Human cerebral hemorrhage DTI image matching method and system based on multi-modal fusion enabling CT

Abstract

The invention provides a human cerebral hemorrhage DTI image matching method and system based on multi-modal fusion enabling CT, and relates to the technical field of medical image processing analysis. A multi-modal feature database is constructed according to physiological parameter information, skull feature information and DTI images of a plurality of cerebral apoplexy individuals; according to the physiological parameter information and the CT image of the target individual, through a first difference index and a second difference index, a candidate individual is rapidly and accurately matched from a multi-modal feature database, and the first difference index is used for representing the difference degree between skull feature information; the second difference index is used for representing the difference degree between the physiological parameter information. According to the method, the DTI image of the candidate individual is determined as the DTI image of the target individual, so that self limitation and equipment limitation of the target individual can be eliminated, the DTI image can be rapidly determined only through the CT image of the target individual, the situation that the diagnosis and treatment time is affected due to acquisition delay of the DTI image can be avoided, and the diagnosis and treatment efficiency is improved.

Description

Technical Field

[0001] This invention relates to the field of medical image processing and analysis technology, and in particular to a method and system for matching human brain hemorrhage DTI images based on multimodal fusion-enabled CT. Background Technology

[0002] In the clinical diagnosis and treatment of neurological diseases, DTI (diffusion tensor imaging) technology has become an important tool for assessing the extent of damage to the corticospinal tract (CST) caused by diseases such as cerebral hemorrhage because it can visually present the course and integrity of white matter fiber tracts.

[0003] However, DTI scans typically require more than 30 minutes of acquisition time. Furthermore, due to the low coverage of current detection equipment and the limited availability of 3.0T high-field MRI, coupled with individual limitations such as physical function, metal implants, or claustrophobia, it is difficult to quickly obtain high-quality DTI images. This results in a long acquisition time, which seriously affects the timing of diagnosis and treatment.

[0004] Therefore, there is an urgent need for a method and system for matching DTI images of human brain hemorrhage based on multimodal fusion-enabled CT, to quickly and accurately acquire DTI images of the target individual, reduce acquisition time, and improve acquisition efficiency. This would avoid delays in DTI image acquisition from affecting diagnosis and treatment time, and improve diagnostic and treatment efficiency. Summary of the Invention

[0005] This invention provides a method and system for matching DTI images of human brain hemorrhage based on multimodal fusion-enabled CT. It can quickly and accurately match DTI images of a target individual based on a pre-built database, effectively eliminating limitations imposed by the individual's own conditions and equipment, reducing the time required to acquire DTI images, and improving acquisition efficiency. This avoids delays in DTI image acquisition affecting diagnosis and treatment time, thus improving diagnostic and treatment efficiency.

[0006] To achieve the above objectives, the embodiments of the present invention adopt the following technical solutions: Firstly, a method for matching DTI images of human brain hemorrhage based on multimodal fusion-enabled CT is provided. The method includes: acquiring physiological parameter information, CT images of the head region, and DTI images of multiple stroke individuals; the physiological parameter information includes age, sex, duration of illness, hematoma location, hematoma morphology, and hematoma volume; determining skull feature information for each stroke individual based on the CT images of the head region of the multiple stroke individuals; the skull feature information includes the thickness of different parts of the skull and the three-dimensional coordinates of each skull landmark in a preset Cartesian coordinate system; and constructing a model based on the skull feature information, physiological parameter information, and DTI images of the multiple stroke individuals. A multimodal feature database is established. Based on the physiological parameters and skull features of the target individual, the DTI images of candidate individuals are selected from the multimodal feature database as the DTI images of the target individual. The candidate individual is one of multiple stroke individuals. The first difference index between the target individual and the candidate individual is less than a preset threshold, and the second difference index between the target individual and the candidate individual is the minimum value among the multiple second difference indices between the target individual and the multiple stroke individuals. The first difference index is positively correlated with the degree of difference in skull features between the target individual and the candidate individual, and the second difference index is positively correlated with the degree of difference in physiological parameters between the target individual and the candidate individual.

[0007] The method provided by this invention constructs a multimodal feature database based on physiological parameters, skull features, and DTI images of multiple stroke individuals. It then rapidly and accurately matches candidate individuals from the multimodal feature database using a first difference index and a second difference index, based on the target individual's physiological parameters and CT images. The first difference index characterizes the degree of difference between skull features, while the second difference index characterizes the degree of difference between physiological parameters. Identifying candidate individuals' DTI images as the target individual's effectively eliminates limitations imposed by the target individual's own conditions and equipment, reduces the time required to acquire DTI images, and improves acquisition efficiency. This avoids delays in DTI image acquisition that could impact treatment time, thus improving diagnostic and treatment efficiency.

[0008] In one possible implementation of the first aspect, the different parts that make up the skull include the frontal bone, parietal bone, temporal bone, occipital bone and sphenoid bone, and the landmarks include the anterior point, posterior point, left temporal point, right temporal point, apex and base of the skull cavity, as well as the hemorrhage point, glabella point, anterior fontanelle point, chevron point, center of external auditory canal and external occipital protuberance.

[0009] The method provided by this invention, by defining the components and landmarks of the skull, can quickly and accurately determine the thickness of different parts of the skull and the three-dimensional coordinates of different landmarks. This can effectively reduce the ambiguity of feature extraction, achieve accurate acquisition of skull feature information of different individuals, and thus improve the matching degree between the DTI images of candidate individuals and the DTI images of target individuals, thereby improving the accuracy of matching.

[0010] In one possible implementation of the first aspect, determining the skull feature information of each stroke individual among multiple stroke individuals based on CT images of the head region of multiple stroke individuals includes: determining a binarized skull image of each stroke individual based on the CT images of the head region of multiple stroke individuals; determining a three-dimensional skull model corresponding to the binarized skull image of each stroke individual based on a three-dimensional reconstruction algorithm; determining the thickness of different parts of the skull based on the distance between each point corresponding to the three-dimensional contour of different parts of the skull on the three-dimensional skull model and the center point of the three-dimensional skull model; and determining the three-dimensional coordinates of each landmark point of the skull in a preset spatial rectangular coordinate system, wherein the preset spatial rectangular coordinate system has the center point of the three-dimensional skull model as the origin.

[0011] The method provided by this invention converts CT images into binary skull images and then establishes a three-dimensional skull model corresponding to each binary skull image. This ensures that the skull features of different individuals have a unified reference, eliminating the influence of positioning deviation on thickness calculation and landmark coordinate extraction. The method provided by this invention calculates skull thickness based on three-dimensional contours and combines it with precise landmark coordinates to achieve a quantitative description of skull morphology, providing a high-precision data foundation for subsequent difference calculations, thereby improving the reliability and accuracy of matching.

