20 | 江西比格威医疗科技有限公司 | A kind of retina OCT image Bruch film angle of release automatic testing method |
| CN110472530A (en) * | 2019-07-29 | 2019-11-19 | 中山大学 | Retina OCT image classification method based on wavelet transformation and transfer learning |
| CN111184508A (en) * | 2020-01-19 | 2020-05-22 | 武汉大学 | Electrocardiosignal detection device and analysis method based on joint neural network |
| CN111329445A (en) * | 2020-02-20 | 2020-06-26 | 广东工业大学 | Atrial fibrillation identification method based on group convolution residual error network and long-term and short-term memory network |
| CN111611877A (en) * | 2020-04-30 | 2020-09-01 | 杭州电子科技大学 | Age interference resistant face recognition method based on multi-temporal-spatial information fusion |
Non-Patent Citations (1)
| Title |
|---|
| 光学相干断层扫描血管成像观察抗血管内皮生长因子治疗渗出性老年性黄斑变性的临床疗效;苏强等;《中国药物与临床》;20200831;第20卷(第16期);第2683-2686页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111920375A (en) | 2020-11-13 |
| WO2021139446A1 (en) | 2021-07-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20220076420A1 (en) | Retinopathy recognition system | |
| Araújo et al. | DR| GRADUATE: Uncertainty-aware deep learning-based diabetic retinopathy grading in eye fundus images | |
| Hogarty et al. | Current state and future prospects of artificial intelligence in ophthalmology: a review | |
| Tang et al. | Detection of diabetic retinopathy from ultra-widefield scanning laser ophthalmoscope images: a multicenter deep learning analysis | |
| CN110544274B (en) | Multispectral-based fundus image registration method and system | |
| Han et al. | Automated AI labeling of optic nerve head enables insights into cross-ancestry glaucoma risk and genetic discovery in> 280,000 images from UKB and CLSA | |
| US20210391056A1 (en) | Health big data service method and system based on remote fundus screening | |
| Keenan et al. | DeepLensNet: deep learning automated diagnosis and quantitative classification of cataract type and severity | |
| US20210225048A1 (en) | Systems and methods for reconstructing cardiac images | |
| CN111920375B (en) | Vascular Endothelial Growth Factor (VEGF) resistance curative effect prediction device and method | |
| CN110415245A (en) | Optical data determines method, model training method and equipment | |
| Tahvildari et al. | Application of artificial intelligence in the diagnosis and management of corneal diseases | |
| Wu et al. | Deep learning-based risk classification and auxiliary diagnosis of macular edema | |
| Dong et al. | Deep learning-based estimation of axial length and subfoveal choroidal thickness from color fundus photographs | |
| Liu et al. | A new ultra-wide-field fundus dataset to diabetic retinopathy grading using hybrid preprocessing methods | |
| CN113658097B (en) | Training method and device for fundus image quality enhancement model | |
| CN107203758A (en) | Diabetes patient's retinal vascular images dividing method | |
| CN106446805A (en) | Segmentation method and system for optic cup in eye ground photo | |
| Giancardo | Automated fundus images analysis techniques to screen retinal diseases in diabetic patients | |
| CN115908274A (en) | Device, equipment and medium for detecting focus | |
| Taş et al. | Detection of retinal diseases from ophthalmological images based on convolutional neural network architecture. | |
| CN112734701A (en) | Fundus focus detection method, fundus focus detection device and terminal equipment | |
| Hassan et al. | Identifying the level of diabetic retinopathy using deep convolution neural network | |
| Huang et al. | Development of a deep learning-based tool to assist wound classification | |
| CN111951219A (en) | Thyroid eye disease screening method, system and equipment based on orbit CT image |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |