An intelligent ear disease recognition method and system applicable to multiple types of devices

Abstract

The present invention discloses an intelligent ear disease recognition method and system applicable to multiple types of devices, comprising: obtaining an ear lesion image dataset, preprocessing the ear lesion image dataset to obtain a preprocessed image; constructing an intelligent ear disease recognition model, wherein the intelligent recognition model includes: a downsampling module, a feature extraction module, and an ear injury feature fusion module; based on the downsampling module, downsampling the preprocessed image to obtain a first feature map and a second feature map; based on the feature extraction module, high-level semantic feature extraction is performed on the first feature map to obtain a first feature information map; low-level semantic feature extraction is performed on the second feature map to obtain a second feature information map; based on the ear injury feature fusion module, feature fusion is performed on the first feature information map and the second feature information map to obtain a fused image. The present invention can improve the fidelity of the fused image, simultaneously capture global and local spatial information, and alleviate the problem of low precision.

Description

Technical Field

[0001] The present invention belongs to the technical field of ear disease recognition, and in particular relates to an ear disease intelligent recognition method and system applicable to multiple types of equipment. Background Art

[0002] The ear is one of the most important organs in the human body. It not only controls hearing, but also plays an important role in maintaining body balance. At the same time, the ear is also an organ that is often affected by diseases. If not detected and treated in time, ear diseases may lead to hearing loss and lifelong complications. Like other diseases, timely detection and appropriate treatment of ear diseases can help prevent the condition from worsening and avoid unnecessary side effects of certain drugs (such as broad-spectrum antibiotics). Compared with other diseases, examining ear diseases is a simple and non-invasive process. In clinical practice, otoscopy is a commonly used ear examination method. However, due to differences in professional knowledge and experience levels, misdiagnosis by medical professionals often occurs, especially in non-specialized clinics (pediatrics, general practitioners, emergency departments, primary care physicians, and telemedicine) and areas with limited medical resources.

[0003] To alleviate these challenges, there is an urgent need to develop a new generation of ear disease recognition devices that can provide timely, convenient, and accurate identification. Fortunately, otoscopes, commonly used for ear examinations, are easily replaceable devices that can be operated without specialized training. This opens the door to integrating self-recognition capabilities into devices using emerging deep learning technologies. While some studies have demonstrated promising results using deep learning models for intelligent recognition of various types of otoendoscopic images, these efforts often lack extensive data validation and focus solely on model development and accuracy without considering their applicability in real-world scenarios. In addition to focusing on recognition performance, intelligent ear disease recognition models should also balance the trade-offs between recognition accuracy, model parameter size, and model response speed. Clinical identification often requires repeated observation of the affected ear canal area. Therefore, a model that achieves extremely fast inference speed could enable individual users and otolaryngologists to perform thorough and in-depth scans of ear lesions in a real-time video format. This approach is more rigorous and scientific than relying solely on intelligent recognition based on a single static image, and is consistent with the actual clinical identification process. Furthermore, this model can be deployed on edge devices with limited computing power, such as low-end computers, smartphones, and tablets. Smaller model parameter size means minimal memory and storage requirements, enabling easy deployment on resource-constrained devices such as smartphones and embedded systems. Specifically, the key to using AI to identify ear diseases lies in developing a system that possesses the aforementioned features and is compatible with different devices. Such a system does not require expensive computing equipment or network services and can be reliably applied to a variety of local devices and scenarios, facilitating early detection and treatment of ear diseases for a wide range of patients. This, in turn, effectively reduces the overall impact and losses caused by ear diseases. Summary of the Invention

[0004] In order to solve the above technical problems, the present invention proposes an intelligent ear disease recognition method and system applicable to multiple types of devices to solve the problems existing in the above-mentioned prior art.

[0005] To achieve the above objectives, the present invention provides an intelligent ear disease recognition method applicable to multiple types of devices, comprising:

[0006] Acquire an ear lesion image dataset, and preprocess the ear lesion image dataset to obtain a preprocessed image;

[0007] Constructing an intelligent recognition model for ear diseases, wherein the intelligent recognition model includes: a downsampling module, a feature extraction module, and an ear injury feature fusion module;

[0008] Downsampling the preprocessed image based on the downsampling module to obtain a first feature map and a second feature map;

[0009] Based on the feature extraction module, high-level semantic feature extraction is performed on the first feature map to obtain a first feature information map; low-level semantic feature extraction is performed on the second feature map to obtain a second feature information map;

[0010] Based on the ear injury feature fusion module, the first feature information map and the second feature information map are feature fused to obtain a fused image; based on the fused image, intelligent identification of ear diseases is achieved.