[0012] In one possible implementation of the first aspect, the formula for determining the first difference index D1 is: D1=α·D thickness +(1-α)·D landmark ; Where α is the overall weight coefficient, D thickness D is a score for differences in skull thickness. landmark Scoring for differences in landmark location; Score for differences in skull thickness (D) thickness The formula for determining it is: N represents the number of parts that make up the skull, T target,i T represents the thickness of the i-th part of the skull of the target individual; candidate,i The thickness of the i-th part of the skull that makes up an individual with a stroke; Marker location difference score D landmark The formula for determining it is: M is the number of markers, P target,j P represents the three-dimensional coordinates of the j-th marker point of the target individual; candidate,j Let j be the three-dimensional coordinates of the j-th landmark point in an individual with stroke, and normalize_factor be the normalize_factor. j This is the normalization factor.

[0013] The method provided by this invention decomposes skull feature differences into thickness differences and landmark position differences, and obtains a first difference index through weighted summation. Thickness differences are calculated by comparing the thickness of each skull segment between the target individual and the candidate individual, while landmark differences are evaluated based on Euclidean distance in three-dimensional coordinates, and a normalization factor is introduced to eliminate the influence of measurement scale. In this way, different dimensions of skull features contribute differently to the matching results, and separating thickness and landmark differences allows for refined evaluation; the weighted summation method assigns reasonable weights to both, enabling the method provided by this invention to comprehensively consider both morphological and spatial differences; and the normalization process ensures the comparability of difference scores for different landmarks, avoiding the over- or under-sighting of certain features due to scale issues, ultimately achieving scientific quantification and accurate matching of skull feature differences.

[0014] In one possible implementation of the first aspect, different parts of the skull are each configured with a first weighting coefficient, and each marker is configured with a second weighting coefficient. Skull thickness difference score D thickness The formula for determining it is: Marker location difference score D landmark The formula for determining it is: Among them, w i v is the first weighting coefficient for the i-th part that makes up the skull; i It is the second weight coefficient of the i-th marker.

[0015] The method provided by this invention assigns differentiated weights to different parts of the skull and landmarks, thereby improving the way to determine thickness differences and landmark position differences. When different skull regions and landmarks have different importance in DTI images, the weight configuration can enhance the matching priority of key structures and weaken the interference of secondary features. At the same time, the adjustability of the weights allows the algorithm to adapt to different use scenarios, improving flexibility and practicality.

[0016] In one possible implementation of the first aspect, the formula for determining the preset threshold μ is: μ = μ0·ε; Where μ0 is the skull difference benchmark threshold, and ε is the skull complexity weight; When the overall complexity index of the target user is less than 0.7, the skull complexity weight ε = 0.8; when the overall complexity index of the target user is greater than or equal to 0.7 and less than 1.3, the skull complexity weight ε = 1; when the overall complexity index of the target user is greater than 1.3, the skull complexity weight ε = 1.2. The formula for determining the overall complexity index v is: Where A is the balance coefficient, Dref is the dispersion reference value; CV thickness η is the coefficient of variation for skull thickness. thickness σ represents the standard deviation of the different parts that make up the skull. thickness Mean values ​​of different parts that make up the skull, DV landmark P represents the dispersion of skull landmarks. j Let P be the three-dimensional coordinates of the j-th marker point of the target user, and let P be the three-dimensional coordinates of the geometric center of all marker points of the target user.

[0017] The method provided by this invention can dynamically adjust a preset threshold based on the skull complexity of a target individual. By calculating the coefficient of variation of skull thickness and the dispersion of landmark points for the target individual, a comprehensive complexity index is constructed. Different skull complexity weights are then mapped according to the index interval, thereby achieving dynamic adjustment of the preset threshold for the target individual. The logic is that individual differences in skull morphology naturally exist, and a fixed threshold cannot meet the matching requirements of complex skull structures. The dynamic threshold mechanism quantifies skull complexity, applying a strict threshold to individuals with simple structures to ensure high matching accuracy, while appropriately relaxing the standard for complex individuals to avoid having no usable matching cases due to overly strict thresholds. This adaptive adjustment strategy expands the scope of application while ensuring matching quality, especially significantly improving the success rate and feasibility of matching when dealing with special individuals.

[0018] In one possible implementation of the first aspect, the formula for determining the second difference index D2 is: K represents the number of parameters included in the physiological parameter information; β k Score is the weighting coefficient of the k-th parameter included in the physiological parameter information. k The difference score for the k-th parameter included in the physiological parameter information; Physiological parameter information includes age difference score age for: A target A represents the age of the target individual. candidate For the age of the individual who suffered a stroke, Amax and A min The maximum and minimum ages of multiple stroke individuals; Physiological parameters include a score indicating gender differences. gender Score: When the target individual and the stroke patient are of the same sex, gender =0; Scoregender =1 when the target individual and the stroke patient are of different sexes; Physiological parameter information includes the score for the difference in the duration of illness. time for: T target The duration of illness for the target individual; T candidate The duration of onset for an individual with stroke; T max and T min The maximum and minimum duration of onset for multiple stroke individuals; Physiological parameters include a score for the difference in hematoma location. location Score: When the hematoma areas of the target individual and the stroke patient are located in the same region, location =0; Score = 0 when the hematoma areas of the target individual and the stroke individual are adjacent. location =0.5; Score = 0.5; when the hematoma areas of the target individual and the stroke individual are located in different and non-adjacent areas. location =1; Physiological parameters include a score for differences in hematoma morphology. shape for: L target L represents the length of the long axis of the hematoma in the target individual. candidate L represents the length of the long axis of the hematoma in an individual with stroke. max and L min The maximum and minimum lengths of the long axis of the hematoma are given for multiple stroke individuals; W target W represents the short axis length of the hematoma in the target individual. candidate W represents the short axis length of the hematoma in an individual with stroke. max and W min H represents the maximum and minimum short axis lengths of hematoma in multiple stroke individuals; target For the hematoma height of the target individual, H candidate The height of the hematoma in an individual with stroke; H max and H min The maximum and minimum hematoma heights are for multiple stroke patients. Physiological parameters include a score for the difference in hematoma volume. volume for: V target V represents the hematoma volume of the target individual; candidate This refers to the hematoma volume in an individual with a stroke.

[0019] The method provided by this invention quantifies the differences in physiological parameters such as age, gender, and duration of illness, and obtains a second difference index through weighted summation. Specifically, age and duration of illness are calculated using normalization to determine difference scores, gender is assessed using a 0-1 binary classification, and hematoma location is scored based on regional correlation grading. This achieves complementarity between physiological parameters and skull features, addressing the issue of poor accuracy in single skull matching. By quantifying the differences in each parameter, the method provided by this invention can assess the similarity between candidate and target individuals in terms of physiological parameters, thereby effectively eliminating limitations imposed by the target individual's own conditions and equipment, reducing the time required to acquire DTI images, and improving acquisition efficiency. This avoids the impact of DTI image acquisition delays on diagnosis and treatment time, improving diagnostic and treatment efficiency.

[0020] In one possible implementation of the first aspect, the above method further includes: determining the timeliness weight coefficient corresponding to each stroke individual based on the skull feature information, physiological parameter information and storage time of DTI images of each stroke individual in the multimodal feature database; The timeliness weighting coefficient γ for the i-th stroke individual norm，i The formula for determining it is: Where t is the current time, t0 is the storage time, and λ is the decay coefficient; The formula for determining the second difference index D2 is: Here, θ is the basic difference threshold.

