Medical image map construction method and system, storage medium and electronic equipment

Abstract

The invention provides a medical image map construction method and system, a storage medium and electronic equipment. The method comprises the following steps: acquiring medical images of a plurality of individuals; extracting high-dimensional features of the medical image based on an encoder; performing feature fusion on the high-dimensional features, and calculating a feature center of the high-dimensional features after feature fusion; and converting the feature center into a medical image map based on a decoder. According to the medical image atlas construction method and system, the storage medium and the electronic equipment, the efficiency of medical image atlas construction is effectively improved based on the deep learning algorithm, and the accuracy and applicability of the medical image atlas are enhanced.

Description

Technical Field

[0001] This invention belongs to the technical field of deep learning, and in particular relates to a method, system, storage medium and electronic device for constructing medical image atlases. Background Technology

[0002] In the field of medical image analysis, constructing accurate atlases is crucial for understanding human anatomy, disease diagnosis, and treatment planning. While technological advancements have made the acquisition of medical imaging data more efficient and precise, they have also brought challenges to data processing and analysis. To better understand, diagnose, and treat diseases, constructing accurate medical image atlases has become a critical requirement. Atlases not only provide standardized anatomical references for medical research but also help improve the accuracy of disease diagnosis and the effectiveness of treatment plans.

[0003] However, existing atlas construction methods have limitations that restrict their application across different medical imaging modalities and organs. Population-averaged atlases are currently the most commonly used atlases, used to represent anatomical variations among multiple subjects. The construction of population-averaged atlases typically involves two steps: iteratively registering population images to a common space, and then averaging these registered images pixel by pixel. To obtain a clearer atlas, iterative registration and pixel averaging are often necessary. Current techniques for constructing medical image atlases still primarily rely on image registration and pixel-level averaging methods, which suffer from the following technical problems:

[0004] (1) Limitations of image registration: Traditional image registration methods are difficult to accurately align complex or small anatomical structures, especially under different imaging conditions.

[0005] (2) Individual differences: There are differences in the anatomical structure between different individuals in the population. Even after registration, individual differences in anatomical structure still exist.

[0006] (3) Information loss due to averaging: pixel-level averaging may blur the anatomical differences between individuals, leading to a decrease in the representativeness and accuracy of the atlas.

[0007] (4) Lack of feature-level optimization: Existing methods do not optimize the map at the feature level, and cannot make full use of the advantages of advanced technologies such as deep learning.

[0008] (5) Lack of application of advanced deep learning algorithms: With the development of deep learning technology, its application in medical image analysis is becoming more and more widespread. Deep learning methods can efficiently capture and generate image information and have potential in map construction.

[0009] With the development of deep learning technology, its application in medical image analysis is becoming increasingly widespread. Deep learning models, such as Convolutional Neural Networks (CNNs) and Visual Transformers (ViTs), have demonstrated superior performance in image recognition, classification, and segmentation tasks. These models can automatically learn complex feature representations of images, providing new possibilities for medical image analysis. Deep learning methods can effectively solve the following problems:

[0010] (1) Deep learning algorithms can learn more representative features from large-scale samples and better reflect the general anatomical morphology of the population.

[0011] (2) Deep learning algorithms can more flexibly model nonlinear relationships and accurately characterize the complexity of anatomical structures.

[0012] (3) Deep learning algorithms can achieve automated processing without manual intervention, and can efficiently process large amounts of medical image data, saving time and labor costs.

[0013] Overall, compared to previous methods, using deep learning to construct standardized brain atlases can better reflect the general brain morphology of a population and obtain more accurate, comprehensive, and representative standardized atlases. However, the application of deep learning technology in the field of atlas construction is still in its early stages. Existing research mainly uses generative methods to construct atlases, but these methods often lead to blurred images and a lack of realism when dealing with high-frequency deformable regions, and may not be able to fully capture the fine structural changes of complex or highly variable populations. Summary of the Invention

[0014] In view of the shortcomings of the prior art described above, the purpose of this invention is to provide a method, system, storage medium and electronic device for constructing medical image atlases, which effectively improves the efficiency of medical image atlas construction and enhances the accuracy and applicability of medical image atlases based on deep learning algorithms.

[0015] In a first aspect, the present invention provides a method for constructing a medical image atlas, the method comprising the following steps: acquiring medical images of multiple individuals; extracting high-dimensional features of the medical images based on an encoder; performing feature fusion on the high-dimensional features and calculating the feature centers of the fused high-dimensional features; and converting the feature centers into a medical image atlas based on a decoder.

