Method and device for determining arterial input function curve and computer equipment

Abstract

The application relates to a method and device for determining an arterial input function curve and a computer device, wherein the method for determining the arterial input function curve comprises the following steps: acquiring CT perfusion (CTP) scanning data and CT angiography (CTA) scanning data of an imaging device for a region of interest, and the CTP scanning region and the CTA scanning region have an overlapping region; determining an initial arterial input function curve based on the CTP scanning data; and supplementing the initial arterial input function curve based on the overlapping region data to obtain a target arterial input function curve, wherein the overlapping region is CTA scanning data of the overlapping region. Through the application, the problem that the arterial input function curve is inaccurate is solved, and the technical effect of improving the accuracy of the arterial input function curve is achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the medical imaging field, in particular to a method and device for determining an arterial input function curve and a computer device. BACKGROUND

[0002] CTA (Computed Tomography Angiography, blood vessel computed tomography) technology and CTP (Computed Tomography Perfusion, perfusion computed tomography) technology are important means for checking stroke-related diseases at present. The medical images obtained by scanning the brain using CTA and CTP can be collectively referred to as brain CT medical images. The brain arterial CTA image can assist in diagnosing brain arterial vascular stenosis and occlusion and other disease information, and the brain CTP image can assist in diagnosing brain tissue blood supply information.

[0003] Through automatic positioning and tracking technology, the cranial CTP and head and neck CTA data are obtained simultaneously in one scan, which reduces the contrast agent dose while ensuring that the cranial CTP and head and neck CTA scan is in the best state, which can make the subsequent CTP and CTA post-processing obtain more stable and accurate results. However, due to the addition of head and neck CTA at the arterial peak period, part of the information of the cranial CTP rising to the peak period is lost, resulting in an inaccurate arterial input function curve, which affects the accuracy of the subsequent perfusion parameter calculation results.

[0004] At present, there is no effective solution to the problem of inaccurate arterial input function curve in the related art. SUMMARY

[0005] In this embodiment, a method, device and computer equipment for determining an arterial input function curve are provided to solve the problem of inaccurate arterial input function curve in the related art.

[0006] In a first aspect, a method for determining an arterial input function curve is provided in this embodiment. The method includes: obtaining CTP scan data and CTA scan data of an imaging device for a region of interest, and the CTP scan region and the CTA scan region have an overlapping region; determining an initial arterial input function curve based on the CTP scan data; and supplementing the initial arterial input function curve based on the overlapping region data to obtain a target arterial input function curve, the overlapping region being CTA scan data of the overlapping region.

[0007] In one of the embodiments, the supplementing the initial arterial input function curve based on the overlap region data to obtain the target arterial input function curve comprises: determining a reference arterial curve based on the overlap region data; determining a correlation parameter of the reference arterial curve and the initial arterial input function curve; determining a supplement curve based on the correlation parameter and the reference arterial curve; and supplementing the initial arterial input function curve based on the supplement curve to obtain the target arterial input function curve.

[0008] In one of the embodiments, the determining the correlation parameter of the reference arterial curve and the initial arterial input function curve comprises: constructing a weight array based on the reference arterial curve and the initial arterial input function curve; performing linear interpolation on the reference arterial curve, the initial arterial input function curve and the weight array to obtain a standard reference arterial curve, a standard initial arterial input function curve and a standard weight array; and processing the standard reference arterial curve, the standard initial arterial input function curve and the standard weight array by using a numerical optimization method to determine the correlation parameter.

[0009] In one of the embodiments, the supplementing the initial arterial input function curve based on the overlap region data to obtain the target arterial input function curve comprises: inputting the overlap region data and the initial arterial input function curve into a trained neural network to obtain a supplement curve; and supplementing the initial arterial input function curve based on the supplement curve to obtain the target arterial input function curve.