[0021] The method provided by this invention introduces a timeliness weight coefficient, calculates the decay degree based on the data storage time, and adjusts a second difference index in conjunction with a basic difference threshold. By reducing the weight of older data through an exponential decay formula, the algorithm prioritizes recent matching cases, ensuring the timeliness of the results. Simultaneously, the constraint of the basic threshold prevents over-reliance on new data from ignoring essential differences in physiological parameters, such as preventing new data from being misjudged as the best match due to data collection errors. This combination achieves a dynamic balance between data freshness and matching quality, improving the timeliness and reliability of the matching results.

[0022] Secondly, this invention provides a DTI image matching system for human brain hemorrhage based on multimodal fusion-enabled CT. The system includes: a data acquisition module for acquiring physiological parameter information, CT images of the head region, and DTI images of multiple stroke individuals; the physiological parameter information includes age, sex, duration of illness, hematoma location, hematoma morphology, and hematoma volume; an information determination module for determining the skull feature information of each stroke individual based on the CT images of the head region; the skull feature information includes the thickness of different parts of the skull and the three-dimensional coordinates of each skull landmark in a preset rectangular coordinate system; and a database construction module for constructing a database based on the skull feature information, physiological parameters, and head region data of multiple stroke individuals. A multimodal feature database is constructed using parameter information and DTI images. An image matching module is used to identify the DTI images of candidate individuals from the multimodal feature database based on the physiological parameter information and skull feature information of the target individual. The candidate individual is one of multiple stroke individuals. The first difference index between the target individual and the candidate individual is less than a preset threshold, and the second difference index between the target individual and the candidate individual is the minimum value among multiple second difference indices corresponding to the target individual and multiple stroke individuals. The first difference index is positively correlated with the degree of difference in skull feature information between the target individual and the candidate individual, and the second difference index is positively correlated with the degree of difference in physiological parameter information between the target individual and the candidate individual.

[0023] Thirdly, an electronic device is provided, the electronic device including a memory and one or more processors; the memory is coupled to the processors; wherein the memory stores computer program code, the computer program code including computer instructions, which, when executed by the processor, cause the electronic device to perform the method as described in any implementation of the first aspect.

[0024] Fourthly, a computer-readable storage medium is provided, including computer instructions that, when executed on an electronic device, cause the electronic device to perform a method as described in any implementation of the first aspect.

[0025] Fifthly, a computer program product is provided that, when run on a computer, causes the computer to perform the method in any implementation of the first aspect.

[0026] Understandably, the beneficial effects achieved by the system of the second aspect, the electronic device of the third aspect, the computer-readable storage medium of the fourth aspect, and the computer program product of the fifth aspect provided above can be referred to with reference to the beneficial effects of the first aspect and any of its possible design embodiments, which will not be repeated here. Attached Figure Description

[0027] Figure 1This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention; Figure 2 A flowchart of a DTI image matching method for human brain hemorrhage based on multimodal fusion-enabled CT provided in an embodiment of the present invention; Figure 3 A flowchart of another DTI image matching method for human brain hemorrhage based on multimodal fusion-enabled CT provided in an embodiment of the present invention; Figure 4 This invention provides an application scenario diagram of a DTI image matching method for human brain hemorrhage based on multimodal fusion-enabled CT. Figure 5 This is a schematic diagram of the structure of an image matching system provided in an embodiment of the present invention. Detailed Implementation

[0028] The technical solutions of the embodiments of the present invention will be described below with reference to the accompanying drawings. In the description of the present invention, unless otherwise stated, " / " indicates that the objects before and after are in an "or" relationship. For example, A / B can represent A or B. The "or" in the present invention is merely a description of the relationship between the related objects, indicating that three relationships can exist. For example, A or B can represent: A alone, A and B simultaneously, and B alone. A and B can be singular or plural. Furthermore, in the description of the present invention, unless otherwise stated, "multiple" refers to two or more. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items.

[0029] Furthermore, to facilitate a clear description of the technical solutions of the embodiments of the present invention, the terms "first" and "second" are used in the embodiments of the present invention to distinguish identical or similar items with essentially the same function and effect. Those skilled in the art will understand that the terms "first" and "second" do not limit the quantity or execution order, and that the terms "first" and "second" are not necessarily different.

[0030] In this embodiment of the invention, the terms "exemplary" or "for example" are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" or "for example" in this embodiment of the invention should not be construed as superior or more advantageous than other embodiments or designs. Specifically, the use of terms such as "exemplary" or "for example" is intended to present the relevant concepts in a concrete manner for ease of understanding.

[0031] In the clinical diagnosis and treatment of neurological diseases, DTI (diffusion tensor imaging) technology has become an important tool for assessing the extent of damage to the corticospinal tract (CST) caused by diseases such as cerebral hemorrhage because it can visually present the course and integrity of white matter fiber tracts.

[0032] However, DTI scans typically require more than 30 minutes of acquisition time. Furthermore, due to the low coverage of current detection equipment and the limited availability of 3.0T high-field MRI, coupled with individual limitations such as physical function, metal implants, or claustrophobia, it is difficult to quickly obtain high-quality DTI images. This results in a long acquisition time, which seriously affects the timing of diagnosis and treatment.

[0033] Therefore, there is an urgent need for a method and system for matching DTI images of human brain hemorrhage based on multimodal fusion-enabled CT, to quickly and accurately acquire DTI images of the target individual, reduce acquisition time, and improve acquisition efficiency. This would avoid delays in DTI image acquisition from affecting diagnosis and treatment time, and improve diagnostic and treatment efficiency.

[0034] In view of this, embodiments of the present invention provide a method for matching DTI images of human brain hemorrhage based on multimodal fusion-enabled CT. The method includes: acquiring physiological parameter information, CT images of the head region, and DTI images of multiple stroke individuals. The physiological parameter information includes age, sex, duration of onset, hematoma location, hematoma morphology, and hematoma volume; determining the skull feature information of each stroke individual based on the CT images of the head region of the multiple stroke individuals. The skull feature information includes the thickness of different parts of the skull and the three-dimensional coordinates of each landmark point of the skull in a preset spatial rectangular coordinate system; and matching the skull feature information, physiological parameter information, and DTI images of the multiple stroke individuals. The process involves constructing a multimodal feature database. Based on the physiological parameters and skull features of the target individual, DTI images of candidate individuals are selected from the multimodal feature database as the target individual's DTI images. The candidate individual is one of multiple stroke individuals. The first difference index between the target individual and the candidate individual is less than a preset threshold, and the second difference index between the target individual and the candidate individual is the minimum value among the multiple second difference indices between the target individual and the multiple stroke individuals. The first difference index is positively correlated with the degree of difference in skull features between the target individual and the candidate individual, and the second difference index is positively correlated with the degree of difference in physiological parameters between the target individual and the candidate individual.

