A method, device and computer equipment for measuring hemodynamic heterogeneity

Abstract

This specification relates to the field of magnetic resonance imaging (MRI) technology, and more particularly to a method, apparatus, and computer device for measuring hemodynamic heterogeneity. The method for measuring hemodynamic heterogeneity includes determining a set of temporal signal intensity data corresponding to pixels in a region of interest within a received dynamic contrast-enhanced MRI image; processing the temporal signal intensity data set to obtain a set of characterization information data; processing the characterization information data using a classification function corresponding to the characterization information data included in the characterization information data set to obtain a characterization information set; and determining heterogeneity prediction results based on the characterization information set. Using the embodiments of this specification, heterogeneity prediction results are achieved based on the signal intensity corresponding to pixels in a dynamic contrast-enhanced MRI image. This reduces errors caused by inter-individual differences, making the heterogeneity prediction results more accurate.

Description

Technical Field

[0001] This specification relates to the field of magnetic resonance imaging technology, and in particular to a method, apparatus, and computer equipment for measuring hemodynamic heterogeneity. Background Technology

[0002] Currently, magnetic resonance imaging (MRI) technology plays an irreplaceable role in the early detection and accurate diagnosis of breast cancer due to its ability to provide structural and functional information. Dynamic contrast-enhanced MRI (VCE) is a routine sequence in breast MRI examinations and a key technology for reflecting the spatial heterogeneity of hemodynamic information. Methods for determining the spatial heterogeneity of hemodynamic information based on VCE images obtained through MRI include using different mathematical models to fit the time-concentration curves of contrast agents within the lesion to reflect the spatial heterogeneity of hemodynamic information within the lesion. However, due to the lack of standardization of MRI acquisition parameters and the difficulty in uniformly collecting corresponding datasets, the predicted heterogeneity results have large errors. Another method for determining the spatial heterogeneity of hemodynamic information based on VCE images is to use deep learning models to predict heterogeneity results. When using deep learning models to predict heterogeneity results, the lack of training sample sets and inter-individual differences lead to large errors in the predicted heterogeneity results.

[0003] How to reduce the error in predicting heterogeneous results from dynamic contrast-enhanced magnetic resonance images obtained based on magnetic resonance technology is a problem that urgently needs to be solved in the current technology. Summary of the Invention

[0004] To address the problems in the prior art, this specification provides a method, apparatus, computer device, and storage medium for measuring hemodynamic heterogeneity. It enables the prediction of heterogeneity results based on the signal intensity corresponding to pixels in dynamically contrast-enhanced magnetic resonance images. This reduces errors caused by inter-individual differences, resulting in more accurate heterogeneity predictions.

[0005] To solve the above-mentioned technical problems, the specific technical solution in this specification is as follows:

[0006] On the one hand, embodiments of this specification provide a method for measuring hemodynamic heterogeneity, including,

[0007] Based on the received dynamic contrast-enhanced magnetic resonance image, determine the set of time-series signal intensity data corresponding to the pixels in the region of interest in the dynamic contrast-enhanced magnetic resonance image;

[0008] The time-series signal strength data set is processed to obtain a characterization information data set;

[0009] The representation information data is processed using a classification function corresponding to the representation information data included in the representation information data set to obtain the representation information set; and

[0010] Based on the set of characterization information, the heterogeneity prediction results are determined.

[0011] Furthermore, based on the received dynamic contrast-enhanced magnetic resonance image, determining the time-series signal intensity data set corresponding to pixels in the region of interest within the dynamic contrast-enhanced magnetic resonance image further includes:

[0012] Based on the received dynamic contrast-enhanced magnetic resonance image, determine the signal intensity data of each pixel in the region of interest within the dynamic contrast-enhanced magnetic resonance image; and

[0013] Based on the signal strength data corresponding to the pixel at the same position at each time, construct the time-series signal strength data set corresponding to each pixel.

[0014] Furthermore, the time-series signal strength data set is processed to obtain a characterization information data set, which further includes:

[0015] The data in the time-series signal strength data set is processed using at least one of the wash-in rate formula, the wash-out rate formula, and the wash-out stability rate formula to obtain the characterization information data set.

[0016] Furthermore, the washout rate formula further includes:

[0017]

[0018] Wherein, the I p The T represents the peak signal enhancement ratio data included in the time-series signal strength data set. P The I0 characterizes the time corresponding to the peak signal enhancement ratio data, and the I0 characterizes the signal enhancement ratio data corresponding to the initial time included in the time-series signal strength data set.

[0019] Furthermore, the elution rate formula further includes,

[0020]

[0021] Wherein, the I last The signal enhancement ratio data corresponding to the termination time included in the time-series signal strength data set, and the I p The peak signal enhancement ratio data included in the time-series signal strength data set.

[0022] Furthermore, the elution stability formula further includes,

[0023]

[0024] Wherein, n represents the first total number of times included in the time-series signal intensity data set, and T n The I represents the termination time included in the time-series signal strength data set. p The peak signal enhancement ratio data included in the time-series signal strength data set represents the signal enhancement ratio peak data, α represents the average value of the signal enhancement ratio data included in the time-series signal strength data set, and I represents the peak signal enhancement ratio peak data. i The signal enhancement ratio data corresponding to time i included in the time-series signal strength data set.

[0025] Furthermore, after determining the heterogeneity prediction result information based on the set of representation information, the process further includes:

[0026] The identical heterogeneous prediction results in the heterogeneity prediction results are aggregated to obtain a second total corresponding to each identical heterogeneous prediction result; and

[0027] Based on the second total and the third total, the voxel composition ratio is determined, wherein the third total value is determined by the number of heterogeneity prediction result information.