[0016] In one implementation of the first aspect, the encoder is a CNN encoder or a ViT encoder.

[0017] In one implementation of the first aspect, the decoder employs a CNN decoder or a ViT decoder.

[0018] In one implementation of the first aspect, the encoder and the decoder are structurally symmetrically arranged to achieve opposite functions.

[0019] In one implementation of the first aspect, the feature center is the average position of the high-dimensional features after feature fusion.

[0020] In one implementation of the first aspect, the medical image is a medical image of a specific organ of the individual.

[0021] In one implementation of the first aspect, the medical image is an MRI image, a CT image, or a PET image.

[0022] In a second aspect, the present invention provides a medical image atlas construction system, the system comprising an acquisition module, an extraction module, a calculation module, and a construction module;

[0023] The acquisition module is used to acquire medical images of multiple individuals;

[0024] The extraction module is used to extract high-dimensional features of the medical image based on the encoder;

[0025] The calculation module is used to perform feature fusion on the high-dimensional features and calculate the feature center of the high-dimensional features after feature fusion.

[0026] The building module is used to convert the feature centers into medical image atlases based on the decoder.

[0027] Thirdly, the present invention provides an electronic device, the electronic device comprising: a processor and a memory;

[0028] The memory is used to store computer programs;

[0029] The processor is used to execute the computer program stored in the memory, so that the electronic device performs the above-described medical image atlas construction method.

[0030] Fourthly, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by an electronic device, implements the above-described method for constructing medical image atlases.

[0031] As described above, the medical image atlas construction method, system, storage medium, and electronic device of the present invention have the following beneficial effects:

[0032] (1) By mapping medical images to a high-dimensional feature space and calculating the center point, the accuracy and representativeness of the atlas are improved, while preserving individual anatomical differences, providing richer information for medical research and clinical applications;

[0033] (2) The graph construction model is placed on the feature space and the deep learning model (such as CNN and ViT) is used to automatically learn image features, which improves the efficiency and quality of graph construction and enhances the generalization ability of the graph. This not only saves time and labor costs significantly, but also reduces the errors that may be introduced by manual operation, and improves the overall efficiency and accuracy of graph construction.

[0034] (3) With the high adaptability and powerful feature extraction capability of deep learning models, it can process different imaging modal data including MRI, CT, PET, etc., and is widely applicable to image analysis of various important organs including the brain, heart, liver, etc., thereby realizing cross-organ and cross-modal medical image analysis and construction of general atlas;

[0035] (4) The constructed atlas can not only provide accurate anatomical references for medical research, but also serve as an information aid to help radiologists screen diseases more effectively in clinical practice, thus supporting personalized and precision medicine. Attached Figure Description

[0036] Figure 1 The flowchart shown is an embodiment of the medical image atlas construction method of the present invention;

[0037] Figure 2 The diagram shown is an architectural schematic of a medical image atlas construction method of the present invention in one embodiment;

[0038] Figure 3 A schematic diagram showing a fetal brain atlas from week 21 to week 38 of gestation in one embodiment;

[0039] Figure 4 The diagram shown is a structural schematic of the medical image atlas construction system of the present invention in one embodiment.

[0040] Figure 5 The diagram shown is a structural schematic of an embodiment of the electronic device of the present invention. Detailed Implementation

[0041] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that, unless otherwise specified, the following embodiments and features described therein can be combined with each other.

[0042] It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. Therefore, the drawings only show the components related to the present invention and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0043] The following embodiments of the present invention provide a method for constructing medical image atlases, which can be applied to electronic devices. The electronic devices described in this invention may include mobile phones with wireless charging capabilities, tablet computers, laptops, wearable devices, in-vehicle devices, augmented reality (AR) / virtual reality (VR) devices, ultra-mobile personal computers (UMPCs), netbooks, personal digital assistants (PDAs), etc. The embodiments of the present invention do not impose any limitations on the specific type of electronic device.

[0044] For example, the electronic device may be a station (STAION, ST) in a WLAN with wireless charging capability, a cellular phone, cordless phone, Session Initiation Protocol (SIP) phone, Wireless Local Loop (WLL) station, Personal Digital Assistant (PDA) device, handheld device with wireless charging capability, computing device or other processing device, computer, laptop computer, handheld communication device, handheld computing device, and / or other devices for communication over a wireless system, as well as next-generation communication systems, such as mobile terminals in 5G networks, mobile terminals in future evolved Public Land Mobile Networks (PLMNs), or mobile terminals in future evolved Non-terrestrial Networks (NTNs).