[0035] The method provided by this invention constructs a multimodal feature database based on physiological parameters, skull features, and DTI images of multiple stroke individuals. It then rapidly and accurately matches candidate individuals from the multimodal feature database using a first difference index and a second difference index, based on the target individual's physiological parameters and CT images. The first difference index characterizes the degree of difference between skull features, while the second difference index characterizes the degree of difference between physiological parameters. Identifying candidate individuals' DTI images as the target individual's effectively eliminates limitations imposed by the target individual's own conditions and equipment, reduces the time required to acquire DTI images, and improves acquisition efficiency. This avoids delays in DTI image acquisition that could impact treatment time, thus improving diagnostic and treatment efficiency.

[0036] In some embodiments, the human brain hemorrhage DTI image matching method based on multimodal fusion-enabled CT provided in this embodiment of the invention can be executed by a human brain hemorrhage DTI image matching system 100 based on multimodal fusion-enabled CT (hereinafter referred to as image matching system 100).

[0037] As an example, the image matching system 100 can be any electronic device 200 with data processing capabilities, such as a general-purpose computer, personal computer, laptop computer, switch, or tablet computer. The specific implementation of the image matching system 100 is not limited here.

[0038] Figure 1 A schematic diagram of the hardware structure of an electronic device provided in an embodiment of the present invention is shown. The electronic device 200 includes a processor 210, a memory 220, and a communication interface 230.

[0039] Processor 210 may include one or more processing cores. Processor 210 connects to various parts within electronic device 200 using various interfaces and lines, and performs various functions and processes data of electronic device 200 by running or executing instructions, programs, code sets, or instruction sets stored in memory 220, and by calling data stored in memory 220. Optionally, processor 210 may be implemented using at least one of the following hardware forms: Central Processing Unit (CPU), Graphics Processing Unit (GPU), Digital Signal Processing (DSP), Field-Programmable Gate Array (FPGA), and Programmable Logic Array (PLA).

[0040] The memory 220 may include random access memory (RAI) or read-only memory (ROI). Optionally, the memory 220 may include non-transitory computer-readable storage ledger. The memory 220 may be used to store instructions, programs, code, code sets, or instruction sets. The memory 220 may include a stored program area. The stored program area may store instructions for implementing an operating system, instructions for implementing at least one function (such as data acquisition, database construction, and image matching), and instructions for implementing the various method embodiments described above.

[0041] The communication interface 230 is used to communicate with other devices, equipment, or communication networks, such as data storage devices, image processing devices, or Ethernet, wireless access networks (RAN), wireless local area networks (WLAN), etc.

[0042] In terms of physical implementation, the aforementioned devices (such as processor 210, memory 220, and communication interface 230) can each be devices within the same device (such as a laptop computer). Alternatively, at least two of these devices can be located within the same device, i.e., as different devices within the same device, similar to the deployment of devices or components in a distributed system.

[0043] It is understood that the structure illustrated in this embodiment does not constitute a specific limitation on the electronic device 200. In other embodiments of the present invention, the electronic device 200 may include more or fewer components than illustrated, or combine some components, or split some components, or have different component arrangements. The illustrated components may be implemented in hardware, software, or a combination of software and hardware.

[0044] The following description, in conjunction with the accompanying drawings, illustrates a method for matching DTI images of human brain hemorrhage based on multimodal fusion-enabled CT, according to an embodiment of the present invention.

[0045] Figure 2 This is a flowchart illustrating a method for matching DTI images of human brain hemorrhage based on multimodal fusion-enhanced CT, as provided in an embodiment of the present invention. Optionally, this method can be... Figure 1 The illustrated electronic device 200 performs this function. The method may include the following steps: S1. Obtain physiological parameter information, CT images and DTI images of the head region from multiple stroke individuals. Physiological parameter information includes age, sex, duration of onset, hematoma location, hematoma morphology and hematoma volume.

[0046] Specifically, the location, morphology, and volume of the hematoma in the aforementioned multiple stroke individuals are determined based on CT images of the head region of the multiple stroke individuals. This embodiment of the invention does not impose any particular restrictions on the specific method of determining the location, morphology, and volume of the hematoma.

[0047] S2. Determine the skull features of each stroke individual based on CT images of the head region of multiple stroke individuals.

[0048] Specifically, the skull feature information includes the thickness of different parts that make up the skull and the three-dimensional coordinates of each landmark point of the skull in a preset spatial rectangular coordinate system.

[0049] In one possible implementation, the different parts that make up the skull include the frontal bone, parietal bone, temporal bone, occipital bone, and sphenoid bone, and the landmarks include the anterior point, posterior point, left temporal point, right temporal point, apex, and base of the skull cavity, as well as the hemorrhage point, glabella point, anterior fontanelle point, lambdoid point, center of external auditory canal, and external occipital protuberance.

[0050] Specifically, the anterior point is the anterior point of the cranial cavity corresponding to the frontal bone, the posterior point is the posterior point of the cranial cavity corresponding to the occipital bone, and the left and right temporal points are the lateral points of the cranial cavities corresponding to the left and right temporal bones, respectively. The apex is the highest point of the cranial cavity corresponding to the parietal bone. The base of the skull is the lowest point of the cranial cavity corresponding to the sphenoid bone.

[0051] It should be understood that the above markers are merely illustrative examples, and operators may set more or fewer markers based on different usage scenarios. This embodiment of the invention does not impose any particular limitations on this.

[0052] The method provided in this invention, by defining the components and landmarks of the skull, can quickly and accurately determine the thickness of different parts of the skull and the three-dimensional coordinates of different landmarks. This can effectively reduce the ambiguity of feature extraction, achieve accurate acquisition of skull feature information of different individuals, and thus improve the matching degree between the DTI images of candidate individuals and the DTI images of target individuals, thereby improving the accuracy of matching.

[0053] In some embodiments, S2 includes: Based on CT images of the head region of multiple stroke patients, a binarized image of the skull for each stroke patient is determined. A 3D model of the skull corresponding to the binarized image is determined based on a 3D reconstruction algorithm. The thickness of different parts of the skull is determined based on the distance between the points corresponding to the 3D contours of different parts of the skull on the 3D model and the center point of the 3D model. The 3D coordinates of each landmark point of the skull in a preset Cartesian coordinate system are determined, where the center point of the 3D model is the origin.

[0054] For example, the above-mentioned method determines the binarized skull image of each stroke individual based on CT images of the head region of multiple stroke individuals; and determines the three-dimensional skull model corresponding to the binarized skull image of each stroke individual based on a three-dimensional reconstruction algorithm, including: Noise reduction was performed on CT images of the head region from multiple stroke patients. Median filtering was used to remove noise, and Gaussian filtering was applied to smooth the images. Simultaneously, histogram equalization was used to enhance image contrast and expand the dynamic range of grayscale values ​​between the skull and surrounding soft tissues, as well as air, making the skull more clearly identifiable. Thresholding segmentation was used to initially separate the skull from the CT images based on specific grayscale value ranges, obtaining an initial skull region. A region growing algorithm was employed, selecting uniformly grayscale points within the skull region as seed points. Based on a set grayscale similarity threshold, surrounding pixels meeting the criteria were gradually merged into the skull region. The segmented skull region was then binarized, assigning a value of 1 to pixels corresponding to the skull tissue and 0 to the background region, forming a binary image containing only the skull and background. Morphological operations were applied: erosion was used to remove small burrs and isolated noise points at the edges of the skull region, and dilation was used to fill small internal holes, smoothing the skull contour edges. Based on the processed binarized image of the skull, a 3D reconstruction algorithm is used to construct a 3D model of the skull from the 2D binarized image. The geometric center point of the 3D skull model is used as the origin of a preset spatial rectangular coordinate system. According to the standard anatomical orientation of the skull, the anterior-posterior direction of the skull is set as the X-axis, the left-right direction as the Y-axis, and the up-down direction as the Z-axis.