[0011] Preferably, the process of obtaining the first characteristic map and the second characteristic map includes:

[0012] Downsampling the preprocessed image multiple times to generate a downsampled image;

[0013] The downsampled image is subjected to multi-stage downsampling to obtain a first feature map and a second feature map.

[0014] Preferably, the process of performing multi-stage downsampling on the downsampled image includes:

[0015] Performing a second stage of downsampling on the downsampled image to obtain a second feature map;

[0016] Performing a third-stage downsampling on the second feature map to obtain a third-stage image;

[0017] The third stage image is downsampled in the fourth stage to obtain the first feature map.

[0018] Preferably, the process of obtaining the first characteristic information graph includes:

[0019] The first feature map is processed by the point-by-point convolution layer, the batch normalization layer, and the linear rectification function layer in sequence to obtain the first feature information map.

[0020] Preferably, the process of obtaining the second characteristic information graph includes:

[0021] The second feature map is processed by multiple average pooling, point-by-point convolution, regularization, efficient channel attention, batch normalization, and linear rectification function to obtain the second feature information map.

[0022] Preferably, the process of efficient channel attention processing includes:

[0023] The image input to the efficient channel attention is the efficient channel first image. After the efficient channel first image is processed by global average pooling and the first activation function layer in sequence, the high school channel second image is obtained. The efficient channel first image and the high school channel second image enter the multiplication layer to obtain the multiplication layer output image, which is the efficient channel attention output image.

[0024] Preferably, the process of obtaining the fused image includes:

[0025] Adding elements of the first feature information map and the second feature information map to obtain a first fused feature map;

[0026] Extracting a first weight and a second weight of the first fused feature map;

[0027] Based on the first multiplication layer, the first weight and the second feature information map are element-wise multiplied to calculate a first weighted fusion feature map;

[0028] Based on the second multiplication layer, the second weight and the first feature information map are element-wise multiplied to calculate a second weighted fusion feature map;

[0029] Based on the addition layer, the first weighted fusion feature map and the second weighted fusion feature map are element-wise added to obtain a fused image.

[0030] Preferably, the method for intelligently identifying ear diseases further comprises:

[0031] Performing loss calculation on the second feature map to obtain a first loss value;

[0032] Performing loss calculation on the third-stage image to obtain a second loss value;

[0033] performing loss calculation on the fused image to obtain a third loss value;

[0034] Based on the loss function, the first loss value, the second loss value, and the third loss value are calculated to obtain a final loss value for optimizing the intelligent recognition model.

[0035] To achieve the above technical objectives, the present invention further provides an intelligent ear disease recognition system applicable to multiple types of devices, for implementing any of the above-mentioned intelligent ear disease recognition methods applicable to multiple types of devices, the system comprising:

[0036] An image acquisition module is used to acquire an ear lesion image dataset, preprocess the ear lesion image dataset, and obtain a preprocessed image;

[0037] A model building module is used to build an intelligent recognition model for ear diseases, wherein the intelligent recognition model includes: a downsampling module, a feature extraction module and an ear injury feature fusion module;

[0038] A downsampling module, configured to downsample the preprocessed image to obtain a first feature map and a second feature map;

[0039] A feature extraction module is configured to extract high-level semantic features from the first feature map to obtain a first feature information map; and extract low-level semantic features from the second feature map to obtain a second feature information map;

[0040] The ear injury feature fusion module is used to fuse the first feature information map and the second feature information map to obtain a fused image; based on the fused image, intelligent ear disease recognition is achieved.

[0041] The present invention also provides an electronic device, comprising: a processor and a memory storing computer program instructions; when the processor executes the computer program instructions, the method for intelligently identifying ear diseases applicable to multiple types of devices is implemented.