[0028] On the other hand, embodiments of this specification also provide an apparatus for measuring hemodynamic heterogeneity, including,

[0029] The first determining unit is used to determine, based on the received dynamic contrast-enhanced magnetic resonance image, a set of time-series signal intensity data corresponding to pixels in the region of interest in the dynamic contrast-enhanced magnetic resonance image;

[0030] The first processing unit is used to process the time-series signal strength data set to obtain a characterization information data set;

[0031] The second processing unit is configured to process the representation information data using a classification function corresponding to the representation information data included in the representation information data set, to obtain the representation information set; and

[0032] The second determining unit is used to determine the heterogeneity prediction result information based on the set of characterization information.

[0033] On the other hand, embodiments of this specification also provide a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above-described method.

[0034] On the other hand, embodiments of this specification also provide a computer-readable storage medium having computer instructions stored thereon, which, when executed by a processor, implement the above-described method.

[0035] Using the embodiments of this specification, for a received dynamic contrast-enhanced magnetic resonance image, a time-series signal intensity data set for each pixel within the region of interest is determined; based on this time-series signal intensity data set, a characterization information data set is determined; each characterization information data is processed using a classification function corresponding to each characterization information data included in the characterization information data set to obtain the characterization information set; and then, based on this characterization information set, heterogeneity prediction results are determined. This achieves the prediction of blood heterogeneity based on the signal intensity data of each pixel within the region of interest of a dynamic contrast-enhanced magnetic resonance image and the acquisition time. This reduces errors caused by inter-individual differences, making the heterogeneity prediction results more accurate. Attached Figure Description

[0036] To more clearly illustrate the technical solutions in the embodiments or prior art of this specification, the drawings used in the description of the embodiments or prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this specification. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0037] Figure 1 The diagram shown is a schematic representation of an implementation system for a method of measuring hemodynamic heterogeneity according to an embodiment of this specification.

[0038] Figure 2 The diagram shown is a flowchart of a method for measuring hemodynamic heterogeneity according to an embodiment of this specification;

[0039] Figure 3 The diagram shown is a flowchart of a method for determining the numerical value of a timing signal strength data set according to an embodiment of this specification.

[0040] Figure 4 The diagram shown is a flowchart of a method for determining the voxel composition ratio according to an embodiment of this specification.

[0041] Figure 5A The diagram shown is a schematic diagram of a method for measuring hemodynamic heterogeneity according to another embodiment of this specification.

[0042] Figure 5B The diagram shown is a schematic representation of heterogeneity prediction results according to an embodiment of this specification.

[0043] Figure 5C The diagram shown is a schematic representation of heterogeneity prediction results according to another embodiment of this specification.

[0044] Figure 6A The diagram shown is a structural schematic of a device for measuring hemodynamic heterogeneity according to an embodiment of this specification.

[0045] Figure 6B The diagram shown is a structural schematic of a device for measuring hemodynamic heterogeneity according to another embodiment of this specification.

[0046] Figure 7 This is a schematic diagram of the structure of a computer device according to an embodiment of this specification.

[0047] [Explanation of Labels in the Attached Image]

[0048] 101. Data Acquisition Terminal;

[0049] 102. Server;

[0050] 103. User terminal;

[0051] 501. Multiple dynamic contrast-enhanced magnetic resonance images;

[0052] 502. Multiple time-series signal strength data sets;

[0053] 511. Wash-in rate formula;

[0054] 512. Washout rate formula;

[0055] 513. Formula for rinsing stability;

[0056] 520. Multiple sets of representational information data;

[0057] 530. Multiple sets of representational information;

[0058] 540. Information on multiple heterogeneity prediction results;

[0059] 550, multiple voxel composition ratio values;

[0060] 561. Washing rate sub-curve region;

[0061] 562. Washout rate sub-curve region;

[0062] 563. Wash out the stability sub-curve region;

[0063] 1. Slow-rise-stable curve;

[0064] 2. Slow-rising-unstable curve;

[0065] 3. Slow-plateau-stability curve;

[0066] 4. Slow-plateau-unstable curve;

[0067] 5. Slow-decline-stable curve;

[0068] 6. Slow-declining-unstable curve;

[0069] 7. Slow-rise-steady curve;

[0070] 8. Slow-rise-unstable curve;

[0071] 9. Slow-plateau-stability curve;

[0072] 10. Slow-plateau-unstable curve;

[0073] 11. Slow-decline-steady curve;

[0074] 12. Slow-decreasing-unstable curve;

[0075] 13. Rapid-rise-stable curve;

[0076] 14. Rapid-rise-unstable curve;

[0077] 15. Fast-Platform-Stability Curve;

[0078] 16. Fast-Plateau-Instability Curve;

[0079] 17. Rapid-decline-stable curve;

[0080] 18. Rapid-decreasing-unstable curve;

[0081] 19. No enhancement curve;

[0082] 610. First Determined Unit;

[0083] 620. First processing unit;

[0084] 630. Second processing unit;

[0085] 640. Second Determined Unit;

[0086] 650. Summary Unit;

[0087] 660. The third unit to be determined;

[0088] 702. Computer equipment;

[0089] 704. Processing equipment;

[0090] 706. Storage resources;

[0091] 708. Drive mechanism;

[0092] 710. Input / Output Module;

[0093] 712. Input devices;

[0094] 714. Output devices;

[0095] 716. Presentation equipment;

[0096] 718. Graphical User Interface;

[0097] 720. Network interface;

[0098] 722. Communication link;

[0099] 724. Communication bus. Detailed Implementation

[0100] The technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this specification, and not all embodiments. Based on the embodiments in this specification, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this specification.

