Diabetes assessment devices and diabetes assessment methods

Abstract

This application provides a diabetes assessment device and a diabetes assessment method, relating to the field of image processing technology. The diabetes assessment device includes: an image determination module configured to determine a fundus image to be assessed, wherein the fundus image is used to characterize the physiological structures of the fundus; and a diabetes assessment module configured to determine a diabetes assessment result based on the fundus image to be assessed. By setting the image determination module, the fundus image to be assessed can be determined; by setting the diabetes assessment module, the diabetes assessment result can be determined based on the fundus image to be assessed. That is, invasive examinations such as blood tests are unnecessary; the diabetes assessment result can be determined solely based on the fundus image to be assessed, without causing harm to the patient's physical or mental health.

Description

Technical Field

[0001] This application relates to the field of image processing technology, specifically to a diabetes assessment device, a diabetes assessment method, a computer-readable storage medium, and an electronic device. Background Technology

[0002] Currently, checking whether a patient has diabetes or what their condition is requires blood tests, which are invasive procedures and can cause physical and psychological harm to the patient. There are currently no non-invasive methods or devices to determine whether a patient has diabetes or what their specific condition is. Summary of the Invention

[0003] In view of this, embodiments of this application provide a diabetes assessment device, a diabetes assessment method, a computer-readable storage medium, and an electronic device, which solve the problem that invasive examinations can easily cause harm to the physical and mental health of patients.

[0004] According to one aspect of this application, an embodiment of this application provides a diabetes assessment device, comprising: an image determination module configured to determine a fundus image to be assessed, wherein the fundus image to be assessed is used to characterize the physiological structure of the fundus of the eyeball; and a diabetes assessment module configured to determine a diabetes assessment result based on the fundus image to be assessed.

[0005] In one embodiment of this application, the diabetes assessment module includes: a suspected lesion area determination module, configured to determine a suspected lesion area in the fundus image to be assessed based on the fundus image to be assessed; a suspected lesion information determination module, configured to determine suspected lesion information in the suspected lesion area based on the suspected lesion area; and an assessment result determination module, configured to determine the diabetes assessment result based on the suspected lesion information.

[0006] In one embodiment of this application, the suspected lesion information determination module includes: a suspected lesion feature information determination unit, configured to determine suspected lesion feature information of the suspected lesion region based on the suspected lesion region, wherein the suspected lesion feature information includes at least one of shape information, color information, topological information, location information, and distribution information of the suspected lesion region; and a suspected lesion information determination unit, configured to determine the suspected lesion information of the suspected lesion region based on the suspected lesion feature information.

[0007] In one embodiment of this application, the evaluation result determination module is further configured to determine whether the subject to be evaluated corresponding to the fundus image to be evaluated has diabetes based on the suspected lesion information.

[0008] In one embodiment of this application, the suspected lesion information includes non-lesion information and / or lesion information, and the lesion information includes at least one of microaneurysm information, hemorrhage information and exudation information.

[0009] In one embodiment of this application, the evaluation result determination module is further configured to determine that the subject to be evaluated corresponding to the fundus image to be evaluated does not have diabetes if all the suspected lesion information is the non-lesion information.

[0010] In one embodiment of this application, the assessment result determination module is further configured to determine the diabetes condition information of the subject to be assessed corresponding to the fundus image to be assessed based on the lesion information if the suspected lesion information includes the lesion information, wherein the diabetes condition information includes risk information of having diabetes and / or severity information of diabetes and / or degree of damage to the body caused by diabetes.

[0011] In one embodiment of this application, the evaluation result determination module includes: a quantization unit configured to quantify the lesion information to obtain a quantification result; and a condition information determination unit configured to determine the diabetes condition information of the subject to be evaluated corresponding to the fundus image to be evaluated based on the quantification result.

[0012] In one embodiment of this application, the quantization unit includes: a lesion feature information determination subunit, configured to determine lesion feature information based on the lesion information, wherein the lesion feature information includes at least one of lesion location information, lesion quantity information, lesion area information, distribution information, and morphological information; and a quantization result determination subunit, configured to determine the quantization result based on the lesion feature information.

[0013] In one embodiment of this application, the diabetes assessment device further includes: a preprocessing module configured to perform preprocessing operations on the fundus image to be assessed to obtain a preprocessed image; wherein, the diabetes assessment module is further configured to determine the diabetes assessment result based on the preprocessed image.

[0014] In one embodiment of this application, the preprocessing module includes at least one of the following units: an image quality assessment unit configured to perform a quality assessment operation on the fundus image to be evaluated to obtain the preprocessed image; a denoising processing unit configured to perform a denoising processing operation on the fundus image to be evaluated to obtain the preprocessed image; a normalization processing unit configured to perform a normalization processing operation on the fundus image to be evaluated to obtain the preprocessed image; and an enhancement processing unit configured to perform an enhancement processing operation on the fundus image to be evaluated to obtain the preprocessed image.

[0015] According to another aspect of this application, one embodiment of this application provides a method for assessing diabetes, comprising: determining a fundus image to be assessed, wherein the fundus image to be assessed is used to characterize the physiological structure of the fundus of the eyeball; and determining a diabetes assessment result based on the fundus image to be assessed.

[0016] According to another aspect of this application, one embodiment of this application provides a computer-readable storage medium storing a computer program for performing the diabetes assessment method described in the above embodiment.

[0017] According to another aspect of this application, one embodiment of this application provides an electronic device, the electronic device comprising: a processor; and a memory for storing processor-executable instructions; the processor being configured to perform the diabetes assessment method described in the above embodiment.

[0018] This application provides a diabetes assessment device, a diabetes assessment method, a computer-readable storage medium, and an electronic device. By setting an image determination module, the device can determine the fundus image to be assessed. By setting a diabetes assessment module, the device can determine the diabetes assessment result based on the fundus image. That is, there is no need for invasive examinations such as blood tests; the diabetes assessment result can be determined solely based on the fundus image, without causing harm to the patient's physical or mental health. Attached Figure Description

[0019] Figure 1 The diagram shown is a scenario applicable to an embodiment of this application.