[0042] Compared with the prior art, the present invention has the following advantages and technical effects:

[0043] The present invention provides an intelligent ear disease recognition method applicable to multiple types of equipment, comprising: first, obtaining an ear lesion image dataset, and preprocessing the ear lesion image dataset to obtain a preprocessed image; secondly, constructing an intelligent ear disease recognition model, wherein the intelligent recognition model comprises: a downsampling module, a feature extraction module, and an ear damage feature fusion module; further, based on the downsampling module, downsampling the preprocessed image to obtain a first feature map and a second feature map; then, based on the feature extraction module, high-level semantic feature extraction is performed on the first feature map to obtain a first feature information map; low-level semantic feature extraction is performed on the second feature map to obtain a second feature information map; finally, based on the ear damage feature fusion module, feature fusion is performed on the first feature information map and the second feature information map to obtain a fused image; based on the fused image, intelligent ear disease recognition is achieved.

[0044] This invention improves the fidelity of the fused image by integrating multiple pixel dimensions and multi-weighted fusion factors before fusing the images. The ear injury feature fusion module simultaneously captures global and local spatial information and guides the network to focus on key areas in the feature map at each level, thereby alleviating the problem of low precision. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] The accompanying drawings, which constitute part of this application, are intended to provide a further understanding of this application. The exemplary embodiments and descriptions of this application are intended to explain this application and do not constitute an improper limitation on this application. In the accompanying drawings:

[0046] Figure 1 is a flow chart of a method according to an embodiment of the present invention;

[0047] Figure 2 A schematic diagram of a system according to an embodiment of the present invention;

[0048] Figure 3 This is a unit-level framework diagram of an embodiment of the present invention;

[0049] Figure 4 This is a sub-unit level framework diagram of an embodiment of the present invention;

[0050] Figure 5 A framework diagram of a class header function model according to an embodiment of the present invention;

[0051] Figure 6 This is a framework diagram of an efficient channel attention layer according to an embodiment of the present invention;

[0052] Figure 7 This is a framework diagram of the local-global spatial feature fusion unit according to an embodiment of the present invention. DETAILED DESCRIPTION

[0053] It should be noted that, in the absence of conflict, the embodiments and features of the embodiments in this application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0054] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0055] Example 1

[0056] like Figure 1 and Figure 3 As shown, this embodiment provides an intelligent ear disease recognition method applicable to multiple types of devices, including:

[0057] S1. Input an input image to be detected; wherein the input image is a pre-established ear lesion image dataset;

[0058] S2. Perform multi-class feature extraction on the input image to obtain a multi-class feature image, and then fuse the image after weight assignment to obtain a fused image; wherein the multi-class feature extraction includes: high-level semantic feature extraction and low-level semantic feature extraction, and the multi-class feature image includes: a first feature information map after high-level semantic feature extraction and a second feature information map after low-level semantic feature extraction;

[0059] Further, such as Figure 4 As shown, the S2 method includes:

[0060] S2.1. Obtain an input image. Normalize the input image if it meets the preset conditions and generate it as a preprocessed image input S2.3. Otherwise, proceed to S2.2. The preset condition is that the input image size is 3×224×224, and the normalization is to use the mean [0.485, 0.456, 0.406] and standard deviation [0.229, 0.224, 0.225] to normalize the three channels of the scaled input image.

[0061] S2.2, preprocessing the input image to obtain a preprocessed image;

[0062] Furthermore, the method of S2.2 includes:

[0063] S2.2.1. Scale each input image to 3×224×224 dimensions.

[0064] S2.2.2. The three channels of the scaled input image are normalized using the mean [0.485, 0.456, 0.406] and standard deviation [0.229, 0.224, 0.225] and generated as the preprocessed image.

[0065] S2.3, performing multiple downsampling on the preprocessed image and generating a downsampled image;

[0066] Furthermore, the S2.3 method includes:

[0067] S2.3.1. Perform two 3×3 convolutions on the preprocessed image to obtain a first downsampled image of 24×112×112 dimensions after downsampling twice.

[0068] In step S2.3.1, the first downsampled image is batch normalized and then rectified using a linear rectification function. Finally, the image is subjected to two maximum pooling operations to obtain a 24×56×56 downsampled image. The number of downsampling operations in step S2.3 is preferably four, with the four downsampling operations comprising two 3×3 convolutions and two maximum pooling operations.

[0069] S2.4. After performing multi-stage downsampling on the downsampled image, a first feature map is finally obtained; wherein, the number of multi-stage downsampling in S2.4 is 3 times, the number of downsampling in each stage is 2, and feature extraction is performed after each stage of downsampling.

[0070] Furthermore, the S2.4 method includes:

[0071] S2.4.1. Perform a second downsampling on the downsampled image to obtain a second feature map. Specifically, downsample the 24×56×56-dimensional downsampled image twice to obtain a 48×28×28-dimensional second feature map.