[0010] In one of the embodiments, the inputting the overlap region data and the initial arterial input function curve into the trained neural network to obtain the supplement curve comprises: determining a reference arterial curve based on the overlap region data; extracting curve features of the reference arterial curve and the initial arterial input function curve; determining a feature vector based on data and the curve features of each time point of the reference arterial curve and the initial arterial input function curve; and inputting the feature vector into the trained neural network to obtain the supplement curve.

[0011] In one of the embodiments, the acquiring the CTP scan data and the CTA scan data of the region of interest comprises: performing a first dose scan on the region of interest, and determining CTP scan parameters and CTA scan parameters based on the first dose scan result, wherein the CTA scan parameters comprise a scan trigger time and a scan start position; performing a second dose CTP scan and a second dose CTA scan on the region of interest based on the CTP scan parameters and the CTA scan parameters; when the scan trigger time of the CTA scan is reached, controlling a scan bed to move to a preset position, and performing a second dose CTA scan on the region of interest based on the CTA scan parameters, with the scan start position as a starting point; and obtaining the CTP scan data and the CTA scan data of the region of interest based on the second dose CTP scan result and the second dose CTA scan result.

[0012] In one of the embodiments, after the initial arterial input function curve is supplemented based on the overlap region data to obtain the target arterial input function curve, the method further comprises: determining perfusion parameters based on the target arterial input function curve.

[0013] In a second aspect, an apparatus for determining an arterial input function curve is provided in the embodiments, and the apparatus comprises:

[0014] an acquiring module configured to acquire CTP scan data and CTA scan data of a region of interest acquired by an imaging device, wherein a CTP scan region and a CTA scan region have an overlap region;

[0015] an initial curve determining module configured to determine an initial arterial input function curve based on the CTP scan data;

[0016] a supplementing module configured to supplement the initial arterial input function curve based on overlap region data to obtain a target arterial input function curve, wherein the overlap region data is CTA scan data of the overlap region.

[0017] In a third aspect, a computer device is provided in the embodiments, and the computer device comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the arterial input function curve determining method of the first aspect when executing the computer program.

[0018] In a fourth aspect, a storage medium is provided in the embodiments, and the storage medium stores a computer program, and the computer program is executable on a processor to implement the arterial input function curve determining method of the first aspect.

[0019] Compared with the related art, the method, device and computer equipment for determining an arterial input function curve provided in this embodiment, by acquiring CTP scan data and CTA scan data of an image device for a region of interest, and the CTP scan region and the CTA scan region have an overlapping region, determining an initial arterial input function curve based on the CTP scan data, and supplementing the initial arterial input function curve based on overlapping region data to obtain a target arterial input function curve, the overlapping region data being CTA scan data of the overlapping region, the missing part of the initial arterial input function curve is supplemented by processing the CTA data of the overlapping region, the problem of inaccurate arterial input function curve is solved, and the technical effect of improving the accuracy of the arterial input function curve is achieved.

[0020] The details of one or more embodiments of the present application are presented in the following drawings and description to make other features, objects and advantages of the present application more apparent. BRIEF DESCRIPTION OF DRAWINGS

[0021] The drawings described herein are intended to provide further understanding of the present application, and constitute a part of the present application. The illustrative embodiments of the present application and their description serve to explain the present application, and do not constitute an improper limitation on the present application. In the drawings:

[0022] Figure 1 is a hardware structure block diagram of a terminal of the method for determining an arterial input function curve of this embodiment;

[0023] Figure 2 is a flowchart of the method for determining an arterial input function curve of this embodiment;

[0024] Figure 3 is a schematic diagram of a region of interest in the related art;

[0025] Figure 4 is a schematic diagram of an overlapping region according to an embodiment of the present application;

[0026] Figure 5 is a schematic diagram of scan data acquisition according to an embodiment of the present application;

[0027] Figure 6 is a schematic diagram of an unsupervised arterial input function curve supplementing process according to an embodiment of the present application;

[0028] Figure 7 is a schematic diagram of a supervised arterial input function curve supplementing process according to another embodiment of the present application;

[0029] Figure 8 is a structure block diagram of the determining device for an arterial input function curve of this embodiment. DETAILED DESCRIPTION

[0030] In order to clearly understand the purpose, technical solutions and advantages of the present application, the present application is described and explained below in conjunction with the accompanying drawings and embodiments.