[0020] Figure 2 The diagram shown is a structural schematic of a diabetes assessment device provided in an exemplary embodiment of this application.

[0021] Figure 3 The image shown is a fundus image to be evaluated provided in an exemplary embodiment of this application.

[0022] Figure 4 The diagram shown is a structural schematic of a diabetes assessment device provided in another exemplary embodiment of this application.

[0023] Figure 5 The diagram shown is a structural schematic of a diabetes assessment device provided in another exemplary embodiment of this application.

[0024] Figure 6 The diagram shown is a structural schematic of a diabetes assessment device provided in another exemplary embodiment of this application.

[0025] Figure 7 The diagram shown is a structural schematic of a diabetes assessment device provided in another exemplary embodiment of this application.

[0026] Figure 8The diagram shown is a structural schematic of a diabetes assessment device provided in another exemplary embodiment of this application.

[0027] Figure 9 The image shown is a fundus image to be evaluated provided in another exemplary embodiment of this application.

[0028] Figure 10 The diagram shown is a flowchart of a diabetes assessment method provided in an exemplary embodiment of this application.

[0029] Figure 11 The diagram shown is a flowchart of a diabetes assessment method provided in another exemplary embodiment of this application.

[0030] Figure 12 The diagram shown is a flowchart of a diabetes assessment method provided in another exemplary embodiment of this application.

[0031] Figure 13 The diagram shown is a flowchart of a diabetes assessment method provided in another exemplary embodiment of this application.

[0032] Figure 14 The diagram shown is a flowchart of a diabetes assessment method provided in another exemplary embodiment of this application.

[0033] Figure 15 The diagram shown is a flowchart of a diabetes assessment method provided in another exemplary embodiment of this application.

[0034] Figure 16 The diagram shown is a flowchart of a diabetes assessment method provided in another exemplary embodiment of this application.

[0035] Figure 17 The diagram shown is a structural schematic of an electronic device provided in an exemplary embodiment of this application. Detailed Implementation

[0036] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0037] Exemplary System

[0038] Figure 1 The diagram shown illustrates a scenario applicable to an embodiment of this application. Figure 1As shown, the scenario applicable to this application embodiment includes an image acquisition device 100 and a computer device 200, wherein there is a communication connection between the image acquisition device 100 and the computer device 200. The communication connection can be a wired connection or a wireless connection.

[0039] Specifically, the image acquisition device 100 is used to acquire fundus images to be evaluated. The image acquisition device 100 can be a fundus camera or other devices with image acquisition capabilities, as long as they can acquire fundus images. This application does not specifically limit the structure of the image acquisition device 100.

[0040] Computer device 200 is used to receive fundus images to be evaluated acquired by image acquisition device 100, and to determine the diabetes assessment result based on the fundus images. Computer device 200 can be a general-purpose computer or a computer device composed of dedicated integrated circuits, etc., and this application embodiment does not specifically limit this. For example, computer device 200 can be a mobile terminal device such as a mobile phone or tablet computer, or a portable computer, desktop computer, etc., and this application does not specifically limit the type of computer device 200. The number of computer devices 200 can be one or more, and the types of multiple computer devices 200 can be the same or different, and this application does not specifically limit the number and type of computer devices 200. Computer device 200 can be used to determine the diabetes assessment result corresponding to the fundus images to be evaluated, without the need for invasive examinations such as blood tests; the diabetes assessment result can be determined solely based on the fundus images to be evaluated, without causing harm to the patient's physical or mental health.

[0041] Exemplary device

[0042] Figure 2 The diagram shown is a structural schematic of a diabetes assessment device provided in an exemplary embodiment of this application. Figure 2 As shown in the embodiment of this application, the diabetes assessment device 20 includes the following modules.

[0043] Image determination module 21 is configured to determine fundus images to be evaluated.

[0044] For example, the fundus image to be evaluated is used to characterize the physiological structures at the base of the eyeball. The fundus image to be evaluated can be a color fundus image, a confocal fundus image, a fluorescence angiography fundus image, etc., and this application does not specifically limit the type of fundus image to be evaluated.

[0045] Figure 3The image shown is a fundus image to be evaluated according to an exemplary embodiment of this application. The fundus image to be evaluated can be an image obtained from the fundus of the subject's eyeball, captured using general or specialized imaging equipment. The fundus image to be evaluated may include a vascular image region 4, an optic disc image region 3, and a macular image region (…). Figure 3 Image regions of non-pathological physiological structures (not shown in the image) may also include image regions of microaneurysms (not shown in the image). Figure 3 Images of lesion structures, such as hemorrhage image region 2 and exudation image region 1 (not shown in the image), are included. The fundus images to be evaluated can be analog or digital images; this application does not impose any specific limitations. The subjects to be evaluated can be individuals suspected of having diabetes or other animals.

[0046] The diabetes assessment module 22 is configured to determine the diabetes assessment result based on the fundus image to be assessed.

[0047] By setting up an image determination module, the fundus image to be evaluated can be determined. By setting up a diabetes assessment module, the diabetes assessment result can be determined based on the fundus image. This means that invasive examinations such as blood tests are unnecessary; the diabetes assessment result can be determined solely based on the fundus image, causing no harm to the patient's physical or mental well-being.

[0048] Figure 4 The diagram shown is a structural schematic of a diabetes assessment device provided in another exemplary embodiment of this application. Figure 2 This application extends from the embodiments shown. Figure 4 The illustrated embodiment will be described in detail below. Figure 4 The illustrated embodiments and Figure 2 The differences between the embodiments shown are not repeated here, and the similarities are not repeated here.

[0049] like Figure 4 As shown, in the diabetes assessment device provided in this application embodiment, the diabetes assessment module 22 includes the following modules.

[0050] The suspected lesion area determination module 221 is configured to determine the suspected lesion area of ​​the fundus image to be evaluated based on the fundus image to be evaluated.

[0051] The suspected lesion area can be an image area containing lesion structures such as microaneurysm, hemorrhage 2, and exudation 1 in the fundus image to be evaluated, or it can be an image area containing non-lesion structures such as small blood vessel image areas.