[0101] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, apparatus, product, or device that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0102] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0103] In the technical solutions of this specification, the collection, storage, use, processing, transmission, provision, disclosure, and application of users' personal information and their transit address information comply with the provisions of relevant laws and regulations, necessary confidentiality measures have been taken, and they do not violate public order and good morals.

[0104] Figure 1The diagram illustrates an implementation system for a method of measuring hemodynamic heterogeneity according to an embodiment of this specification. The system may include a data acquisition terminal 101, a server 102, and a user terminal 103. The data acquisition terminal 101, server 102, and user terminal 103 communicate via a network, which may include a Local Area Network (LAN), a Wide Area Network (WAN), the Internet, or a combination thereof, and is connected to a website, user equipment (e.g., a computing device), and a backend system. The data acquisition terminal 101 includes a sensor. The data acquisition terminal 101 acquires signals through the sensor and sends the acquired information to the server 102. Upon receiving the signal, the server 102 generates a dynamic contrast-enhanced magnetic resonance image based on the information and determines the time-series signal intensity data set corresponding to pixels in the region of interest within the dynamic contrast-enhanced magnetic resonance image. Then, based on this time-series signal intensity data set, it determines heterogeneity prediction results and sends these results to the user terminal 103. It should be noted that after determining the heterogeneity prediction results, the voxel composition ratio can be determined based on the heterogeneity prediction results, and the voxel composition ratio can be sent to the user terminal 103.

[0105] Alternatively, server 102 may be a node of a cloud computing system (not shown in the figure), or each server 102 may be a separate cloud computing system comprising multiple computers interconnected by a network and operating as a distributed processing system.

[0106] In an optional embodiment, the user terminal 103 may include electronic devices, including but not limited to smartphones, data acquisition devices, desktop computers, tablets, laptops, smart speakers, digital assistants, augmented reality (AR) / virtual reality (VR) devices, smart wearable devices, and other similar electronic devices. Optionally, the operating system running on the electronic device may include, but is not limited to, Android, iOS, Linux, Windows, etc.

[0107] In addition, it should be noted that, Figure 1 The example shown is merely one application environment provided in this manual. In actual applications, it may include multiple user terminals 103, and this manual does not impose any restrictions.

[0108] like Figure 2The diagram shows a flowchart of a method for measuring hemodynamic heterogeneity according to an embodiment of this specification. This diagram illustrates the process of predicting blood heterogeneity, but may include more or fewer steps based on conventional or non-inventive work. The order of steps listed in the embodiment is merely one possible execution order among many and does not represent the only possible order. In actual system or device products, the method can be executed sequentially or in parallel according to the embodiment or the accompanying drawings. Specifically, as shown... Figure 2 As shown, the method may include:

[0109] S210, Based on the received dynamic contrast-enhanced magnetic resonance image, determine the set of time-series signal intensity data corresponding to the pixels in the region of interest in the dynamic contrast-enhanced magnetic resonance image;

[0110] S220, processes the time-series signal strength data set to obtain a characterization information data set;

[0111] S230, using the classification function corresponding to the representation information data included in the representation information data set, the representation information data is processed to obtain the representation information set;

[0112] S240, based on the set of representation information, determine the heterogeneity prediction results information.

[0113] Using the embodiments of this specification, for a received dynamic contrast-enhanced magnetic resonance image, a time-series signal intensity data set for each pixel within the region of interest is determined; based on this time-series signal intensity data set, a characterization information data set is determined; each characterization information data is processed using a classification function corresponding to each characterization information data included in the characterization information data set to obtain the characterization information set; and then, based on this characterization information set, heterogeneity prediction results are determined. This achieves the prediction of blood heterogeneity based on the signal intensity data of each pixel within the region of interest of a dynamic contrast-enhanced magnetic resonance image and the acquisition time. This reduces errors caused by inter-individual differences, making the heterogeneity prediction results more accurate.

[0114] According to one embodiment of this specification, dynamic contrast-enhanced magnetic resonance (MRI) images are images of target organs or tissues acquired sequentially based on a fast magnetic resonance sequence while a contrast agent is injected intravenously. The region of interest (ROI) is the area in the acquired MRI image that includes the target organ, and this region includes multiple pixels. Each pixel has corresponding signal intensity data and an acquisition time. This acquisition time can be, for example, characterization data or time data used to represent moments before and after the acquisition time. For example, if the signal intensity data corresponding to pixel A is m, and the acquisition time is 0, it represents that at time 0 (the initial time), the signal intensity of pixel A is m. It should be noted that the number corresponding to the initial time can be any number, and this specification does not limit this. For example, if MRI images of five moments are acquired within the time period from b:n to b:n+k, the five moments can be encoded, for example, 0, 1, 2, 3, 4 or 1, 2, 3, 4, 5, etc., or the real time corresponding to each moment can be used for encoding.

[0115] The dynamic contrast-enhanced magnetic resonance image is associated with the time when the dynamic contrast-enhanced magnetic resonance image was acquired.

[0116] After determining the region of interest (ROI), for each pixel within the ROI, the corresponding time and signal strength are determined. Signal strength data for pixels at the same location are then determined at multiple time points, and these multiple signal strength data points and their corresponding time points are aggregated to obtain a time-series signal strength data set corresponding to that pixel.

[0117] The representation data determination formula is used to process the temporal signal intensity data set corresponding to each pixel point to obtain the corresponding representation data set. This representation data determination formula includes a formula for processing the temporal signal intensity data set to obtain representation information data. This representation data determination formula may include one or more data determination formulas. Furthermore, for each data determination formula, a corresponding classification function is pre-configured. This classification function is a function that processes the representation information data obtained by the data determination formula to determine the representation information corresponding to that representation information data.