[0072] S2.4.2. Perform the third-stage downsampling on the second feature map to obtain a third-stage image. Specifically, perform the third-stage downsampling on the 48×28×28-dimensional second-stage image twice to obtain a 96×14×14-dimensional third-stage image.

[0073] S2.4.3. Perform the fourth downsampling on the third-stage image to obtain a first feature map. Specifically, the 96×14×14-dimensional third-stage image is downsampled twice to obtain a 192×7×7-dimensional first feature map.

[0074] S2.5. Perform feature extraction on the first feature map to obtain a first feature information map after high-level semantic feature extraction;

[0075] Furthermore, the S2.5 method includes:

[0076] The first feature map is processed sequentially through a point-by-point convolution layer, a batch normalization layer, and a linear rectifier layer to produce the first feature information map. The point-by-point convolution layer converts the 192×7×7-dimensional second feature map to a 384×7×7-dimensional map. Specifically, at the end of the optimal ear network model backbone, the first feature map output after the fourth stage of downsampling undergoes final high-level semantic feature extraction using a convolutional layer with a kernel size of 3x3. Batch normalization and ReLU activation are then performed before outputting the first feature information map.

[0077] S2.6. Obtain a second feature map obtained by the second stage of downsampling in S2.4, and perform feature extraction on the second feature map to obtain a second feature information map after low-level semantic feature extraction;

[0078] Furthermore, the S2.6 method includes the following steps: The second feature map obtained by the second stage of downsampling in S2.4 is the second feature map obtained in S2.4.1. This second feature map is then processed sequentially through multiple layers of average pooling (Avg Pool), pointwise convolution (PWConv), regularization (Drop Block), efficient channel attention (ECA), batch normalization (Batch Normal), and linear rectification (ReLU) to obtain the second feature information map. The number of average pooling layers in S2.6 is four, and after four layers of average pooling and one pointwise convolution, the 48×28×28-dimensional second feature map is converted to 384×7×7 dimensions. Specifically, the second feature map is not directly passed to the local-global spatial feature fusion unit (LGSFF), but is instead processed through a separate branch path. First, an average pooling layer and a pointwise convolution layer are applied to resize the second feature map from 48×28×48 to 384×7×7. Then, a regularization layer is used to regularize the feature map. Afterwards, Efficient Channel Attention (ECA) is applied to enable the network to focus more on the low-level feature information in the second feature map. Batch normalization and linear rectification function activation functions are applied to the second feature map features.

[0079] Furthermore, if Figure 6 As shown, the image input to the efficient channel attention is the efficient channel first image. The efficient channel first image is globally average pooled and then passes through the first activation function layer to obtain the high school channel second image. The efficient channel first image and the high school channel second image enter the multiplication layer, and the image output by the multiplication layer is the image output by the efficient channel attention.

[0080] S2.7. Perform feature fusion on the first feature information map and the second feature information map to obtain a fused image.

[0081] Furthermore, if Figure 7 As shown, the S2.7 feature fusion method includes:

[0082] S2.7.1. Add elements of the first feature information map and the second feature information map to obtain a first fused feature map. Specifically, add elements of the first feature information map having a dimension of 384 × 7 × 7 and elements of the second feature information map having a dimension of 384 × 7 × 7 to obtain a first fused feature map having a dimension of 384 × 7 × 7.

[0083] S2.7.2. Extract a first weight and a second weight of the first fused feature map;

[0084] Furthermore, the S2.7.2 method includes:

[0085] S2.7.2.1. Apply grouped pointwise convolution, batch normalization, and linear rectification to the first fused feature map of 384 × 7 × 7 dimensions to obtain a second fused feature map of 384 × 7 × 7 dimensions.

[0086] S2.7.2.2. Combine the 1×7×7-dimensional maximum-pooled second fused feature map and the 1×7×7-dimensional average-pooled second fused feature map (cat) according to the channel dimension to obtain a 2×7×7-dimensional third fused feature map.

[0087] S2.7.2.3. Perform a 3×3 convolution on the third fused feature map (2×7×7 dimensions) and then combine it with the second fused feature map (384×7×7 dimensions) to form a 1×7×7 fourth fused feature map.

[0088] S2.7.2.4. Perform weight calculation on the fourth fused feature map of 1×7×7 dimensions to obtain a first weight and a second weight.