[0052] Specifically, suspected lesion areas can be identified by extracting the color, brightness, and shape of different regions in the fundus image to be evaluated. For example, Figure 3As shown, the grayscale of hemorrhage region 2 is smaller than that of the surrounding region, meaning the color of hemorrhage region 2 is darker than the surrounding region. Therefore, the hemorrhage region can be extracted by grayscale difference or color difference. Additionally, the microaneurysm image region ( Figure 3 The gray level of the hemangioma image (not shown) is also smaller than that of the surrounding area, meaning that the color of the hemangioma image is also darker than that of the surrounding area. Therefore, the suspected lesion areas such as the hemorrhage image area 2 and the microaneurysm image area can be extracted by the gray level difference or color difference of the fundus image to be evaluated.

[0053] The suspected lesion information determination module 222 is configured to determine suspected lesion information based on the suspected lesion area.

[0054] Specifically, after identifying a suspected lesion area, other characteristics of the suspected lesion area can be used to further determine the suspected lesion information.

[0055] In one embodiment, suspected lesion information includes non-lesion information and / or lesion information. Lesion information may include microaneurysm information, hemorrhage information, exudation information, cotton wool spot information, beaded vein information, intraretinal microvascular abnormality information, neovascularization information, etc. Non-lesion information may include short segment blood vessel information, nerve fiber information, etc.

[0056] The assessment result determination module 223 is configured to determine the diabetes assessment result based on suspected lesion information.

[0057] For example, the diabetes assessment result can be whether the subject corresponding to the fundus image being assessed has diabetes, or it can be the diabetes condition information of the subject corresponding to the fundus image being assessed. The diabetes condition information can be information on the severity of diabetes, or information on the extent of damage diabetes causes to the patient's body. Information on the extent of damage diabetes causes to the patient's body can be information on the extent of damage diabetes causes to the patient's kidneys, eyes, etc.

[0058] In one embodiment, the evaluation result determination module 223 is further configured to determine whether the subject to be evaluated, corresponding to the fundus image to be evaluated, has diabetes based on suspected lesion information.

[0059] In one embodiment, the evaluation result determination module 223 is further configured to: if all suspected lesion information is non-lesion information, determine that the subject to be evaluated corresponding to the fundus image to be evaluated does not have diabetes. For example, since the fundus image to be evaluated can be a partial fundus image, if all suspected lesion information is non-lesion information, it can be determined that the probability of the subject to be evaluated corresponding to the fundus image to be evaluated having diabetes is low. The evaluation result determination module 223 can be configured according to the actual application scenario to obtain an accurate judgment result.

[0060] In one embodiment, the evaluation result determination module 223 is further configured to: if the suspected lesion information includes lesion information, determine the diabetic condition information of the subject to be evaluated corresponding to the fundus image to be evaluated based on the lesion information.

[0061] For example, diabetes information can be risk information about developing diabetes or information about the severity of diabetes.

[0062] For example, information on the severity of diabetes may include: mild, moderate, moderate to severe, and severe.

[0063] Mild severity can be a result where the fundus image being evaluated only includes microaneurysm regions. Moderate severity can be a result where the fundus image being evaluated includes less than or equal to a preset number of microaneurysm regions, hemorrhage regions, and exudation regions. Moderate to severe severity can be a result where the fundus image being evaluated includes more than a preset number of hemorrhage regions, more than a preset number of beaded vein changes, or at least one obvious intraretinal microvascular abnormality. Severe severity can be a result where the fundus image being evaluated includes beaded veins. The preset number can be determined based on specific circumstances and is not specifically limited in this application. The criteria for classifying mild, moderate, and moderate to severe severity can also be other criteria, and are not specifically limited in this application.

[0064] By setting up the suspected lesion area determination module 221, the suspected lesion area can be determined based on the fundus image to be evaluated. By setting up the suspected lesion information determination module 222, the suspected lesion information of the suspected lesion area can be determined based on the suspected lesion area. By setting up the evaluation result determination module 223, the diabetes evaluation result can be determined based on the suspected lesion information. This makes the diabetes evaluation result more detailed and accurate, provides doctors with more accurate diagnostic suggestions, reduces the doctor's diagnosis time, and improves the accuracy of the doctor's diagnosis.

[0065] Figure 5 The diagram shown is a structural schematic of a diabetes assessment device provided in another exemplary embodiment of this application. Figure 4 This application extends from the embodiments shown. Figure 5 The illustrated embodiment will be described in detail below. Figure 5 The illustrated embodiments and Figure 4 The differences between the embodiments shown are not repeated here, and the similarities are not repeated here.

[0066] like Figure 5 As shown, in the diabetes assessment device provided in this application embodiment, the suspected lesion information determination module 222 includes the following units.

[0067] The suspected lesion feature information determination unit 2221 is configured to determine the suspected lesion feature information of the suspected lesion area based on the suspected lesion area.

[0068] For example, the suspected lesion feature information includes at least one of the following: shape information, color information, topological information, location information, and distribution information of the suspected lesion area.

[0069] For example, based on the suspected lesion area, the suspected lesion feature information of the suspected lesion area can be determined, and the edge of the suspected lesion area can be extracted to obtain the shape information of the suspected lesion area. The edge of the suspected lesion area can be extracted using the maximum bounding rectangle algorithm, the maximum bounding circle algorithm, edge feature extraction algorithm, etc., and this application does not make specific limitations.

[0070] Based on the suspected lesion area, the suspected lesion feature information of the suspected lesion area can be determined. Alternatively, the color (including brightness) of the suspected lesion area can be extracted to obtain the color information of the suspected lesion area. The color of the suspected lesion area can be extracted by extracting the sum of the gray values ​​of each pixel or by extracting the average gray value of each pixel. This application does not make any specific limitation.

[0071] Based on the suspected lesion area, the characteristic information of the suspected lesion area can be determined. Topological information of the suspected lesion area can also be extracted, for example, extracting regions of the same color as the suspected lesion area in the surrounding area. Based on the characteristic information of the suspected lesion area, the location information of the suspected lesion area can also be extracted. The location information can be the position of the geometric center of the suspected lesion area relative to the optic disc, or the absolute position of the geometric center of the suspected lesion area in the fundus image to be evaluated. The characteristic information of the suspected lesion can include one or more of shape information, color information, topological information, and location information; this application does not impose specific limitations.