[0031] Unless otherwise defined, technical terms or scientific terms used in the present application shall have the general meaning understood by a person skilled in the art to which the present application belongs. In the present application, "one", "a", "an", "the", "these" and similar words do not represent a quantitative limitation, and they can be singular or plural. In the present application, the terms "include", "contain", "have" and any variants thereof are intended to cover non-exclusive inclusion; for example, a process, method and system, product or device containing a series of steps or modules (units) are not limited to the listed steps or modules (units), but can include steps or modules (units) not listed, or can include other steps or modules (units) inherent to the process, method, product or device. In the present application, the terms "connected", "connected", "coupled" and the like do not limit to physical or mechanical connection, but can include electrical connection, whether direct or indirect. In the present application, "multiple" means two or more. The association between the associated objects is described by "and / or", which means that there can be three relationships, for example, "A and / or B" can mean that A exists alone, A and B exist together, and B exists alone. In general, the character " / " represents the relationship between the front and rear associated objects as "or". In the present application, the terms "first", "second", "third" and the like are only used to distinguish similar objects, and do not represent a specific order for the objects.

[0032] The method embodiments provided in the present embodiment can be executed in a terminal, a computer or a similar computing device. For example, the method embodiments are executed on a terminal, Figure 1 is a hardware structure block diagram of the terminal of the determination method of the arterial input function curve of the present embodiment. As shown in Figure 1 , the terminal can include one or more (only one is shown in Figure 1 ) processor 102 and memory 104 for storing data, wherein the processor 102 can include but not limited to processing devices such as microprocessor MCU or programmable logic device FPGA. The above terminal can also include transmission device 106 for communication function and input / output device 108. Those skilled in the art can understand that Figure 1 The structure shown is only schematic, which does not limit the structure of the above terminal. For example, the terminal can include more or less components than Figure 1 shown, or have a different configuration from Figure 1 shown.

[0033] The memory 104 can be used to store computer programs, such as software programs of application software and modules, such as the computer program corresponding to the arterial input function curve determination method in the present embodiment. The processor 102 can execute various functional applications and data processing, i.e., implement the method described above, by running the computer program stored in the memory 104. The memory 104 can include a high-speed random access memory, and can further include a non-volatile memory, such as one or more magnetic storage devices, flash memories, or other non-volatile solid-state memories. In some examples, the memory 104 can further include a memory remotely arranged with respect to the processor 102, which can be connected to the terminal through a network. Examples of the network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0034] The transmission device 106 is configured to receive or send data via a network. The network includes a wireless network provided by a communication provider of the terminal. In an example, the transmission device 106 includes a network interface controller (NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In an example, the transmission device 106 can be a radio frequency (RF) module, which is configured to communicate with the Internet in a wireless manner.

[0035] In recent years, perfusion has been increasingly widely used in the diagnosis, prognosis, and efficacy evaluation of tumors, especially brain tumors. Perfusion imaging can be used to quantitatively analyze the microvascular distribution and blood perfusion state of tissues, and can provide diagnosis for brain perfusion abnormal diseases such as stroke, and can also provide effective reference for the formulation of treatment decisions. For example, in clinical practice, CT (Computer Tomography) and CTA (Computed Tomography Angiography) are usually combined to diagnose stroke, but there is a large radiation dose. CT perfusion imaging (CTP) is to collect dynamic images at multiple time points in a selected region to record the change of contrast agent concentration in the tissue of the region over time. In theory, CTP contains the information of CTA, and the blood vessel region can be obtained by analyzing the segmentation of CTP, so as to provide blood vessel information for diagnosis and reduce the radiation dose of patients to a certain extent.