[0118] For example, when the formula for determining the representation data includes a first formula and a second formula, there exists a first classification function corresponding to the first formula and a second classification function corresponding to the second formula. When the time-series signal intensity data set corresponding to pixel A is (e,r,t,y,u,p,0,1,2,3,4), the first formula is used to process this time-series signal intensity data set to obtain representation information data R. The second formula is then used to process this time-series signal intensity data set to obtain representation data S. Furthermore, the first classification function is used to process the representation information data R to obtain corresponding first representation information, and the second classification function is used to process the representation information data S to obtain corresponding second representation information. Based on the first and second representation information, the corresponding representation information set is obtained as (first representation information, second representation information).

[0119] It should be noted that the first characterization information and the second characterization information can both be information on any characterization feature in the same class. For example, the first characterization information or the second characterization information can be fast rinsing, slow rinsing, slow rinsing, etc.

[0120] A corresponding preset curve image is pre-configured for each characterization information. For example, when the first characterization information is fast rinsing, slow rinsing, and sluggish rinsing, corresponding preset curve images are pre-configured for each of these three speeds.

[0121] The heterogeneity prediction result information is a time-signal intensity signal that can characterize the corresponding pixel, used to assist in determining heterogeneity. For example, the heterogeneity prediction result information can be time-signal intensity information (a set of characterization information). Alternatively, it can be a time-signal intensity curve corresponding to each set of characterization information.

[0122] In cases where the heterogeneity prediction result information can also be, for example, the time-signal intensity curve corresponding to each set of characterization information, the corresponding heterogeneity prediction result information is determined based on the set of characterization information. This includes obtaining a corresponding preset curve image for each characterization information included in the set of characterization information, and stitching the preset curve images together in a predetermined order to obtain a target curve image, and using the target curve image as the heterogeneity prediction result information.

[0123] For example, for each set of representation information, at least one preset curve image is determined, and these preset curve images are sorted in a predetermined order. Then, according to this sorting, the first and last images are stitched together to obtain the target curve image. When the set of representation information is (first representation information, second representation information), a preset curve image T corresponding to the first representation information and a preset curve image W corresponding to the second representation information are determined. When the predetermined order is first representation information - second representation information, the starting point of the curve included in the preset curve image W is connected to the ending point of the curve included in the preset curve image T to obtain the target curve image.

[0124] According to another embodiment of this specification, processing a time-series signal strength data set to obtain a characterization information data set includes: processing the data in the time-series signal strength data set using at least one of the wash-in rate formula, the wash-out rate formula, and the wash-out stability rate formula to obtain the characterization information data set.

[0125] In other words, the formulas for determining the above characterization data include at least one of the following: the formula for rinsing in rate, the formula for rinsing out rate, and the formula for rinsing stability rate.

[0126] Figure 3 The diagram shows a flowchart of a method for determining the numerical values ​​of a time-series signal strength dataset according to an embodiment of this specification. This diagram illustrates a process for determining a time-series signal strength dataset, but based on conventional or non-inventive methods, it may include more or fewer operational steps. Specifically, as shown... Figure 3 As shown, the method may include:

[0127] S311, Based on the received dynamic contrast-enhanced magnetic resonance image, determine the signal intensity data of each pixel in the region of interest in the dynamic contrast-enhanced magnetic resonance image;

[0128] S312, construct a time-series signal strength data set corresponding to each pixel based on the signal strength data corresponding to the pixel at the same position at each time.

[0129] According to another embodiment of this specification, the region of interest can be, for example, the region occupied by an organ in a dynamic contrast-enhanced magnetic resonance image, which includes multiple pixels. Before performing heterogeneity prediction on the dynamic contrast-enhanced magnetic resonance image, the user specifies the corresponding region of interest in the dynamic contrast-enhanced magnetic resonance image via a user terminal.

[0130] For example, the region of interest is the area occupied by organ Q in a dynamic contrast-enhanced magnetic resonance image. This region includes the first and second positions of organ Q. When acquiring dynamic contrast-enhanced magnetic resonance images at five time points, in the dynamic contrast-enhanced magnetic resonance image corresponding to the initial time 0, the first position corresponds to pixel c0 and the second position corresponds to pixel v0; in the dynamic contrast-enhanced magnetic resonance image corresponding to the second time 1, the first position corresponds to pixel c1 and the second position corresponds to pixel v1; in the dynamic contrast-enhanced magnetic resonance image corresponding to the third time 2, the first position corresponds to pixel c2 and the second position corresponds to pixel v2; in the dynamic contrast-enhanced magnetic resonance image corresponding to the fourth time 3, the first position corresponds to pixel c3 and the second position corresponds to pixel v3; and in the dynamic contrast-enhanced magnetic resonance image corresponding to the final time 4, the first position corresponds to pixel c4 and the second position corresponds to pixel v4. The signal intensity data corresponding to pixels c0, c1, c2, c3, and c4 corresponding to the first position are determined respectively, and the signal intensity data are summarized to obtain the time-series signal intensity data set corresponding to that pixel. When the signal strength data corresponding to pixels c0, c1, c2, c3, and c4 at the first position are g0, g1, g2, g3, and g4, the set of temporal signal strength data corresponding to the pixels at the first position is (g0, g1, g2, g3, g4, 0, 1, 2, 3, 4). Similarly, when the signal strength data corresponding to pixels v0, v1, v2, v3, and v4 at the second position are h0, h1, h2, h3, and h4, the set of temporal signal strength data corresponding to the pixels at the second position is (h0, h1, h2, h3, h4, 0, 1, 2, 3, 4). Therefore, when the region of interest includes both the first and second positions, two sets of temporal signal strength data are generated.