[0072] The suspected lesion information determination unit 2222 is configured to determine suspected lesion information in the suspected lesion area based on the suspected lesion feature information.

[0073] Suspected lesion information includes non-lesion information and / or lesion information. Lesion information can be microaneurysm information, hemorrhage information, and exudation information. Non-lesion information can be information about small segments of blood vessels.

[0074] For example, since the exudation image region is relatively brighter than the microaneurysm image region and the hemorrhage image region in the fundus image to be evaluated, a brightness threshold can be set. When the brightness of the suspected lesion region meets the preset brightness threshold, the suspected lesion information can be exudation information. Regarding grayscale values, the grayscale of the exudation image region is relatively larger than that of the microaneurysm image region and the hemorrhage image region. Therefore, a grayscale threshold can be set. When the grayscale of the suspected lesion region meets the preset grayscale threshold or threshold range, the suspected lesion information can be exudation information.

[0075] For example, the grayscale values ​​and brightness of the microaneurysm and hemorrhage image regions are similar, making them impossible to determine based on color information. However, the edge shapes of the microaneurysm and hemorrhage image regions differ. Since microaneurysm image regions are generally smooth-edged ellipses or circles, when the shape information of a suspected lesion region is a smooth-edged ellipse or circle, the suspected lesion information could be microaneurysm information. Conversely, since hemorrhage image regions are generally ellipses or circles with jagged or spiky edges, when the shape information of a suspected lesion region is a jagged or spiky ellipse or circle, the suspected lesion information could be hemorrhage information.

[0076] For example, small blood vessel image regions and hemorrhage image regions have similar grayscale values ​​and brightness, making them impossible to distinguish using color information. Furthermore, the edge information of small blood vessel segments may resemble that of hemorrhage information; for instance, a small blood vessel segment could be elliptical. Therefore, it is not easy to differentiate between small blood vessel image regions and hemorrhage image regions using color and shape information. However, since blood vessels are continuous, topological information can be used to determine whether suspected lesion information is hemorrhage. When the topological information of the lesion region extracted around the lesion region is a blood vessel and is collinear with the blood vessel, it indicates that the suspected lesion information is small blood vessel information; that is, the small blood vessel segment is only a suspected lesion, i.e., it is not lesion information. When the topological information of the lesion region cannot be extracted around the lesion region, it indicates that the suspected lesion information is hemorrhage information.

[0077] For example, different suspected lesion information may appear in different locations in the fundus image to be evaluated; therefore, suspected lesion information can also be determined by location information.

[0078] By including a suspected lesion feature information determination unit 2221 in the suspected lesion information determination module 222, the lesion feature information of the suspected lesion area can be determined based on the suspected lesion area. By including a suspected lesion information determination unit 2222 in the suspected lesion information determination module 222, the suspected lesion information can be determined based on the lesion feature information, thereby providing more references for the confirmation of suspected lesion information and improving the accuracy of the judgment of suspected lesion information.

[0079] Figure 6 The diagram shown is a structural schematic of a diabetes assessment device provided in another exemplary embodiment of this application. Figure 5 This application extends from the embodiments shown. Figure 6 The illustrated embodiment will be described in detail below. Figure 6 The illustrated embodiments and Figure 5 The differences between the embodiments shown are not repeated here, and the similarities are not repeated here.

[0080] like Figure 6 As shown, in the diabetes assessment device provided in this application embodiment, the assessment result determination module 223 includes the following units.

[0081] The quantization unit 2231 is configured to quantify lesion information and obtain quantification results.

[0082] For example, the quantification result can be obtained by quantifying the quantity of lesion information, or by quantifying other characteristics of the lesion information. For instance, calculating the number of microaneurysm image regions, the number of hemorrhage image regions 2, and the number of low-exudation image regions 1 in the fundus image to be evaluated, etc., means that the quantification result can be the quantity of lesion information. The quantification result can also be the absolute location information of the lesion information in the fundus image to be evaluated; any quantification result obtained by quantifying the lesion information is acceptable. This application does not specifically limit the content of the quantification result.

[0083] The disease information determination unit 2232 is configured to determine the diabetes disease information of the subject to be evaluated based on the quantitative results corresponding to the fundus image to be evaluated.

[0084] For example, information about the condition of diabetes can be determined based on the quantification results. For instance, information about the condition of diabetes can be determined based on the number of lesions or the absolute location of the lesions in the fundus image to be evaluated. As long as the information about the condition of diabetes is determined based on the quantification results, this application does not specifically limit the quantification results used to determine information about the condition of diabetes.

[0085] By including a quantization unit 2231 in the assessment result determination module 223, the lesion information can be quantified to obtain a quantification result. By including a condition information determination unit 2232 in the assessment result determination module 223, the diabetic condition information corresponding to the fundus image to be evaluated can be determined based on the quantification result, thereby further analyzing the lesion information. Since the diabetic condition information is determined based on the lesion information, the accuracy of the diabetic condition information is further improved.

[0086] Figure 7 The diagram shown is a structural schematic of a diabetes assessment device provided in another exemplary embodiment of this application. Figure 6This application extends from the embodiments shown. Figure 7 The illustrated embodiment will be described in detail below. Figure 7 The illustrated embodiments and Figure 6 The differences between the embodiments shown are not repeated here, and the similarities are not repeated here.

[0087] like Figure 7 As shown, in the diabetes assessment device provided in this application embodiment, the quantification unit 2231 includes the following sub-units.

[0088] The lesion feature information determination subunit 22311 is configured to determine lesion feature information based on lesion information.

[0089] For example, the lesion feature information includes at least one of the following: lesion location information, lesion quantity information, lesion area information, distribution information, and morphological information.