[0089] S2.7.3. Element-wise multiplication of the first weight and the second feature information map is performed based on the multiplication layer to obtain a first weighted fusion feature map.

[0090] S2.7.4. Perform element-wise multiplication of the second weight and the regularized first feature information map based on the multiplication layer to obtain a second weighted fusion feature map;

[0091] S2.7.5. Based on the addition layer, the first weighted fusion feature map and the second weighted fusion feature map are element-wise added to obtain a fused image.

[0092] In S2.7, the size of the first and second feature information maps is 384×7×7. The local-global spatial feature fusion unit first performs an addition layer to add the elements of the first and second feature information maps. This is then processed sequentially through grouped pointwise convolution (GPWconv), batch normalization (BatchNormal), and linear rectification (ReLU) to produce a second fused feature map. To enhance the model's perception of spatial information in the input image and improve model performance, a spatial attention module (Woo et al., 2018) is used to process the second fused feature map and obtain a third fused feature map. An activation function (Sigmod) is then calculated based on the third fused feature map to obtain activation weights (first and second weights). The first weight is then multiplied by the second feature information map, and the second weight is multiplied by the first feature information map. Before multiplying the first feature information map by the second weight, a regularization (Dropblock) technique is applied to the first feature information map. Dropblock randomly discards some blocks, that is, groups of adjacent feature map units, instead of randomly discarding individual neurons like Dropout, and forces the model to use other non-discarded areas to learn image features; Dropblock provides stronger regularization, improving the robustness and generalization ability of the model. Finally, the first feature information map and the second feature information map multiplied by the corresponding weights are added to obtain the output, denoted as Out. The output dimension of EFLL, that is, the output dimension of the fused image, is the same as that of the first feature information map and the second feature information map, which is 384×7×7. The entire process of the local-global spatial feature fusion unit is summarized as formula (1), formula (2), and formula (3):

[0093]

[0094] SAout=Conv 3×3 (Cat(Maxpool(Fmerge);Averagepool(Fmerge)))(2)

[0095]

[0096] Where: Fmerge represents the second fusion feature map, Rule represents the linear rectification function, BN represents batch normalization, GPWconv represents grouped point-by-point convolution, Flow represents the second feature information map, represents the addition layer, Fhigh represents the first feature information map, SAout represents the third fusion feature map, Conv 3×3Represents 3×3 convolution, Cat represents the concatenation of feature maps according to the channel dimension, Maxpool represents maximum pooling, Averagepool represents average pooling, Out represents the output of the local-global spatial feature fusion unit, Sigmod represents the activation function, represents the multiplication layer, and DropBlock represents regularization.

[0097] Furthermore, the method also includes: S2.8, calculating the final loss value of the optimal ear net model based on the image and loss function obtained in any one or more steps of S2.1-S2.7.

[0098] Furthermore, the S2.8 method includes:

[0099] S2.8.1. Calculate the loss of the second feature map to obtain a first loss value.

[0100] S2.8.2. Calculate the loss of the third-stage image to obtain a second loss value;

[0101] S2.8.3. Calculate the loss of the fused image to obtain a third loss value.

[0102] S2.8.4. Perform a final loss value calculation on the first loss value, the second loss value, and the third loss value based on the loss function to obtain a final loss value for optimizing the optimal ear net model.

[0103] Among them, the first-class head function, the second-class head function, and the third-class head function all adopt the head-like function model, which adopts the input layer, linear layer, normalized reference function layer, and output layer set in sequence, such as Figure 5 Specifically, multiple auxiliary classification heads are added to the optimal ear network model. The first, second, and third class head functions are all auxiliary classification heads. Compared to a single loss, using multiple auxiliary classification heads introduces additional losses from shallow layers of the network, which increases the gradient signal propagated during backpropagation. This provides additional regularization, improves the robustness of the model, and helps the network better converge and optimize its parameters.

[0104] Furthermore, the method further includes S2.9, performing prediction based on a result obtained by the prediction function through the third type header function.

[0105] S3. Output the fused image.

[0106] Beneficial effects of this embodiment:

[0107] This embodiment provides an intelligent ear disease recognition method applicable to multiple types of devices, constructs a large-scale data set, improves the richness of samples, performs multi-level class feature extraction on the input image, obtains multi-level class feature images, and then fuses them after weight assignment to obtain a fused image. The fidelity of the fused image is improved by the multi-level pixel dimensions and multi-weight fusion factors before the image is fused.