[0131] Figure 4 The diagram shows a flowchart of a method for determining a voxel composition ratio according to an embodiment of this specification. This diagram illustrates a process for determining a voxel composition ratio, but based on conventional or non-inventive methods, it may include more or fewer operational steps. Specifically, as shown... Figure 4 As shown, the method may include:

[0132] S450, summarize the identical heterogeneous prediction results in the heterogeneous prediction results information to obtain a second total corresponding to each identical heterogeneous prediction result information;

[0133] S460, determine the voxel composition ratio based on the second total and the third total, whereby the third total value is determined by the number of heterogeneity prediction results.

[0134] Using the embodiments of this specification, based on the heterogeneity prediction results determined for each pixel, the voxel composition ratio is determined to effectively assist in determining whether there is heterogeneity, thereby assisting in determining the pathological results corresponding to the dynamic contrast-enhanced magnetic resonance image.

[0135] According to another embodiment of this specification, the same heterogeneity prediction result information is the same set of characterization information or the same time-signal intensity curve. The third total value is determined by the number of heterogeneous prediction result information items; for example, the total number of heterogeneous prediction result information items can be used as the third total value.

[0136] After determining the second total and the third total, the second total and the third total are processed based on the following formula (1) to obtain the corresponding voxel composition ratio values.

[0137]

[0138] Among them, Type i The characterization is represented by the voxel composition ratio corresponding to the heterogeneity prediction result information i, num(i) is the second total number corresponding to the heterogeneity prediction result information i, N is the third total number, and x is a positive integer, which is determined by the different categories of heterogeneity prediction result information.

[0139] For example, signal strength data of four pixels were collected at five time points. This yields four time-series information strength data sets, which in turn yield four characterization information data sets, ultimately resulting in four heterogeneous prediction results. The third total is 4. If three of the four heterogeneous prediction results are the same (F), and one is E, then the second total corresponding to heterogeneous prediction result F is 3, and the second total corresponding to heterogeneous prediction result E is 1. Since there are a total of heterogeneous prediction results F and E, x is 2. Therefore, the Type1 corresponding to heterogeneous prediction result F is determined to be 0.75, and the Type2 corresponding to heterogeneous prediction result E is determined to be 0.25.

[0140] Figure 5A The diagram shown is a schematic of a method for measuring hemodynamic heterogeneity according to another embodiment of this specification. Figure 5B The diagram shown is a schematic representation of heterogeneity prediction results from an embodiment of this specification. Figure 5C The diagram shown is a schematic representation of heterogeneity prediction results according to another embodiment of this specification.

[0141] According to another embodiment of this specification, the classification function may include, for example, the mapping relationships shown in Table 1 below.

[0142] Table 1

[0143] Characteristic information data (washout rate value) Representation information (0.1,0.5] slow (0.5,1] slow (1,+∞) fast (-∞,0.1] No enhancement

[0144] According to another embodiment of this specification, the classification function may also include, for example, the mapping relationships shown in Table 2 below.

[0145] Table 2

[0146] Characteristic information data (washout rate value) Representation information (0.05,+∞) rise [-0.05,+0.05] platform (-∞,-0.05) decline

[0147] According to another embodiment of this specification, the classification function may also include, for example, the mapping relationships shown in Table 3 below.

[0148] Table 3

[0149] Characteristic information data (washout stability value) Representation information (0,0.05] Stablize (0.05,+∞) Unstable

[0150] It should be noted that the classification function is determined by the formulas in Tables 1, 2, and 3 based on the characterization data. For example, if the formula for determining the characterization data includes the wash-in rate formula, the classification function is the mapping relationship included in Table 1; if the formula for determining the characterization data includes the wash-out rate formula and the wash-out stability rate formula, the classification function is the mapping relationship included in Tables 2 and 3, and the wash-out rate formula is associated with the mapping relationship included in Table 2, while the wash-out stability rate formula is associated with the mapping relationship included in Table 3; if the formula for determining the characterization data includes the wash-in rate, wash-out rate formula, and wash-out stability rate formula, the classification function is the mapping relationship included in Tables 1, 2, and 3, and the wash-in rate formula is associated with the mapping relationship included in Table 1, the wash-out rate formula is associated with the mapping relationship included in Table 2, and the wash-out stability rate formula is associated with the mapping relationship included in Table 3.

[0151] According to another embodiment of this specification, the wash-in rate formula includes the following formula (2).

[0152]

[0153] Among them, I p The peak signal enhancement ratio (T) data included in the time-series signal strength dataset characterizes the signal enhancement ratio of the dataset. P The peak time corresponding to the signal enhancement ratio data is represented by I0, and the initial time corresponding to the initial time included in the time-series signal strength data set is represented by I0.

[0154] According to the following formula (3), the signal strength data included in the time series signal strength data set are processed to obtain the signal enhancement ratio data corresponding to each signal strength data.

[0155]

[0156] Among them, S i Let I be the signal strength data at time i. iTo match the signal strength data S i The corresponding signal enhancement ratio data, S0 is the signal strength data corresponding to the initial time.

[0157] After processing the signal strength data in the time-series signal strength dataset to obtain the corresponding signal enhancement ratio data, the original time-series signal strength data is updated by replacing the signal strength data in the dataset with this signal enhancement ratio data. Thus, each time step is associated with the corresponding signal enhancement ratio data.

[0158] From multiple signal enhancement ratio data in a time-series signal strength dataset that includes signal enhancement ratio data, the data with the largest signal enhancement ratio is identified as the peak signal enhancement ratio data.

[0159] According to another embodiment of this specification, the washout rate formula includes the following formula (4).