[0090] For example, lesion location information can be the location information of the geometric center of the lesion area corresponding to the lesion information. For example, when the lesion information is microaneurysm information, the lesion location information can be the absolute location information of the geometric center of the lesion area corresponding to the microaneurysm information in the fundus image to be evaluated. Lesion quantity information can be the number of lesion areas corresponding to the lesion information. For example, when the lesion information is microaneurysm information, the lesion quantity information can be the number of lesion areas corresponding to the microaneurysm information. Lesion area information can be the total area or average area of ​​the lesion areas corresponding to the lesion information. For example, when the lesion information is microaneurysm information, the lesion area information can be the total area or average area of ​​the lesion areas corresponding to the microaneurysm information. Distribution information can be the distribution of the lesion areas corresponding to the lesion information. For example, when the lesion areas are distributed near the macula, the lesion information has a greater impact on the diabetic condition. Morphological information can be the shape and orientation of the lesion areas corresponding to the lesion information. For example, the shape can be circular, elliptical, serrated, etc., and the orientation can be the roundness of a circle.

[0091] The quantification result determination subunit 22312 is configured to determine the quantification result based on lesion feature information.

[0092] After determining the quantitative results based on the lesion characteristic information, the diabetes condition information can be determined based on the quantitative results. For example, the diabetes condition information can include mild, moderate, moderate-severe, and severe. Mild can be a result where the fundus image to be evaluated only includes microaneurysm image regions, and the area of ​​the microaneurysm image regions is less than a first preset area threshold. Moderate can be a result where the fundus image to be evaluated includes less than or equal to a preset number of microaneurysm image regions, hemorrhage image regions, and exudation image regions, and the area of ​​the microaneurysm image regions is less than a second preset area threshold. Moderate-severe can be a result where the fundus image to be evaluated includes more than a preset number of hemorrhage image regions, and the area of ​​the microaneurysm image regions is greater than the second preset area threshold; it can also be a result where the fundus image to be evaluated includes more than a preset number of vein beading change images; or it can be a result where the fundus image to be evaluated includes at least one obvious intraretinal microvascular abnormality image. The preset number can be determined according to specific circumstances, and this application does not impose a specific limitation. The criteria for classifying mild, moderate, moderate-severe, and severe conditions can also be other criteria. For example, moderate condition can also be defined as a result where the fundus image to be evaluated includes less than or equal to a preset number of microaneurysm image areas, hemorrhage image areas, and exudation image areas, and the area of ​​the hemorrhage image area is less than a preset hemorrhage area threshold. This application does not specifically limit the criteria for classifying mild, moderate, and moderate-severe conditions.

[0093] For example, after determining the quantitative result based on the lesion feature information, the diabetes condition information can be determined based on the quantitative result, such as the duration of diabetes. When the fundus image to be evaluated only includes a microaneurysm image region, and the area of ​​the microaneurysm image region is less than a first preset area threshold, the duration of diabetes can be determined to be one year. When the fundus image to be evaluated includes more than a preset number of venous beading changes, the duration of diabetes can be determined to be five years. This application does not specifically limit the relationship between the quantitative result and the duration of diabetes.

[0094] By including a lesion feature information determination subunit 22311 in the quantification unit 2231, the lesion feature information corresponding to the lesion information can be determined based on the lesion information. By including a quantification result determination subunit 22312 in the quantification unit 2231, the quantification result can be determined based on the lesion feature information, thereby making the quantification result more accurate. Since the diabetes condition information is determined based on the quantification result, the accuracy of the diabetes condition information is further improved.

[0095] Figure 8 The diagram shown is a structural schematic of a diabetes assessment device provided in another exemplary embodiment of this application. Figure 2 This application extends from the embodiments shown. Figure 8 The illustrated embodiment will be described in detail below. Figure 8The illustrated embodiments and Figure 2 The differences between the embodiments shown are not repeated here, and the similarities are not repeated here.

[0096] like Figure 8 As shown in the embodiments of this application, the diabetes assessment device also includes the following modules.

[0097] The preprocessing module 224 is configured to perform preprocessing operations on the fundus image to be evaluated to obtain a preprocessed image.

[0098] Preprocessing operations include at least one of denoising, normalization, and enhancement. For example, denoising can be performed first to obtain a denoised fundus image to be evaluated. Then, normalization can be performed on the denoised fundus image to obtain a normalized fundus image to be evaluated. Finally, enhancement can be performed on the normalized fundus image to obtain an enhanced fundus image to be evaluated. Alternatively, only denoising, normalization, or enhancement operations can be performed on the fundus image to be evaluated. That is, the above preprocessing operations can be selected according to actual needs, and this application does not impose specific limitations.

[0099] In one embodiment, the preprocessing module 224 includes at least one of the following units.

[0100] The image quality assessment unit 2241 is configured to perform a quality assessment operation on the fundus image to be assessed to obtain the preprocessed image.

[0101] For example, image quality assessment can be to evaluate whether the fundus image to be assessed is indeed a fundus image; if it is not a fundus image, it can be directly removed. Image quality assessment can also be to evaluate whether the fundus image to be assessed is clear; if the clarity is insufficient, it can also be directly removed.

[0102] The denoising processing unit 2242 is configured to perform denoising processing on the fundus image to be evaluated.

[0103] For example, the noise reduction operation can be to remove noise generated during the acquisition of the fundus image to be evaluated, such as... Figure 9 As shown, during the acquisition of the fundus image to be evaluated, there is a large area with low grayscale due to eyelid occlusion 5. Figure 3 The grayscale of the exudation image region 1 shown is... Figure 9The grayscale values ​​of the occluded region 5 shown are similar, but the area of ​​the exudate image region 1 is relatively smaller than that of the occluded region 5. Therefore, regions smaller than a preset noise grayscale threshold can be extracted to obtain suspected noise regions. The area of ​​the suspected noise regions is then calculated, and regions with areas larger than a preset noise area threshold are identified as noise regions. Finally, the noise regions are removed. The denoising process can also remove other information unrelated to the lesion region from the fundus image to be evaluated. For example, personal information text of the patient and text indicating the shooting time in the fundus image to be evaluated. The denoising process can also employ multi-scale denoising algorithms based on the Shearlet framework, multi-scale denoising algorithms based on Ridgelet transform, etc. This application does not specifically limit the algorithm for the denoising process. The denoising process can reduce the interference of noise such as occluded region 5, personal information text of the patient, and text indicating the shooting time on the determination of the lesion region, thereby improving the accuracy of lesion region determination.