[0045] For example, the electronic device can communicate with networks and other devices wirelessly. The wireless communication can use any communication standard or protocol, including but not limited to Global System for Mobile Communication (GSM), General Packet Radio Service (GPRS), Code Division Multiple Access (CDMA), Wideband Code Division Multiple Access (WCDMA), Long Term Evolution (LTE), email, Short Messaging Service (SMS), BT, GNSS, WLAN, NFC, FM, and / or IR technologies. The GNSS can include Global Positioning System (GPS), Global Navigation Satellite System (GLONASS), BeiDou Navigation Satellite System (BDS), Quasi-Zenith Satellite System (QZSS), and / or Satellite Based Augmentation Systems (SBAS).

[0046] The technical solutions of the present invention will now be described in detail with reference to the accompanying drawings.

[0047] like Figure 1 and Figure 2 As shown, in one embodiment, the medical image atlas construction method of the present invention includes steps S1-S4.

[0048] Step S1: Acquire medical images of multiple individuals.

[0049] Specifically, medical images of multiple individuals are acquired. These medical images can be MRI, CT, or PET images. Furthermore, the medical images are specific to a particular organ of each individual, such as brain, heart, or liver images. Therefore, the medical image atlas construction method of this invention enables cross-organ, cross-modal medical image analysis and the construction of a universal atlas. This not only provides precise anatomical references for medical research but also serves as an information support tool, helping radiologists screen diseases more effectively in clinical practice and supporting personalized and precision medicine.

[0050] Step S2: Extract high-dimensional features of the medical image based on the encoder.

[0051] Specifically, the medical images of the multiple individuals are input into an encoder to obtain high-dimensional features for each medical image, thereby avoiding the inaccurate alignment problem of complex or minute anatomical structures in traditional methods. This invention operates directly at the feature level through deep learning, rather than relying on pixel-level image registration and averaging, effectively solving the limitations of traditional image registration methods when aligning complex anatomical structures.

[0052] In one embodiment, the encoder is a CNN (Convolutional Neural Networks) encoder or a ViT (Vision Transformer) encoder.

[0053] Step S3: Perform feature fusion on the high-dimensional features and calculate the feature center of the high-dimensional features after feature fusion.

[0054] Specifically, the high-dimensional features are fed into a feature fusion module for feature fusion. This feature fusion not only comprehensively considers the anatomical differences between individuals but also ensures that individual differences are effectively represented through optimized algorithms, thus avoiding the registration inaccuracies caused by individual differences in traditional methods. After feature fusion, the average position of the fused high-dimensional features, i.e., the feature center, is calculated to determine the position of the atlas in the feature space. The feature center not only considers the statistical information of the image group but also effectively preserves the anatomical differences between individuals through the feature representation capabilities of deep learning. This method solves the information loss problem that may occur during traditional averaging, ensuring that the atlas can truly and accurately reflect the anatomical features of the group.

[0055] Step S4: Convert the feature centers into medical image atlases based on the decoder.

[0056] Specifically, the decoder is used to transform the feature center back into image space, thereby generating the medical image atlas. The medical image atlas is a group statistical average representation of the input medical images, located at the center in the feature space, and can be used in fields such as medical research, disease diagnosis, and education.

[0057] In one embodiment, the decoder is a CNN decoder or a ViT decoder. It should be noted that the structure of the decoder is symmetrical to that of the encoder, but its function is opposite. It maps the high-dimensional features back to the original image dimensions, constructing a map representing the group's features, effectively ensuring the map's high accuracy and high representativeness.

[0058] It should be noted that a loss function is introduced during the training of the encoder and decoder to optimize the quality of the spectrogram, ensuring that the spectrogram maintains consistency with the input image population in terms of structure and intensity. This method fully leverages the feature-level optimization capabilities of deep learning, addressing the lack of feature-level optimization in existing methods.

[0059] The medical image atlas construction method of the present invention will be further illustrated below through specific embodiments.