[0055] The method provided in this invention converts CT images into binary skull images and then establishes a three-dimensional skull model corresponding to each binary skull image. This ensures that the skull features of different individuals have a unified reference, eliminating the influence of positioning deviations on thickness calculation and landmark coordinate extraction. The method provided in this invention calculates skull thickness based on three-dimensional contours and combines it with precise landmark coordinates to achieve a quantitative description of skull morphology, providing a high-precision data foundation for subsequent difference calculations, thereby improving the reliability and accuracy of matching.

[0056] S3. Construct a multimodal feature database based on skull feature information, physiological parameter information and DTI images of multiple stroke individuals.

[0057] S4. Based on the physiological parameters and skull features of the target individual, the DTI images of the candidate individuals are identified as the DTI images of the target individual from the multimodal feature database.

[0058] Specifically, the candidate individual is one of multiple stroke individuals. The first difference index between the target individual and the candidate individual is less than a preset threshold, and the second difference index between the target individual and the candidate individual is the minimum value among the multiple second difference indices between the target individual and the multiple stroke individuals. The first difference index is positively correlated with the degree of difference in skull feature information between the target individual and the candidate individual, and the second difference index is positively correlated with the degree of difference in physiological parameter information between the target individual and the candidate individual.

[0059] In one example, the preset threshold ranges from greater than or equal to 0.15 to less than or equal to 0.3.

[0060] In some embodiments, see Figure 3 The above S4 includes: S41. Based on the skull feature information of the target individual, determine the initial candidate individual set from the multimodal feature database, and the first difference index between the target individual and each initial candidate individual in the initial candidate individual set is less than a preset threshold.

[0061] In one possible implementation, the formula for determining the first difference index D1 is: D1=α·D thickness +(1-α)·D landmark ; Where α is the overall weight coefficient, D thickness D is a score for differences in skull thickness. landmark The score is given for the difference in the location of the marker; in one example, the overall weighting coefficient α is 0.4.

[0062] Skull thickness difference score D thickness The formula for determining it is: N represents the number of parts that make up the skull, T target,i T represents the thickness of the i-th part of the skull of the target individual; candidate,i The thickness of the i-th part of the skull that makes up an individual with a stroke; Marker location difference score D landmark The formula for determining it is: M is the number of markers, P target,j P represents the three-dimensional coordinates of the j-th marker point of the target individual; candidate,j Let j be the three-dimensional coordinates of the j-th landmark point in an individual with stroke, and normalize_factor be the normalize_factor. j This is the normalization factor.

[0063] Specifically, the normalization factor is the overall size of the skull or the average distance from the landmark point to the center of the skull, which can avoid the influence of scale on different individuals.

[0064] The method provided in this invention decomposes skull feature differences into thickness differences and landmark position differences, and obtains a first difference index through weighted summation. Thickness differences are calculated by comparing the thickness of each skull segment between the target individual and the candidate individual, while landmark differences are evaluated based on Euclidean distance in three-dimensional coordinates, with a normalization factor introduced to eliminate the influence of measurement scale. In this way, different dimensions of skull features contribute differently to the matching results, and separating thickness and landmark differences allows for refined evaluation; the weighted summation method assigns reasonable weights to both, enabling the method provided in this invention to comprehensively consider both morphological and spatial differences; and the normalization process ensures the comparability of difference scores for different landmark points, avoiding the over- or under-sighting of certain features due to scale issues, ultimately achieving scientific quantification and accurate matching of skull feature differences.

[0065] In another possible implementation, different parts of the skull are each assigned a first weighting coefficient, and each marker is assigned a second weighting coefficient.

[0066] In one example, the first weighting coefficient for the frontal bone is 0.25, the first weighting coefficient for the parietal bone is 0.25, the first weighting coefficient for the temporal bone is 0.2, the first weighting coefficient for the occipital bone is 0.2, and the first weighting coefficient for the sphenoid bone is 0.10.

[0067] In another example, the second weighting coefficient for the anterior point of the cranial cavity is 0.1, the second weighting coefficient for the posterior point is 0.1, the second weighting coefficient for the left temporal point is 0.07, the second weighting coefficient for the right temporal point is 0.07, the second weighting coefficient for the vertex is 0.08, the second weighting coefficient for the skull base is 0.03, the second weighting coefficient for the hemorrhage point is 0.2, the second weighting coefficient for the glabella is 0.1, the second weighting coefficient for the anterior fontanelle is 0.05, the second weighting coefficient for the zygomatic point is 0.05, the second weighting coefficient for the center point of the external auditory canal is 0.1, and the second weighting coefficient for the external occipital protuberance is 0.05.

[0068] Skull thickness difference score D thickness The formula for determining it is: Marker location difference score D landmark The formula for determining it is: Among them, w i v is the first weighting coefficient for the i-th part that makes up the skull; i It is the second weight coefficient of the i-th marker.

[0069] The method provided in this invention assigns differentiated weights to different parts of the skull and landmarks, thereby improving the determination of thickness differences and landmark position differences. It can enhance the matching priority of key structures and weaken the interference of secondary features by configuring weights when different skull regions and landmarks have different importance in DTI images. At the same time, the adjustability of weights allows the algorithm to adapt to different use scenarios, improving flexibility and practicality.

[0070] Because of the differences between individuals, using the same preset threshold leads to a decrease in matching accuracy.

[0071] To address this issue, in some embodiments, the method provided by the present invention further includes: Determine the preset threshold corresponding to the target individual.

[0072] Specifically, the formula for determining the preset threshold μ is: μ = μ0·ε; Where μ0 is the skull difference benchmark threshold, and ε is the skull complexity weight; When the overall complexity index of the target user is less than 0.7, the skull complexity weight ε = 0.8; when the overall complexity index of the target user is greater than or equal to 0.7 and less than 1.3, the skull complexity weight ε = 1; when the overall complexity index of the target user is greater than 1.3, the skull complexity weight ε = 1.2. The formula for determining the overall complexity exponent τ is: Where A is the balance coefficient, Dref is the dispersion reference value; CV thickness η is the coefficient of variation for skull thickness. thickness σ represents the standard deviation of the different parts that make up the skull. thickness Mean values ​​of different parts that make up the skull, DV landmark P represents the dispersion of skull landmarks. j Let P be the three-dimensional coordinates of the j-th marker point of the target user, and let P be the three-dimensional coordinates of the geometric center of all marker points of the target user.