[0104] The normalization processing unit 2243 is configured to perform normalization processing on the fundus image to be evaluated.

[0105] For example, normalization can limit values ​​of different magnitudes to a single range without affecting their relative magnitudes. For instance, normalization can convert the grayscale values ​​of multiple pixels to the range [0,1] while maintaining the relative magnitudes of these values. For example, normalization can be performed using the formula (x-min) / (max-min), where x represents the grayscale value of a pixel, max represents the maximum grayscale value among the pixels in the fundus image to be evaluated, and min represents the minimum grayscale value among the pixels in the fundus image to be evaluated. By limiting the grayscale values ​​of multiple pixels to a single range through normalization, subsequent calculations are facilitated.

[0106] Enhancement processing unit 2244 is configured to perform enhancement processing operations on the fundus image to be evaluated.

[0107] For example, enhancement processing can involve adding information or transforming data in the original image to selectively highlight features of interest or suppress unwanted features. For instance, if a region with a small grayscale value is a region of interest in the fundus image to be evaluated, its grayscale value can be set to 0. Specifically, regions with grayscale values ​​less than 10 can have their grayscale values ​​set to 0. Conversely, if a region with a large grayscale value is a region of interest in the fundus image to be evaluated, its grayscale value can be set to 255. Specifically, regions with grayscale values ​​greater than 200 can have their grayscale values ​​set to 255. Enhancement processing can also involve multiplying the grayscale values ​​of all pixels in the fundus image to be evaluated by a certain value, such as 3 or 5. The specific multiplication value can be chosen according to actual needs and is not specifically limited in this application. Enhancement processing can make lesion areas easier to extract and identify, further improving the accuracy of lesion area determination.

[0108] The suspected lesion area determination module 221 is further configured to: determine the suspected lesion area based on the preprocessed image.

[0109] Exemplary methods

[0110] Figure 10 The diagram shown is a flowchart illustrating a diabetes assessment method provided in an exemplary embodiment of this application. Figure 10 As shown in the embodiments of this application, the diabetes assessment method includes the following steps.

[0111] Step 1001: Determine the fundus image to be evaluated.

[0112] For example, the fundus images to be evaluated are used to characterize the physiological structures at the base of the eyeball.

[0113] Step 1002: Determine the diabetes assessment result based on the fundus image to be evaluated.

[0114] Figure 11 The diagram shown is a flowchart illustrating a diabetes assessment method provided in another exemplary embodiment of this application. Figure 10 This application extends from the embodiments shown. Figure 11 The illustrated embodiment will be described in detail below. Figure 11 The illustrated embodiments and Figure 10 The differences between the embodiments shown are not repeated here, and the similarities are not repeated here.

[0115] like Figure 11 As shown in the embodiments of this application, the method for assessing diabetes, which determines the diabetes assessment result based on the fundus image to be assessed, includes the following steps.

[0116] Step 1101: Determine the suspected lesion area in the fundus image to be evaluated based on the fundus image to be evaluated.

[0117] Step 1102: Determine the suspected lesion information based on the suspected lesion area.

[0118] For example, suspected lesion information includes non-lesion information and / or lesion information. Lesion information may include at least one of microaneurysm information, hemorrhage information, and exudation information. Lesion information may also include at least one of cotton wool spot information, beaded vein information, intraretinal microvascular abnormality information, and neovascularization information.

[0119] Step 1103: Determine the diabetes assessment result based on suspected lesion information.

[0120] In one embodiment, determining the diabetes assessment result based on suspected lesion information can be done by determining whether the subject to be assessed, corresponding to the fundus image to be assessed, has diabetes based on the suspected lesion information.

[0121] In one embodiment, determining the diabetes assessment result based on suspected lesion information can be as follows: if all suspected lesion information is non-lesion information, it is determined that the subject to be assessed corresponding to the fundus image to be assessed does not have diabetes.

[0122] In one embodiment, determining the diabetes assessment result based on suspected lesion information may be as follows: if the suspected lesion information includes lesion information, the diabetes condition information of the subject to be assessed corresponding to the fundus image to be assessed is determined based on the lesion information, wherein the diabetes condition information includes risk information for developing diabetes and / or severity information for diabetes.

[0123] Figure 12 The diagram shown is a flowchart illustrating a diabetes assessment method provided in another exemplary embodiment of this application. Figure 11 This application extends from the embodiments shown. Figure 12 The illustrated embodiment will be described in detail below. Figure 12 The illustrated embodiments and Figure 11 The differences between the embodiments shown are not repeated here, and the similarities are not repeated here.

[0124] like Figure 12 As shown in the embodiments of this application, the method for assessing diabetes includes the following steps for determining suspected lesion information based on suspected lesion areas.

[0125] Step 1201: Determine the suspected lesion feature information of the suspected lesion area based on the suspected lesion area.

[0126] For example, the suspected lesion feature information includes at least one of the following: shape information, color information, topological information, and location information of the suspected lesion area.

[0127] Step 1202: Determine the suspected lesion information of the suspected lesion area based on the suspected lesion feature information.

[0128] Figure 13 The diagram shown is a flowchart illustrating a diabetes assessment method provided in another exemplary embodiment of this application. Figure 11 This application extends from the embodiments shown. Figure 13 The illustrated embodiment will be described in detail below. Figure 13 The illustrated embodiments and Figure 11 The differences between the embodiments shown are not repeated here, and the similarities are not repeated here.

[0129] like Figure 13 As shown in the embodiments of this application, the method for assessing diabetes, which determines the assessment result based on suspected lesion information, includes the following steps.

[0130] Step 1301: Quantify the lesion information to obtain the quantification results.

[0131] Step 1302: Determine the diabetes condition information of the subject to be evaluated corresponding to the fundus image to be evaluated based on the quantification results.