[0060] Taking the construction of fetal brain MRI atlases as an example, firstly, fetal brain MRI images from multiple individuals are mapped to a high-dimensional feature space using a deep learning encoder. The encoder processes each fetal MRI image to extract high-dimensional features reflecting the complex anatomical structure of the fetal brain. Next, the extracted high-dimensional features are fed into a feature fusion module to comprehensively consider the differences in anatomical structure among different fetuses. An optimized algorithm ensures that individual differences are effectively represented, avoiding registration inaccuracies caused by individual fetal differences in traditional methods. Then, the average position of all extracted high-dimensional features, i.e., the feature center, is calculated to determine the position of the atlas in the feature space. Finally, a decoder transforms the feature center back into image space to generate the fetal brain atlas. Figure 3 The image shows the fetal brain atlas from week 21 to week 38 of gestation, which has a clear anatomical structure and successfully captures the changes in the fetal brain's anatomical structure as the gestational age increases.

[0061] In the large dataset collected in this experiment, the present invention outperforms traditional methods in both Peak Signal-to-Noise Ratio (PSNR) and Structure Similarity Index Measure (SSIM), demonstrating better image quality and structural detail. Through visual comparison, the atlases generated by the present invention are closer to real anatomical structures in terms of detail and overall morphology.

[0062] The scope of protection of the medical image atlas construction method described in this embodiment is not limited to the execution order of the steps listed in this embodiment. Any solution implemented by adding, subtracting, or replacing steps in the prior art based on the principles of this invention is included within the scope of protection of this invention.

[0063] This invention also provides a medical image atlas construction system, which can implement the medical image atlas construction method described in this invention. However, the implementation device of the medical image atlas construction system described in this invention includes, but is not limited to, the structure of the medical image atlas construction system listed in this embodiment. All structural modifications and substitutions of the prior art made in accordance with the principles of this invention are included within the protection scope of this invention.

[0064] like Figure 4 As shown, in one embodiment, the medical image atlas construction system of the present invention includes an acquisition module 41, an extraction module 42, a calculation module 43, and a construction module 44.

[0065] The acquisition module 41 is used to acquire medical images of multiple individuals.

[0066] The extraction module 42 is connected to the acquisition module 41 and is used to extract high-dimensional features of the medical image based on the encoder.

[0067] The calculation module 43 is connected to the extraction module 42 and is used to perform feature fusion on the high-dimensional features and calculate the feature center of the high-dimensional features after feature fusion.

[0068] The construction module 44 is connected to the calculation module 43 and is used to convert the feature center into a medical image atlas based on the decoder.

[0069] The structure and principle of the acquisition module 41, extraction module 42, calculation module 43 and construction module 44 correspond one-to-one with the steps in the above-mentioned medical image atlas construction method, so they will not be described again here.

[0070] In the embodiments provided by this invention, it should be understood that the disclosed systems, apparatuses, or methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For instance, the division of modules / units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or units may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection of apparatuses or modules or units may be electrical, mechanical, or other forms.

[0071] The modules / units described as separate components may or may not be physically separate. The components shown as modules / units may or may not be physical modules; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules / units can be selected to achieve the objectives of the embodiments of the present invention, depending on actual needs. For example, the functional modules / units in the various embodiments of the present invention may be integrated into one processing module, or each module / unit may exist physically separately, or two or more modules / units may be integrated into one module / unit.

[0072] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0073] This invention also provides a computer-readable storage medium. Those skilled in the art will understand that all or part of the steps in the methods of the above embodiments can be implemented by a program instructing a processor. The program can be stored in a computer-readable storage medium, which is a non-transitory medium, such as random access memory, read-only memory, flash memory, hard disk, solid-state drive, magnetic tape, floppy disk, optical disk, and any combination thereof. The storage medium can be any available medium accessible to a computer or a data storage device such as a server or data center that integrates one or more available media. This available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., digital video disc (DVD)), or a semiconductor medium (e.g., solid-state drive (SSD)).

[0074] This invention also provides an electronic device. The electronic device includes a processor and a memory.

[0075] The memory is used to store computer programs.

[0076] The memory includes various media capable of storing program code, such as ROM, RAM, magnetic disk, USB flash drive, memory card, or optical disk.

[0077] The processor is connected to the memory and is used to execute the computer program stored in the memory so that the electronic device performs the above-described medical image atlas construction method.

[0078] Preferably, the processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0079] like Figure 5 As shown, the electronic device of the present invention is embodied in the form of a general-purpose computing device. The components of the electronic device may include, but are not limited to: one or more processors or processing units 51, a memory 52, and a bus 53 connecting different system components (including the memory 52 and the processing unit 51).