[0073] The method provided in this invention can dynamically adjust a preset threshold based on the skull complexity of a target individual. By calculating the coefficient of variation of skull thickness and the dispersion of landmark points for the target individual, a comprehensive complexity index is constructed. Different skull complexity weights are then mapped according to the index range, thereby achieving dynamic adjustment of the preset threshold for the target individual. The logic is that individual differences in skull morphology naturally exist, and a fixed threshold cannot meet the matching requirements of complex skull structures. The dynamic threshold mechanism quantifies skull complexity, applying a strict threshold to individuals with simple structures to ensure high matching accuracy, while appropriately relaxing the standard for complex individuals to avoid having no usable matching cases due to overly strict thresholds. This adaptive adjustment strategy expands the scope of application while ensuring matching quality, especially significantly improving the success rate and feasibility of matching when dealing with special individuals.

[0074] S42. Based on the physiological parameter information of the target individual, determine the second difference index between the target individual and each initial candidate individual in the initial candidate individual set, determine the initial candidate individual with the minimum value of the second difference index as the candidate individual, and determine the DTI image of the candidate individual as the DTI image of the target individual.

[0075] In one possible implementation, the formula for determining the second difference index D2 is: K represents the number of parameters included in the physiological parameter information; β k Score is the weighting coefficient of the k-th parameter included in the physiological parameter information. k The difference score for the k-th parameter included in the physiological parameter information; Specifically, the weighting coefficient for age is 0.2, the weighting coefficient for gender is 0.1, the weighting coefficient for duration of illness is 0.25, the weighting coefficient for hematoma location is 0.2, the weighting coefficient for hematoma morphology is 0.1, and the weighting coefficient for hematoma volume is 0.15.

[0076] Physiological parameter information includes age difference score age for: A target A represents the age of the target individual. candidate For the age of the individual who suffered a stroke, A max and A min The maximum and minimum ages of multiple stroke individuals; Physiological parameters include a score indicating gender differences. gender for: When the target individual and the stroke patient are of the same sex, Score gender =0; When the target individual and the stroke patient are of different sexes, Scoregender = 1; Physiological parameter information includes the score for the difference in the duration of illness. time for: T target The duration of illness for the target individual; T candidate The duration of onset for an individual with stroke; T max and T min The maximum and minimum duration of onset for multiple stroke individuals; Physiological parameters include a score for the difference in hematoma location. location for: When the hematoma areas of the target individual and the stroke patient are located in the same region, Score location =0; When the hematoma areas of the target individual and the stroke individual are located in adjacent areas, Score location =0.5; When the hematoma areas of the target individual and the stroke patient are located in different and non-adjacent areas, Score location =1; Physiological parameters include a score for differences in hematoma morphology. shape for: L target L represents the length of the long axis of the hematoma in the target individual. candidate L represents the length of the long axis of the hematoma in an individual with stroke. max and L min The maximum and minimum lengths of the long axis of the hematoma are given for multiple stroke individuals; W target W represents the short axis length of the hematoma in the target individual. candidate W represents the short axis length of the hematoma in an individual with stroke. max and W min H represents the maximum and minimum short axis lengths of hematoma in multiple stroke individuals; target For the hematoma height of the target individual, H candidate The height of the hematoma in an individual with stroke; H max and H min This represents the maximum and minimum hematoma height among multiple stroke patients.

[0077] Physiological parameters include a score for the difference in hematoma volume. volume for: Vtarget V represents the hematoma volume of the target individual; candidate This refers to the hematoma volume in an individual with a stroke.

[0078] As described above, the method provided in this invention quantifies the differences in physiological parameters such as age, gender, and duration of illness, and obtains a second difference index through weighted summation. Specifically, age and duration of illness are normalized to calculate difference scores, gender is assessed using a 0-1 binary system, and hematoma location is graded based on regional correlation. This achieves complementarity between physiological parameters and skull features, addressing the issue of poor accuracy in single skull matching. By quantifying the differences in each parameter, the method provided by this invention can assess the similarity between candidate and target individuals in terms of physiological parameters, thereby effectively eliminating limitations imposed by the target individual's own conditions and equipment, reducing the time required to acquire DTI images, and improving acquisition efficiency. This avoids delays in DTI image acquisition affecting treatment time and improves diagnostic and treatment efficiency.

[0079] In another possible implementation, the method provided in this embodiment of the invention further includes: The timeliness weight coefficient for each stroke individual is determined based on the skull feature information, physiological parameter information, and storage time of DTI images of each stroke individual in the multimodal feature database. The timeliness weighting coefficient γ for the i-th stroke individual norm，i The formula for determining it is: Where t is the current time, t0 is the storage time, and λ is the decay coefficient.

[0080] Furthermore, the formula for determining the second difference index D2 is as follows: Here, θ is the basic difference threshold.

[0081] In one example, the attenuation coefficient is 0.001 to 0.01 per day, and the baseline variability threshold is 0.2.

[0082] The method provided by this invention introduces a timeliness weight coefficient, calculates the decay degree based on the data storage time, and adjusts a second difference index in conjunction with a basic difference threshold. By reducing the weight of older data through an exponential decay formula, the algorithm prioritizes recent matching cases, ensuring the timeliness of the results. Simultaneously, the constraint of the basic threshold prevents over-reliance on new data from ignoring essential differences in physiological parameters, such as preventing new data from being misjudged as the best match due to data collection errors. This combination achieves a dynamic balance between data freshness and matching quality, improving the timeliness and reliability of the matching results.

[0083] As described in S1-S4 above, the method provided in this embodiment of the invention constructs a multimodal feature database based on the physiological parameter information, skull feature information, and DTI images of multiple stroke individuals. Based on the physiological parameter information and CT images of the target individual, it quickly and accurately matches candidate individuals from the multimodal feature database using a first difference index and a second difference index. The first difference index characterizes the degree of difference between skull feature information, and the second difference index characterizes the degree of difference between physiological parameter information. Identifying the DTI images of candidate individuals as the DTI images of the target individual effectively eliminates the limitations imposed by the target individual's own conditions and equipment, reduces the time required to acquire DTI images, and improves acquisition efficiency. It avoids delays in DTI image acquisition from affecting treatment time, thus improving treatment efficiency.

[0084] In one example, see Figure 4 , Figure 4 This diagram illustrates an application scenario of a DTI image matching method for human brain hemorrhage based on multimodal fusion-enabled CT, as shown in an embodiment of the present invention. The method provided in this embodiment is applied to an image matching system, which includes a server 400 and multiple terminal devices 410. The server 400 communicates with each of the multiple terminal devices 410.

[0085] Server 400 acquires physiological parameter information, CT images and DTI images of the head region of multiple stroke individuals. The physiological parameter information includes age, sex, duration of onset, hematoma location, hematoma morphology and hematoma volume. Server 400 determines the skull feature information of each stroke individual based on the CT images of the head region of the multiple stroke individuals. The skull feature information includes the thickness of different parts of the skull and the three-dimensional coordinates of each landmark point of the skull in a preset spatial rectangular coordinate system. Server 400 constructs a multimodal feature database based on the skull feature information, physiological parameter information and DTI images of the multiple stroke individuals.