[0132] Figure 14 The diagram shown is a flowchart illustrating a diabetes assessment method provided in another exemplary embodiment of this application. Figure 13 This application extends from the embodiments shown. Figure 14 The illustrated embodiment will be described in detail below. Figure 14 The illustrated embodiments and Figure 13 The differences between the embodiments shown are not repeated here, and the similarities are not repeated here.

[0133] like Figure 14 As shown in the embodiment of this application, the method for assessing diabetes involves quantifying lesion information to obtain quantification results, including the following steps.

[0134] Step 1401: Determine lesion characteristic information based on lesion information.

[0135] For example, the lesion feature information includes at least one of the following: lesion location information, lesion quantity information, lesion area information, distribution information, and morphological information.

[0136] Step 1402: Determine the quantitative result based on lesion characteristic information.

[0137] Figure 15 The diagram shown is a flowchart illustrating a diabetes assessment method provided in another exemplary embodiment of this application. Figure 10 This application extends from the embodiments shown. Figure 15 The illustrated embodiment will be described in detail below. Figure 15 The illustrated embodiments and Figure 10 The differences between the embodiments shown are not repeated here, and the similarities are not repeated here.

[0138] like Figure 15 As shown in the embodiments of this application, the method for assessing diabetes, which determines the diabetes assessment result based on the fundus image to be assessed, includes the following steps.

[0139] Step 1501: Perform preprocessing operations on the fundus image to be evaluated to obtain a preprocessed image.

[0140] The process of determining the diabetes assessment result based on the fundus image to be evaluated includes the following steps.

[0141] Step 1502: Determine the diabetes assessment results based on the preprocessed images.

[0142] Figure 16 The diagram shown is a flowchart illustrating a diabetes assessment method provided in another exemplary embodiment of this application. Figure 15 This application extends from the embodiments shown. Figure 16 The illustrated embodiment will be described in detail below. Figure 16 The illustrated embodiments and Figure 15 The differences between the embodiments shown are not repeated here, and the similarities are not repeated here.

[0143] like Figure 16 As shown, in the diabetes assessment method provided in this application embodiment, the preprocessing operation of the fundus image to be assessed to obtain the preprocessed image includes at least one of the following steps:

[0144] Step 1601: Perform a quality assessment on the fundus image to be evaluated to obtain a preprocessed image.

[0145] Step 1602: Perform noise reduction on the fundus image to be evaluated to obtain a preprocessed image.

[0146] Step 1603: Normalize the fundus image to be evaluated to obtain a preprocessed image.

[0147] Step 1604: Perform enhancement processing on the fundus image to be evaluated to obtain a preprocessed image.

[0148] Exemplary electronic devices

[0149] Below, for reference Figure 17 This application describes an electronic device according to an exemplary embodiment of the present application. Figure 17 The diagram shown is a structural schematic of an electronic device provided in an exemplary embodiment of this application.

[0150] like Figure 17As shown, the electronic device 170 includes one or more processors 1701 and memory 1702.

[0151] The processor 1701 may be a central processing unit (CPU) or other form of processing unit with data processing capabilities and / or instruction execution capabilities, and may control other components in the electronic device 170 to perform desired functions.

[0152] The memory 1702 may include one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may include, for example, random access memory (RAM) and / or cache memory. The non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc. One or more computer program instructions may be stored on the computer-readable storage medium, and the processor 1701 may execute the program instructions to implement the diabetes assessment methods of the various embodiments of this application described above and / or other desired functions. Various contents, such as fundus images to be evaluated, may also be stored in the computer-readable storage medium.

[0153] In one example, the electronic device 170 may also include an input device 1703 and an output device 1704, which are interconnected via a bus system and / or other forms of connection mechanism (not shown).

[0154] The input device 1703 may include, for example, a keyboard, a mouse, etc.

[0155] The output device 1704 can output various information to the outside, including diabetes assessment results. The output device 1704 may include, for example, a display, a communication network, and a remote output device connected thereto.

[0156] Of course, for the sake of simplicity, Figure 17 Only some of the components of the electronic device 170 relevant to this application are shown in this illustration; components such as buses, input / output interfaces, etc., are omitted. In addition, the electronic device 170 may include any other suitable components depending on the specific application.

[0157] Exemplary computer program products and computer-readable storage media

[0158] In addition to the methods and devices described above, embodiments of this application may also be computer program products comprising computer program instructions that, when executed by a processor, cause the processor to perform the steps in the diabetes assessment methods according to various embodiments of this application as described in the "Exemplary Methods" section of this specification.

[0159] The computer program product can be written in any combination of one or more programming languages ​​to perform the operations of the embodiments of this application. The programming languages ​​include object-oriented programming languages ​​such as Java and C++, as well as conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computing device, partially on the user's computing device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0160] Furthermore, embodiments of this application may also be computer-readable storage media storing computer program instructions thereon, which, when executed by a processor, cause the processor to perform the steps in the diabetes assessment methods according to various embodiments of this application described in the "Exemplary Methods" section above.

[0161] The computer-readable storage medium may be any combination of one or more readable media. A readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may, for example, include, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatuses, or devices, or any combination thereof. More specific examples of readable storage media (a non-exhaustive list) include: electrical connections having one or more wires, portable disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0162] The basic principles of this application have been described above with reference to specific embodiments. However, it should be noted that the advantages, benefits, and effects mentioned in this application are merely examples and not limitations, and should not be considered as essential features of each embodiment of this application. Furthermore, the specific details disclosed above are for illustrative and facilitative purposes only, and are not limitations. These details do not limit the application to the necessity of employing the aforementioned specific details for implementation.

[0163] The block diagrams of devices, apparatuses, devices, and systems involved in this application are merely illustrative examples and are not intended to require or imply that they must be connected, arranged, or configured in the manner shown in the block diagrams. As those skilled in the art will recognize, these devices, apparatuses, devices, and systems can be connected, arranged, and configured in any manner. Words such as “comprising,” “including,” “having,” etc., are open-ended terms meaning “including but not limited to,” and are used interchangeably with them. The terms “or” and “and” as used herein refer to the terms “and / or,” and are used interchangeably with them unless the context clearly indicates otherwise. The term “such as” as used herein refers to the phrase “such as but not limited to,” and is used interchangeably with it.