[0160]

[0161] Among them, I last The signal enhancement ratio data corresponding to the termination time included in the time-series signal strength dataset, and I p Peak signal enhancement ratio data included in a time-series signal strength dataset.

[0162] According to another embodiment of this specification, the elution stability formula includes the following formula (5).

[0163]

[0164] Where n represents the first total number of times included in the time-series signal intensity data set, and T n The termination time included in the time-series signal strength dataset, I p The peak signal enhancement ratio data included in the time-series signal strength dataset represents the peak value of the signal enhancement ratio data included in the time-series signal strength dataset, α represents the average value of the signal enhancement ratio data included in the time-series signal strength dataset, and I i The signal enhancement ratio data corresponding to time i included in the time series signal strength dataset.

[0165] like Figure 5AAs shown, after determining multiple dynamic contrast-enhanced magnetic resonance images 501, pixels at the same location are determined from the region of interest in each dynamic contrast-enhanced magnetic resonance image, and then the temporal signal intensity data set of that pixel is determined, resulting in multiple temporal signal intensity data sets 502. Each temporal signal intensity data set in the multiple temporal signal intensity data sets 502 is input into the washing rate formula 511, the washing out rate formula 512, and the washing out stability formula 513, respectively, to determine the representation information data set corresponding to each temporal signal intensity data set, resulting in multiple representation information data sets 520. Based on the classification function corresponding to the representation information data in each representation information data set, the data in the multiple representation information data sets 520 are processed to obtain multiple representation information sets 530. Then, based on the multiple representation information sets 530, the target curve image corresponding to each representation information set is determined, and each target curve image is used as heterogeneity prediction result information, resulting in multiple heterogeneity prediction result information 540. Then, using the above formula (1), the multiple heterogeneity prediction results 540 are processed to obtain multiple voxel composition ratio values ​​550.

[0166] It should be noted that when the multiple voxel composition ratios obtained in this embodiment are input into the machine learning model for predicting heterogeneity, the prediction accuracy is greater than 80%. This accuracy is higher than that of the three-class classification method based on the mean time-signal intensity over the region of interest; it is also superior to the prediction results based on the mean signal intensity over the region of interest using only the deep learning model for predicting heterogeneity, specifically showing an improvement of 5%-20% or more in each evaluation metric. These evaluation metrics are used to assess the accuracy of the heterogeneity prediction results.

[0167] Furthermore, based on the embodiments of this specification, a method for visually reflecting the dynamic heterogeneity within a lesion is provided for the time-signal intensity curve of each pixel. This method fully utilizes the properties of the time-signal intensity curve and has the following advantages: 1. Low computational time complexity, eliminating the need for extensive calculations and curve fitting, resulting in faster calculation speed; 2. Compared to methods that average out the dynamic heterogeneity within a lesion by taking the mean time-signal intensity curve, the method of the embodiments of this specification is more sensitive to dynamic heterogeneity and provides higher accuracy in assisting heterogeneity judgment.

[0168] According to another embodiment of this specification, such as Figure 5BAs shown, when the formulas for determining the characterization data include the formulas for rinsing in, rinsing out, and rinsing out stability, the classification function is the mapping relationship included in Tables 1, 2, and 3, and the heterogeneity prediction result information is the time-signal intensity curve corresponding to each characterization information set, for each characterization information set, a corresponding preset curve image is obtained, and the preset curve images are stitched together in a predetermined order to obtain a target curve image. The target curve image is used as the heterogeneity prediction result information, including: determining the angle of the corresponding rinsing in sub-curve region 561 based on the rinsing in formula and the rinsing in characterization information obtained in Table 1; determining the width of the corresponding rinsing out sub-curve region 562 based on the rinsing out formula and the rinsing out characterization information obtained in Table 2; and determining the height of the corresponding rinsing out stability sub-curve region 563 based on the rinsing out stability formula and the rinsing out stability characterization information obtained in Table 3.

[0169] Based on the determined input rate, output rate, and rinsing stability rate characterization information, and the corresponding angle of the input rate sub-curve region 561, the width of the output rate sub-curve region 562, and the height of the rinsing stability rate sub-curve region 563, a corresponding preset curve image is determined. The preset curve images are then stitched together in the order of input rate sub-curve, output rate, and rinsing stability rate sub-curve to obtain the target curve image. It should be noted that when the characterization data determination formula includes the input rate and output rate formulas, the angle of the corresponding input rate sub-curve region 561 is determined based on the input rate formula and the input rate characterization information obtained from Table 1; the width of the corresponding output rate sub-curve region 562 is determined based on the output rate formula and the output rate characterization information obtained from Table 2. Then, based on the order of the rinsing rate sub-curve and the rinsing rate sub-curve, the corresponding preset curve images are stitched together to obtain the target curve image.

[0170] According to another embodiment of this specification, when the characterization data determination formula includes the wash-in rate, wash-out rate formula, and wash-out stability rate formula, the classification function is the mapping relationship included in Tables 1, 2, and 3, and the wash-in rate formula is associated with the mapping relationship included in Table 1, the wash-out rate formula is associated with the mapping relationship included in Table 2, the wash-out stability rate formula is associated with the mapping relationship included in Table 3, and the heterogeneity prediction result information is the time-signal intensity curve corresponding to each characterization information set, the mapping relationship with the time-signal intensity curve corresponding to each characterization information set is shown in Table 4 below. Specifically, the mapping relationship between the wash-in rate characterization information, the wash-out rate characterization information, and the wash-out stability rate characterization information, and the angle of the wash-in rate sub-curve region 561, the width of the wash-out rate sub-curve region 562, and the height of the wash-out stability rate sub-curve region 563 is shown in Table 4 below.