[0086] For any terminal device 410 sending an image matching request to the server 400, the image matching request carries the physiological parameter information of the target individual and a CT image of the head region. In response to the image matching request, the server 400 determines the skull feature information of the target individual based on the CT image of the head region. Then, based on the physiological parameter information and skull feature information of the target individual, the server 400 determines candidate individuals from the multimodal feature database and identifies the DTI images of the candidate individuals as the DTI images of the target individual. Finally, the DTI image of the target individual is sent to the terminal device 410.

[0087] The foregoing mainly describes the solutions of the embodiments of the present invention from a methodological perspective. It is understood that, in order to achieve the above-mentioned functions, the image matching system 100 includes at least one of the hardware structures and software modules corresponding to the execution of each function. Those skilled in the art should readily recognize that, in conjunction with the units and algorithm steps of the various examples described in the embodiments disclosed herein, the embodiments of the present invention can be implemented in hardware or a combination of hardware and computer software. Whether a function is executed in hardware or by computer software driving hardware 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 implementation should not be considered beyond the scope of the embodiments of the present invention.

[0088] In this embodiment of the invention, the image matching system 100 can be divided into functional units according to the above method example. For example, the image matching system 100 can be divided into functional units corresponding to various functions, or two or more functions can be integrated into one processing unit. The integrated unit can be implemented in hardware or as a software functional unit. It should be noted that the unit division in this embodiment of the invention is illustrative and only represents one logical functional division; other division methods may be used in actual implementation.

[0089] For example, Figure 5This diagram illustrates the hardware structure of an image matching system according to an embodiment of the present invention. The image matching system 100 includes: a data acquisition module 110, used to acquire physiological parameter information, CT images of the head region, and DTI images of multiple stroke individuals; the physiological parameter information includes age, gender, duration of illness, hematoma location, hematoma morphology, and hematoma volume; an information determination module 120, used to determine the skull feature information of each stroke individual based on the CT images of the head region of the multiple stroke individuals; the skull feature information includes the thickness of different parts of the skull and the three-dimensional coordinates of each skull landmark in a preset Cartesian coordinate system; and a database construction module 130, used to construct a multi-model database based on the skull feature information, physiological parameter information, and DTI images of the multiple stroke individuals. The system includes a multimodal feature database and an image matching module 140, which is used to identify the DTI images of candidate individuals as the DTI images of the target individual from the multimodal feature database based on the physiological parameter information and skull feature information of the target individual. The candidate individual is one of multiple stroke individuals. The first difference index between the target individual and the candidate individual is less than a preset threshold, and the second difference index between the target individual and the candidate individual is the minimum value among the multiple second difference indices between the target individual and the multiple stroke individuals. The first difference index is positively correlated with the degree of difference in skull feature information between the target individual and the candidate individual, and the second difference index is positively correlated with the degree of difference in physiological parameter information between the target individual and the candidate individual.

[0090] It should be understood that specific descriptions of the above-mentioned optional methods can be found in the foregoing method embodiments, and will not be repeated here. Furthermore, explanations of any of the image matching systems 100 provided above, as well as descriptions of their beneficial effects, can be found in the corresponding method embodiments described above, and will not be repeated here.

[0091] This invention also provides a computer-readable storage medium storing at least one computer instruction, which is loaded and executed by a processor to implement the methods of the various embodiments described above. Explanations of the relevant content and descriptions of the beneficial effects of any of the computer-readable storage media provided above can be found in the corresponding embodiments described above, and will not be repeated here.

[0092] This invention also provides a chip. This chip integrates a control circuit for implementing the functions of the image matching system 100 described above, and one or more ports. Optionally, the functions supported by this chip are as described above, and will not be repeated here.

[0093] Those skilled in the art will understand that the program for implementing all or part of the steps of the above embodiments, which can be executed by a program instructing related hardware, can be stored in a computer-readable storage medium. The storage medium mentioned above can be a read-only memory, a random access memory, etc. The processing unit or processor mentioned above can be a central processing unit, a general-purpose processor, an application-specific integrated circuit (ASIC), a microprocessor (DSP), a field-programmable gate array (FPGA), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof.

[0094] This invention also provides a computer program product containing instructions that, when executed on a computer, cause the computer to perform any of the methods described in the above embodiments. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the flow or function according to the embodiments of this invention is generated. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, computer instructions may be transmitted from one website, computer, server, or data center to another via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may 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. The available medium may be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., SSD), etc.

[0095] It should be noted that the devices for storing computer instructions or computer programs provided in the embodiments of the present invention, such as, but not limited to, the aforementioned memory, computer-readable storage medium, and communication chip, are all non-transitory. Those skilled in the art should recognize that the functions described in the embodiments of the present invention in one or more of the above examples can be implemented using hardware, software, firmware, or any combination thereof. When implemented using software, these functions can be stored in a computer-readable storage medium or transmitted as one or more instructions or code on a computer-readable storage medium. Computer-readable storage media include computer storage media and communication media, wherein communication media include any medium that facilitates the transmission of computer programs from one place to another. Storage media can be any available medium accessible to general-purpose or special-purpose computers.

[0096] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A method for matching DTI images of human brain hemorrhage based on multimodal fusion-enabled CT, characterized in that, The method includes: Physiological parameters, CT images and DTI images of the head region of multiple stroke individuals were acquired. The physiological parameters included age, sex, duration of onset, hematoma location, hematoma morphology and hematoma volume. Based on CT images of the head region of the multiple stroke individuals, the skull feature information of each stroke individual is determined. The skull feature information includes the thickness of different parts that make up the skull and the three-dimensional coordinates of each landmark point of the skull in a preset spatial rectangular coordinate system. A multimodal feature database is constructed based on the skull feature information, physiological parameter information, and DTI images of the plurality of stroke individuals. Based on the physiological parameter information and skull feature information of the target individual, the DTI images of candidate individuals are determined from the multimodal feature database as the DTI images of the target individual. The candidate individual is one of the plurality of stroke individuals. The first difference index corresponding to the target individual and the candidate individual is less than a preset threshold, and the second difference index corresponding to the target individual and the candidate individual is the minimum value among the plurality of second difference indices corresponding to the target individual and the plurality of stroke individuals. The first difference index is positively correlated with the degree of difference in the skull feature information of the target individual and the candidate individual, and the second difference index is positively correlated with the degree of difference in the physiological parameter information of the target individual and the candidate individual.

2. The method according to claim 1, characterized in that, The different parts that make up the skull include the frontal bone, parietal bone, temporal bone, occipital bone, and sphenoid bone. The landmarks include the anterior point, posterior point, left temporal point, right temporal point, apex, and base of the skull cavity, as well as the hemorrhage point, glabella point, anterior fontanelle point, chevron point, center point of external auditory canal, and external occipital protuberance point.