[0080] Bus 53 represents one or more of several bus architectures, including a memory bus or memory controller, a peripheral bus, a graphics acceleration port, a processor, or a local bus using any of the various bus architectures. Examples of these architectures include, but are not limited to, the Industry Standard Architecture (ISA) bus, the Micro Channel Architecture (MAC) bus, the Enhanced ISA bus, the Video Electronics Standards Association (VESA) local bus, and the Peripheral Component Interconnect (PCI) bus.

[0081] Electronic devices typically include a variety of computer-readable media. These media can be any available media that can be accessed by the electronic device, including volatile and non-volatile media, and removable and non-removable media.

[0082] Memory 52 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 521 and / or cache memory 522. The electronic device may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, storage system 523 may be used to read and write non-removable, non-volatile magnetic media (…). Figure 5 Not shown; usually referred to as a "hard drive"). Although Figure 5Not shown, a disk drive for reading and writing to a removable non-volatile disk (e.g., a "floppy disk") and an optical disk drive for reading and writing to a removable non-volatile optical disk (e.g., a CD-ROM, DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to bus 53 via one or more data media interfaces. Memory 52 may include at least one program product having a set (e.g., at least one) of program modules configured to perform the functions of the embodiments of the present invention.

[0083] A program / utility 524 having a set (at least one) of program modules 5241 may be stored, for example, in memory 52. ​​Such program modules 5241 include, but are not limited to, an operating system, one or more application programs, other program modules, and program data. Each or some combination of these examples may include an implementation of a network environment. Program modules 5241 typically perform the functions and / or methods described in the embodiments of the present invention.

[0084] The electronic device can also communicate with one or more external devices (e.g., keyboard, pointing device, display, etc.), one or more devices that enable a user to interact with the electronic device, and / or any device that enables the electronic device to communicate with one or more other computing devices (e.g., network interface card, modem, etc.). This communication can be performed through input / output (I / O) interface 54. Furthermore, the electronic device can also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN), and / or public networks, such as the Internet) through network adapter 55. Figure 5 As shown, network adapter 55 communicates with other modules of the electronic device via bus 53. It should be understood that, although not shown in the figure, other hardware and / or software modules can be used in conjunction with the electronic device, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0085] The above embodiments are merely illustrative of the principles and effects of the present invention and are not intended to limit the invention. Any person skilled in the art can modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by those skilled in the art without departing from the spirit and technical concept disclosed in the present invention should still be covered by the claims of the present invention.

Claims

1. A method for constructing a medical image atlas, characterized in that, The method includes the following steps: Acquire medical images of multiple individuals; High-dimensional features of the medical images are extracted based on the encoder; The high-dimensional features are fused, and the feature centers of the fused high-dimensional features are calculated. The feature centers are converted into medical image atlases based on the decoder.

2. The method for constructing medical image atlases according to claim 1, characterized in that: The encoder is either a CNN encoder or a ViT encoder.

3. The method for constructing medical image atlases according to claim 1, characterized in that: The decoder is either a CNN decoder or a ViT decoder.

4. The method for constructing medical image atlases according to claim 1, characterized in that: The encoder and the decoder are symmetrically arranged to achieve opposite functions.

5. The method for constructing medical image atlases according to claim 1, characterized in that: The feature center is the average position of the high-dimensional features after feature fusion.

6. The method for constructing medical image atlases according to claim 1, characterized in that: The medical image refers to a medical image of a specific organ of the individual.

7. The method for constructing medical image atlases according to claim 1, characterized in that: The medical images used are MRI images, CT images, or PET images.

8. A medical image atlas construction system, characterized in that, The system includes an acquisition module, an extraction module, a calculation module, and a construction module; The acquisition module is used to acquire medical images of multiple individuals; The extraction module is used to extract high-dimensional features of the medical image based on the encoder; The calculation module is used to perform feature fusion on the high-dimensional features and calculate the feature center of the high-dimensional features after feature fusion. The building module is used to convert the feature centers into medical image atlases based on the decoder.

9. An electronic device, characterized in that, The electronic device includes: a processor and a memory; The memory is used to store computer programs; The processor is used to execute the computer program stored in the memory to cause the electronic device to perform the medical image atlas construction method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by an electronic device, the program implements the medical image atlas construction method as described in any one of claims 1 to 7.

Images