[0171] Table 4

[0172]

[0173] It should be noted that the numbers in Table 4 are different from those in Table 4. Figure 5C The images corresponding to the same numbers in the table are associated (mapped). Specifically, "1" in Table 4 is associated with... Figure 5C The "slow-rising-stable curve 1" in the diagram has a mapping relationship. That is, given the representation information set as (slow, rising, stable), the target curve image (heterogeneity prediction result information) is... Figure 5C The slow-rise-stable curve 1 in Table 4. Similarly, "2" in Table 4 is related to... Figure 5C The “slow-rising-unstable curve 2” in Table 4 has a mapping relationship; “3” in Table 4 is related to… Figure 5C The “slow-plateau-stable curve 3” in Table 4 has a mapping relationship; “4” in Table 4 is related to… Figure 5C The “slow-plateau-unstable curve 4” in Table 4 has a mapping relationship; “5” in Table 4 is related to… Figure 5C The “slow-decline-stable curve 5” in Table 4 has a mapping relationship; “6” in Table 4 is related to… Figure 5C The “slow-decline-unstable curve 6” in Table 4 has a mapping relationship; “7” in Table 4 is related to… Figure 5C The “slow-rise-stable curve 7” in Table 4 has a mapping relationship; “8” in Table 4 is related to… Figure 5C The “slow-rise-unstable curve 8” in Table 4 has a mapping relationship; “9” in Table 4 is related to… Figure 5C The “slow-plateau-stable curve 9” in Table 4 has a mapping relationship; “10” in Table 4 is related to… Figure 5C The “slow-plateau-unstable curve 10” in Table 4 has a mapping relationship; “11” in Table 4 is related to… Figure 5C The “slow-decline-stable curve 11” in Table 4 has a mapping relationship; “12” in Table 4 is related to… Figure 5C The “slow-decline-unstable curve 12” in Table 4 has a mapping relationship; “13” in Table 4 is related to… Figure 5C The “rapid-rise-stable curve 13” in Table 4 has a mapping relationship; “14” in Table 4 is related to… Figure 5C The “rapid-rise-unstable curve 14” in Table 4 has a mapping relationship; “15” in Table 4 is related to… Figure 5C The “fast-plateau-stable curve 15” in Table 4 has a mapping relationship; “16” in Table 4 is related to… Figure 5C The “fast-plateau-unstable curve 16” in Table 4 has a mapping relationship; “17” in Table 4 is related to… Figure 5C The “rapid-decline-stable curve 17” in Table 4 has a mapping relationship; “18” in Table 4 is related to… Figure 5CThe “rapid-decline-unstable curve 18” in the table has a mapping relationship; and “19” in Table 4 has a mapping relationship with… Figure 5C The "unenhanced curve 19" in the figure has a mapping relationship. It should be noted that, for... Figure 5C Each curve in the graph can be assigned a different color.

[0174] Figure 6A The diagram shown is a structural schematic of a device for measuring hemodynamic heterogeneity according to an embodiment of this specification. Figure 6A As shown, including,

[0175] The first determining unit 610 is used to determine, based on the received dynamic contrast-enhanced magnetic resonance image, a set of time-series signal intensity data corresponding to pixels in the region of interest in the dynamic contrast-enhanced magnetic resonance image;

[0176] The first processing unit 620 is used to process the time-series signal strength data set to obtain a characterization information data set;

[0177] The second processing unit 630 is used to process the characterization information data using a classification function corresponding to the characterization information data included in the characterization information data set, so as to obtain the characterization information set.

[0178] The second determining unit 640 is used to determine the heterogeneity prediction result information based on the set of representation information.

[0179] Since the principle of the above-mentioned device in solving the problem is similar to that of the above-mentioned method, the implementation of the above-mentioned device can refer to the implementation of the above-mentioned method, and the repeated parts will not be described again.

[0180] Figure 6B The diagram shown is a structural schematic of a device for measuring hemodynamic heterogeneity according to another embodiment of this specification. Figure 6B As shown, including,

[0181] The aggregation unit 650 is used to aggregate the same heterogeneous prediction results in the heterogeneous prediction results information to obtain a second total corresponding to each of the same heterogeneous prediction results information.

[0182] The third determining unit 660 is used to determine the voxel composition ratio value based on the second total and the third total, wherein the third total value is determined by the number of heterogeneity prediction result information.

[0183] Since the principle of the above-mentioned device in solving the problem is similar to that of the above-mentioned method, the implementation of the above-mentioned device can refer to the implementation of the above-mentioned method, and the repeated parts will not be described again.

[0184] like Figure 7The diagram illustrates the structure of a computer device according to an embodiment of this specification. The apparatus described in this specification can be the computer device in this embodiment, performing the methods described above. The computer device 702 may include one or more processing devices 704, such as one or more central processing units (CPUs), each of which can implement one or more hardware threads. The computer device 702 may also include any storage resource 706 for storing information of any kind, such as code, settings, data, etc. Without limitation, for example, the storage resource 706 may include any one or more combinations of the following: any type of RAM, any type of ROM, flash memory, hard disk, optical disk, etc. More generally, any storage resource can use any technology to store information. Furthermore, any storage resource can provide volatile or non-volatile retention of information. Further, any storage resource may represent a fixed or removable component of the computer device 702. In one case, when the processing device 704 executes associated instructions stored in any storage resource or combination of storage resources, the computer device 702 can perform any operation of the associated instructions. The computer device 702 also includes one or more drive mechanisms 708 for interacting with any storage resource, such as a hard disk drive mechanism, an optical disk drive mechanism, etc.