[0108] This embodiment provides a novel local-global spatial feature fusion module that simultaneously captures global and local spatial information and guides the network to focus on key areas in feature maps at all levels, thereby alleviating the problem of low precision. It is improved using a local-global spatial feature fusion unit and multiple auxiliary loss accumulation strategies to achieve a balance between recognition performance, average FPS, and model parameter size.

[0109] This embodiment adds multiple class head function sub-units in the hidden layer as auxiliary classification heads to calculate the final fused image and the loss values of each downsampling stage. Compared with calculating a single loss, using multiple auxiliary classification heads will introduce additional losses from the shallow layers of the network, which will increase the gradient signal propagated during the backpropagation process; this provides additional regularization, improves the robustness of the model, and helps the network better converge and optimize its parameters.

[0110] In this embodiment, the network uses multiple auxiliary classification heads for effective parameter optimization. With 0.77M parameters, the optimal ear network model achieves an average frame rate of 80 frames per second on a CPU. This system can be deployed on common cloud servers, mobile phones, tablets, and PC electronic devices. By operating a small electronic otoscope, users can use real-time video to fully scan and identify the ear canal, providing a new paradigm for otoendoscopy and other medical endoscopic image recognition applications.

[0111] This embodiment is applied to edge computing-based mobile terminals, cloud server-based mobile terminals, edge computing-based tablet terminals, and edge computing-based PC terminals. This embodiment designs a model that can be easily used in various real-world scenarios and develops corresponding ear lesion intelligent recognition models. To achieve this, the largest ear lesion image dataset ever was first established. Then, drawing on the lightweight network framework as the basic model, a new image feature fusion module called ear lesion feature fusion module (ELFF) was proposed. This network achieves a good trade-off between recognition performance, inference speed, and model parameter size. This embodiment demonstrates ultra-fast inference speed and extremely small model parameter size, making it easy to deploy in different scenarios.

[0112] Importantly, this embodiment demonstrates excellent recognition performance for a wide range of ear diseases, as well as normal ears, and has been validated across diverse populations of varying genders, age groups, and clinical settings. Finally, an application was developed, consisting of four versions: an edge computing-based mobile version, a cloud server-based mobile version, an edge computing-based tablet version, and an edge computing-based PC version. The mobile version is used on mobile devices, the tablet version on tablets, and the PC version on PCs, primarily assisting ordinary users with self-identification and regular checkups. With the help of a compact electronic otoscope, users can identify their ears promptly when they experience discomfort. Simultaneously, the tablet and PC versions can assist healthcare professionals in identifying patients in a variety of settings, including community screenings, community clinics, resource-limited areas, and specialized ENT practices. This helps reduce misdiagnosis, improve recognition efficiency, enhance the patient experience, and reduce the significant human and material costs associated with traditional recognition. Furthermore, a gradient CAM method was employed to visualize the optimal decision-making process of this system. This allows users and physicians to better understand the decision-making process of the deep learning model and helps developers more effectively validate, debug, and improve the model's behavior.

[0113] Example 2

[0114] like Figure 2 As shown, this embodiment provides an intelligent ear disease recognition system applicable to multiple types of devices, which is used to implement the intelligent ear disease recognition method applicable to multiple types of devices described in any one of the first embodiments. The system includes:

[0115] An image acquisition module is used to acquire an ear lesion image dataset, preprocess the ear lesion image dataset, and obtain a preprocessed image;

[0116] A model building module is used to build an intelligent recognition model for ear diseases, wherein the intelligent recognition model includes: a downsampling module, a feature extraction module and an ear injury feature fusion module;

[0117] A downsampling module, configured to downsample the preprocessed image to obtain a first feature map and a second feature map;

[0118] A feature extraction module is configured to extract high-level semantic features from the first feature map to obtain a first feature information map; and extract low-level semantic features from the second feature map to obtain a second feature information map;

[0119] The ear injury feature fusion module is used to fuse the first feature information map and the second feature information map to obtain a fused image; based on the fused image, intelligent ear disease recognition is achieved.

[0120] This embodiment has all the advantages of the first embodiment.

[0121] Implementation Three

[0122] This embodiment provides an electronic device, comprising: a processor and a memory storing computer program instructions; when the processor executes the computer program instructions, the intelligent ear disease recognition method applicable to multiple types of devices described in Example 1 is implemented.