3. The method according to claim 2, characterized in that, The step of determining the skull feature information of each stroke patient based on CT images of the head region of the plurality of stroke patients includes: The skull binarized image of each stroke individual was determined based on CT images of the head region of the multiple stroke individuals. Based on the three-dimensional reconstruction algorithm, the three-dimensional skull model corresponding to the binarized skull image of each stroke individual is determined; The thickness of the different parts of the skull is determined based on the distance between each point corresponding to the three-dimensional contour of the different parts of the skull on the three-dimensional skull model and the center point of the three-dimensional skull model. The three-dimensional coordinates of each landmark point of the skull in a preset spatial rectangular coordinate system are determined, wherein the preset spatial rectangular coordinate system takes the center point of the three-dimensional skull model as its origin.

4. The method according to claim 3, characterized in that, The formula for determining the first difference index D1 is: D1=α·D thickness +(1-a)·D landmark ; Where α is the overall weight coefficient, D thickness D is a score for differences in skull thickness. landmark Scoring for differences in landmark location; Score for differences in skull thickness (D) thickness The formula for determining it is: N represents the number of parts that make up the skull, T target,i T represents the thickness of the i-th part of the skull of the target individual; candidate,i The thickness of the i-th part of the skull that makes up an individual with a stroke; Marker location difference score D landmark The formula for determining it is: M is the number of markers, P target,j P represents the three-dimensional coordinates of the j-th marker point of the target individual; candidate,j Let j be the three-dimensional coordinates of the j-th landmark point in an individual with stroke, and normalize_factor be the normalize_factor. j This is the normalization factor.

5. The method according to claim 4, characterized in that, Different parts of the skull are each assigned a first weighting coefficient, and each marker point is assigned a second weighting coefficient. Skull thickness difference score D thickness The formula for determining it is: Marker location difference score D landmark The formula for determining it is: Among them, w i v is the first weighting coefficient for the i-th part that makes up the skull; i It is the second weight coefficient of the i-th marker.

6. The method according to claim 5, characterized in that, The formula for determining the preset threshold μ is: μ = μ0·ε; Where μ0 is the skull difference benchmark threshold, and ε is the skull complexity weight; When the overall complexity index of the target user is less than 0.7, the skull complexity weight ε = 0.8; when the overall complexity index of the target user is greater than or equal to 0.7 and less than 1.3, the skull complexity weight ε = 1; when the overall complexity index of the target user is greater than 1.3, the skull complexity weight ε = 1.

2. The formula for determining the comprehensive complexity index τ is as follows: Where A is the balance coefficient, Dref is the dispersion reference value; CV thickness η is the coefficient of variation for skull thickness. thickness σ represents the standard deviation of the different parts that make up the skull. thickness Mean values ​​of different parts that make up the skull, DV landmark P represents the dispersion of skull landmarks. j Let P be the three-dimensional coordinates of the j-th marker point of the target user, and let P be the three-dimensional coordinates of the geometric center of all marker points of the target user.

7. The method according to claim 6, characterized in that, The formula for determining the second difference index D2 is: K represents the number of parameters included in the physiological parameter information; β k Score is the weighting coefficient of the k-th parameter included in the physiological parameter information. k The difference score for the k-th parameter included in the physiological parameter information; Physiological parameter information includes age difference score age for: A target A represents the age of the target individual. candidate For the age of the individual who suffered a stroke, A max and A min The maximum and minimum ages of the plurality of stroke individuals; Physiological parameters include a score indicating gender differences. gender for: When the target individual and the stroke patient are of the same sex, Score gender =0; When the target individual and the stroke patient are of different sexes, Scoregender = 1; Physiological parameter information includes the score for the difference in the duration of illness. time for: T target The duration of illness for the target individual; T candidate The duration of onset for an individual with stroke; T max and T min The maximum and minimum duration of onset for the multiple stroke individuals are given. Physiological parameters include a score for the difference in hematoma location. location for: When the hematoma areas of the target individual and the stroke patient are located in the same region, Score location =0; When the hematoma areas of the target individual and the stroke individual are located in adjacent areas, Score location =0.5; Score = 0.5; when the hematoma areas of the target individual and the stroke individual are located in different and non-adjacent areas. location =1; Physiological parameters include a score for differences in hematoma morphology. shape for: L target L represents the length of the long axis of the hematoma in the target individual. candidate L represents the length of the long axis of the hematoma in an individual with stroke. max and L min The maximum and minimum lengths of the long axis of the hematoma are given for multiple stroke patients. W target W represents the short axis length of the hematoma in the target individual. candidate W represents the short axis length of the hematoma in an individual with stroke. max and W min H represents the maximum and minimum short axis lengths of hematoma in multiple stroke individuals; target For the hematoma height of the target individual, H candidate The height of the hematoma in an individual with stroke; H max and H min The maximum and minimum hematoma heights are for multiple stroke patients. Physiological parameters include a score for the difference in hematoma volume. volume for: V target V represents the hematoma volume of the target individual; candidate This refers to the hematoma volume in an individual with a stroke.

8. The method according to claim 7, characterized in that, The method further includes: The timeliness weight coefficient for each stroke individual is determined based on the skull feature information, physiological parameter information, and storage time of DTI images of each stroke individual in the multimodal feature database. The timeliness weighting coefficient γ for the i-th stroke individual norm，i The formula for determining it is: Where t is the current time, t0 is the storage time, and λ is the decay coefficient; The formula for determining the second difference index D2 is: Here, θ is the basic difference threshold.

9. A DTI image matching system for human brain hemorrhage based on multimodal fusion-enabled CT, characterized in that, The system includes: The data acquisition module is used to acquire physiological parameter information, CT images and DTI images of the head region of multiple stroke individuals. The physiological parameter information includes age, gender, duration of onset, hematoma location, hematoma morphology and hematoma volume. The information determination module is used to determine the skull feature information of each of the multiple stroke individuals based on CT images of the head region of the multiple stroke individuals. The skull feature information includes the thickness of different parts that make up the skull and the three-dimensional coordinates of each landmark point of the skull in a preset spatial rectangular coordinate system. The database construction module is used to construct a multimodal feature database based on the skull feature information, physiological parameter information and DTI images of the multiple stroke individuals; The image matching module is used to determine the DTI image of the target individual from the multimodal feature database based on the physiological parameter information and skull feature information of the target individual. The candidate individual is one of the plurality of stroke individuals. The first difference index corresponding to the target individual and the candidate individual is less than a preset threshold, and the second difference index corresponding to the target individual and the candidate individual is the minimum value among the plurality of second difference indices corresponding to the target individual and the plurality of stroke individuals. The first difference index is positively correlated with the degree of difference in skull feature information between the target individual and the candidate individual, and the second difference index is positively correlated with the degree of difference in physiological parameter information between the target individual and the candidate individual.

10. An electronic device, characterized in that, include: processor; Memory used to store the processor's executable instructions; The processor is configured to execute the instructions to implement a human brain hemorrhage DTI image matching method based on multimodal fusion-enabled CT as described in any one of claims 1-8.

Images