[0185] Computer device 702 may also include an input / output module 710 (I / O) for receiving various inputs (via input device 712) and providing various outputs (via output device 714). A specific output mechanism may include a presentation device 716 and an associated graphical user interface (GUI) 718. In other embodiments, the input / output module 710 (I / O), input device 712, and output device 714 may be omitted, and the device may function solely as a computer device within a network. Computer device 702 may also include one or more network interfaces 720 for exchanging data with other devices via one or more communication links 722. One or more communication buses 724 couple the components described above together.

[0186] Communication link 722 can be implemented in any way, such as via a local area network, a wide area network (e.g., the Internet), a point-to-point connection, or any combination thereof. Communication link 722 may include any combination of hardwired links, wireless links, routers, gateway functions, name servers, etc., governed by any protocol or combination of protocols.

[0187] This specification also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method.

[0188] This specification also provides a computer program product, which includes a computer program that, when executed by a processor, implements the above-described method.

[0189] Those skilled in the art will understand that embodiments of this specification can be provided as methods, systems, or computer program products. Therefore, this specification may take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this specification may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0190] This specification is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this specification. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create a machine for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0191] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0192] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0193] The above specific embodiments further illustrate the purpose, technical solutions, and beneficial effects of this specification. It should be understood that the above are merely specific embodiments of this specification and are not intended to limit the scope of protection of this specification. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this specification should be included within the scope of protection of this specification.

Claims

1. A method for measuring hemodynamic heterogeneity, characterized in that, include: Based on the received dynamic contrast-enhanced magnetic resonance image, determine the set of time-series signal intensity data corresponding to the pixels in the region of interest in the dynamic contrast-enhanced magnetic resonance image; The time-series signal strength data set is processed to obtain a characterization information data set; The characterization information data is processed using a classification function corresponding to the characterization information data included in the characterization information data set to obtain the characterization information set. as well as Based on the aforementioned set of characterization information, the heterogeneity prediction results are determined. The processing of the time-series signal strength data set to obtain the characterization information data set includes: The data in the time-series signal strength data set is processed by at least one of the wash-in rate formula, the wash-out rate formula, and the wash-out stability rate formula to obtain the characterization information data set. The wash-in rate formula includes: Wherein, the I p The T represents the peak signal enhancement ratio data included in the time-series signal strength data set. P The I0 characterizes the time corresponding to the peak signal enhancement ratio data, and the I0 characterizes the signal enhancement ratio data corresponding to the initial time included in the time-series signal strength data set. The elution rate formula includes: Wherein, the I last The signal enhancement ratio data corresponding to the termination time included in the time-series signal strength data set, and the I p The peak signal enhancement ratio data included in the time-series signal strength data set; The formula for elution stability includes: Wherein, n represents the first total number of times included in the time-series signal intensity data set, and T n The I represents the termination time included in the time-series signal strength data set. p The peak signal enhancement ratio data included in the time-series signal strength data set represents the signal enhancement ratio peak data, α represents the average value of the signal enhancement ratio data included in the time-series signal strength data set, and I represents the peak signal enhancement ratio peak data. i The signal enhancement ratio data corresponding to time i included in the time-series signal strength data set.

2. The method according to claim 1, characterized in that, The step of determining the time-series signal intensity data set corresponding to pixels in the region of interest of the received dynamic contrast-enhanced magnetic resonance image includes: Based on the received dynamic contrast-enhanced magnetic resonance image, determine the signal intensity data of each pixel in the region of interest within the dynamic contrast-enhanced magnetic resonance image; and Based on the signal strength data corresponding to the pixel at the same position at each time, construct the time-series signal strength data set corresponding to each pixel.

3. The method according to claim 1, characterized in that, After determining the heterogeneity prediction result information based on the set of representation information, the method further includes: The identical heterogeneous prediction results in the heterogeneity prediction results are aggregated to obtain a second total corresponding to each identical heterogeneous prediction result; and Based on the second total and the third total, the voxel composition ratio is determined, wherein the third total value is determined by the number of heterogeneity prediction result information.

4. A device for measuring hemodynamic heterogeneity, characterized in that, include: The first determining unit is used to determine, based on the received magnetic resonance image, a set of time-series signal intensity data corresponding to pixels in the region of interest in the magnetic resonance image; The first processing unit is used to process the time-series signal strength data set to obtain a characterization information data set. The second processing unit is used to process the representation information data using a classification function corresponding to the representation information data included in the representation information data set, so as to obtain the representation information set. as well as The second determining unit is used to determine the heterogeneity prediction result information based on the set of characterization information. The first processing unit processes the data in the time-series signal strength data set by using at least one of the wash-in rate formula, the wash-out rate formula, and the wash-out stability rate formula to obtain the characterization information data set. The wash-in rate formula includes: Wherein, the I p The T represents the peak signal enhancement ratio data included in the time-series signal strength data set. P The I0 characterizes the time corresponding to the peak signal enhancement ratio data, and the I0 characterizes the signal enhancement ratio data corresponding to the initial time included in the time-series signal strength data set. The elution rate formula includes: Wherein, the I last The signal enhancement ratio data corresponding to the termination time included in the time-series signal strength data set, and the I p The peak signal enhancement ratio data included in the time-series signal strength data set; The formula for elution stability includes: Wherein, n represents the first total number of times included in the time-series signal intensity data set, and T n The I represents the termination time included in the time-series signal strength data set. p The peak signal enhancement ratio data included in the time-series signal strength data set represents the signal enhancement ratio peak data, α represents the average value of the signal enhancement ratio data included in the time-series signal strength data set, and I represents the peak signal enhancement ratio peak data. i The signal enhancement ratio data corresponding to time i included in the time-series signal strength data set.

5. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method of any one of claims 1-3.

6. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, performs the method of any one of claims 1-3.

Images