[0123] The above are merely preferred embodiments of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

Claims

1. An intelligent ear disease recognition method applicable to multiple types of devices, characterized by: The following steps are involved: Acquire an ear lesion image dataset, and preprocess the ear lesion image dataset to obtain a preprocessed image; Constructing an intelligent recognition model for ear diseases, wherein the intelligent recognition model includes: a downsampling module, a feature extraction module, and an ear injury feature fusion module; Downsampling the preprocessed image based on the downsampling module to obtain a first feature map and a second feature map; Based on the feature extraction module, high-level semantic feature extraction is performed on the first feature map to obtain a first feature information map; low-level semantic feature extraction is performed on the second feature map to obtain a second feature information map; The process of obtaining the first characteristic information graph includes: The first feature map is processed sequentially through a point-by-point convolution layer, a batch normalization layer, and a linear rectification function layer to obtain a first feature information map; The process of obtaining the second characteristic information graph includes: The second feature map is processed by multiple average pooling, point-by-point convolution, regularization, efficient channel attention, batch normalization, and linear rectification function to obtain the second feature information map; The process of efficient channel attention processing includes: The image input to the efficient channel attention is the efficient channel first image. The efficient channel first image is sequentially subjected to global average pooling and the first activation function layer processing to obtain the efficient channel second image. The efficient channel first image and the efficient channel second image are fed into the multiplication layer to obtain the multiplication layer output image, which is the efficient channel attention output image. Based on the ear injury feature fusion module, the first feature information map and the second feature information map are subjected to feature fusion to obtain a fused image; and based on the fused image, intelligent ear disease recognition is achieved; The process of obtaining the fused image includes: Adding elements of the first feature information map and the second feature information map to obtain a first fused feature map; Extracting a first weight and a second weight of the first fused feature map; Based on the first multiplication layer, the first weight and the second feature information map are element-wise multiplied to calculate a first weighted fusion feature map; Based on the second multiplication layer, the second weight and the first feature information map are element-wise multiplied to calculate a second weighted fusion feature map; Based on the addition layer, the first weighted fusion feature map and the second weighted fusion feature map are element-wise added to obtain a fused image.

2. The method for intelligently identifying ear diseases applicable to multiple types of devices according to claim 1, characterized in that: The process of obtaining the first feature map and the second feature map includes: Downsampling the preprocessed image multiple times to generate a downsampled image; The downsampled image is subjected to multi-stage downsampling to obtain a first feature map and a second feature map.

3. The method for intelligently identifying ear diseases applicable to multiple types of devices according to claim 2, characterized in that: The process of performing multi-stage downsampling on the downsampled image includes: Performing a second stage of downsampling on the downsampled image to obtain a second feature map; Performing a third-stage downsampling on the second feature map to obtain a third-stage image; The third stage image is downsampled in the fourth stage to obtain the first feature map.

4. The method for intelligently identifying ear diseases applicable to multiple types of devices according to claim 3, characterized in that: Also includes: Performing loss calculation on the second feature map to obtain a first loss value; Performing loss calculation on the third-stage image to obtain a second loss value; performing loss calculation on the fused image to obtain a third loss value; Based on the loss function, the first loss value, the second loss value, and the third loss value are calculated to obtain a final loss value for optimizing the intelligent recognition model.

5. An intelligent ear disease recognition system applicable to multiple types of equipment, characterized by: A system for implementing the method for intelligently identifying ear diseases applicable to multiple types of devices according to any one of claims 1 to 4, comprising: An image acquisition module is used to acquire an ear lesion image dataset, preprocess the ear lesion image dataset, and obtain a preprocessed image; A model building module is used to build an intelligent recognition model for ear diseases, wherein the intelligent recognition model includes: a downsampling module, a feature extraction module and an ear injury feature fusion module; A downsampling module, configured to downsample the preprocessed image to obtain a first feature map and a second feature map; A feature extraction module is configured to extract high-level semantic features from the first feature map to obtain a first feature information map; and extract low-level semantic features from the second feature map to obtain a second feature information map; The ear injury feature fusion module is used to fuse the first feature information map and the second feature information map to obtain a fused image; based on the fused image, intelligent ear disease recognition is achieved.

6. An electronic device, characterized in that: The device includes: a processor and a memory storing computer program instructions; when the processor executes the computer program instructions, it implements the intelligent ear disease recognition method applicable to multiple types of devices as described in any one of claims 1 to 4.

Images