[0164] It should also be noted that in the apparatus, equipment, and methods of this application, the components or steps can be disassembled and / or recombined. These disassemblies and / or recombinations should be considered as equivalent solutions of this application.

[0165] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use this application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein can be applied to other aspects without departing from the scope of this application. Therefore, this application is not intended to be limited to the aspects shown herein, but rather to be carried out within the widest scope consistent with the principles and novel features of the invention herein.

[0166] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Any modifications or equivalent substitutions made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A diabetes assessment device, characterized by, The method comprises: an image determining module configured to determine an eye fundus image to be evaluated, wherein the eye fundus image to be evaluated is used to represent physiological structures of the fundus of an eyeball; and a diabetes evaluation module configured to determine a diabetes evaluation result based on the eye fundus image to be evaluated; the diabetes evaluation module comprises: a suspected lesion area determining module configured to determine a suspected lesion area of the eye fundus image to be evaluated based on the eye fundus image to be evaluated; a suspected lesion information determining module configured to determine suspected lesion information of the suspected lesion area based on the suspected lesion area; and an evaluation result determining module configured to determine the diabetes evaluation result based on the suspected lesion information; the evaluation result determining module is further configured to, if the eye fundus image to be evaluated is a partial eye fundus image, determine that a probability of a subject corresponding to the eye fundus image to be evaluated suffering from diabetes is low if all the suspected lesion information is non-lesion information; the evaluation result determining module is further configured to, if the suspected lesion information comprises lesion information, determine diabetes condition information of the subject corresponding to the eye fundus image to be evaluated based on the lesion information, wherein the diabetes condition information comprises risk information of suffering from diabetes and / or severity information of diabetes and / or damage degree information of diabetes to the body, and time of suffering from diabetes.

2. The diabetes evaluation device of claim 1, wherein, the suspected lesion information determining module comprises: a suspected lesion feature information determining unit configured to determine suspected lesion feature information of the suspected lesion area based on the suspected lesion area, wherein the suspected lesion feature information comprises at least one of shape information, color information, topological information and position information of the suspected lesion area; a suspected lesion information determining unit configured to determine the suspected lesion information of the suspected lesion area based on the suspected lesion feature information.

3. The diabetes evaluation device of claim 1, wherein, the evaluation result determining module is further configured to determine whether the subject corresponding to the eye fundus image to be evaluated suffers from diabetes based on the suspected lesion information.

4. The diabetes evaluation device of claim 3, wherein, the suspected lesion information comprises non-lesion information and / or lesion information, and the lesion information comprises at least one of microaneurysm information, hemorrhage information and exudation information.

5. The diabetes evaluation device of claim 4, wherein, the evaluation result determining module is further configured to determine that the subject corresponding to the eye fundus image to be evaluated does not suffer from diabetes if all the suspected lesion information is the non-lesion information.

6. The diabetes evaluation device of claim 1, wherein, the evaluation result determining module comprises: a quantifying unit configured to quantify the lesion information to obtain a quantification result; a condition information determining unit configured to determine the diabetes condition information of the subject corresponding to the eye fundus image to be evaluated based on the quantification result.

7. The diabetes evaluation device of claim 6, wherein, the quantifying unit comprises: a lesion feature information determining sub-unit configured to determine lesion feature information based on the lesion information, wherein the lesion feature information comprises at least one of lesion position information, lesion quantity information and lesion area information, distribution information and morphological information; a quantification result determining sub-unit configured to determine the quantification result based on the lesion feature information.

8. The diabetes assessment device of any one of claims 1 to 7, wherein, the method further comprises: The preprocessing module is configured to perform a preprocessing operation on the to-be-evaluated fundus image to obtain a preprocessed image. The diabetes evaluation module is further configured to determine the diabetes evaluation result based on the preprocessed image.

9. The diabetes evaluation device of claim 8, wherein, The preprocessing module includes at least one of the following units: An image quality evaluation unit configured to perform a quality evaluation operation on the to-be-evaluated fundus image to obtain the preprocessed image. A denoising processing operation unit configured to perform a denoising processing operation on the to-be-evaluated fundus image to obtain the preprocessed image. A normalization processing operation unit configured to perform a normalization processing operation on the to-be-evaluated fundus image to obtain the preprocessed image. An enhancement processing operation unit configured to perform an enhancement processing operation on the to-be-evaluated fundus image to obtain the preprocessed image.

10. A method of diabetes assessment, characterized by, The method comprises: determining a to-be-evaluated fundus image, wherein the to-be-evaluated fundus image is used to represent the physiological structure of the fundus of the eyeball; and determining a diabetes evaluation result based on the to-be-evaluated fundus image; The method of determining the diabetes evaluation result based on the to-be-evaluated fundus image comprises: determining a suspected lesion area of the to-be-evaluated fundus image based on the to-be-evaluated fundus image; determining suspected lesion information of the suspected lesion area based on the suspected lesion area; and determining the diabetes evaluation result based on the suspected lesion information; The method of determining the diabetes evaluation result based on the suspected lesion information comprises: if the to-be-evaluated fundus image is a partial fundus image, if all the suspected lesion information is non-lesion information, it is determined that the to-be-evaluated object corresponding to the to-be-evaluated fundus image has a low probability of suffering from diabetes; if the suspected lesion information includes lesion information, determining diabetes condition information of the to-be-evaluated object corresponding to the to-be-evaluated fundus image based on the lesion information, wherein the diabetes condition information includes risk information of suffering from diabetes and / or severity information of diabetes and / or damage degree information of diabetes to the body, and time of suffering from diabetes.

11. A computer readable storage medium, characterized in that, The storage medium stores a computer program, and the computer program is used to execute the diabetes evaluation method in claim 10.

12. An electronic device, comprising: The electronic device comprises: a processor; and a memory for storing instructions executable by the processor; The processor is configured to execute the diabetes evaluation method in claim 10.

Images