[0036] By means of automatic positioning and tracking technology, the CTP and CTA data of the head and neck can be obtained simultaneously in one scan. While reducing the dose of contrast agent, the CTP and CTA scans of the head and neck are in the best state, which is conducive to obtaining more stable and accurate results in subsequent CTP and CTA post-processing. However, due to the addition of head and neck CTA at the peak of the artery, the information of the intracranial CTP will be lost, which will affect the calculation results of the user's concerned parameters. In actual operation, in order to make the head and neck CTA in the best state, the contrast agent filling state of the head and neck CTA needs to be captured, the scanning bed needs to be moved to the neck, the CTP scan needs to be switched to the head and neck CTA scan, and then the subsequent CTP scan needs to be performed. Due to the limitation of hardware equipment, this scanning method will sacrifice the information of the rising period of the arterial input curve of CTP, especially the information of intracranial arteries, which will be more than 4 seconds, while the CTP calculation generally requires a time interval of less than 2 seconds. Therefore, the accuracy of the subsequent concerned parameter calculation will be affected.

[0037] In this embodiment, a method for determining an arterial input function curve is provided, Figure 2 is a flowchart of the method for determining an arterial input function curve in this embodiment, as Figure 2 shown, the flow includes the following steps:

[0038] In step S201, CTP scan data and CTA scan data of a region of interest are obtained, the CTP scan data and the CTA scan data are obtained by CTP and CTA joint scanning, and the CTP scan region and the CTA scan region have an overlapping region.

[0039] By way of example, the region of interest refers to the region to be scanned, which is generally specified by a technician and includes the lesion position of the subject. The region of interest can be manually selected by the technician or automatically selected by machine through deep learning or other means. CTP, i.e. CT perfusion scanning, is a method of using CT to scan the brain multiple times while injecting contrast agent into the vein of the subject, then reconstructing by computer, and finally obtaining the change of the concentration of the contrast agent in the brain tissue, reflecting the change of the brain perfusion. CTA, also known as CT angiography, is a method of injecting contrast agent into the vein of the subject, and reconstructing the three-dimensional image of the blood vessels by computer processing. In order to reduce the influence of contrast agent on the subject and improve the scanning efficiency, the CTP scan and CTA scan are generally combined in the related art to obtain the scan data. Taking the combined scanning process of head and neck CTA and CTP as an example, Figure 3 is a schematic diagram of the region of interest in the related art, as Figure 3As shown, the scanning direction of the imaging device is from top to bottom, and specifically to the body part of the subject, i.e., from the head to the neck. After the scanning starts, the imaging device first scans the P1 region, i.e., the first region of interest, which is arranged at the brain of the subject, and mainly performs the combined scanning of CTP and CTA on the cerebral arteries to obtain the CTP scanning data and the CTA scanning data of the region. The P2 region, i.e., the second region of interest, is arranged at the neck of the patient, and only CTA scanning is performed on the second region of interest. This is because the head and neck CTA scanning needs to collect the CTA scanning data of the head of the patient and the CTA scanning data of the neck of the patient. However, the intracranial CTP only needs to scan the intracranial region. During the combined scanning, because the contrast agent injected intravenously flows from the carotid artery to the brain along with the blood, the head and neck CTA of the subject will first reach the filling state. At this time, in order to capture the CTA scanning data in this state, the scanning bed is moved to the neck region of interest for CTA scanning. After the CTA scanning is completed, the scanning bed is moved back to the first region of interest for combined scanning. After the brain CTP reaches the filling state, the collection of the brain CTP scanning data is completed. However, in the related art, during the period from the movement of the scanning bed to the neck region of interest for CTA scanning to the movement of the scanning bed to the brain region of interest for CTP scanning, the rising period data of the arterial input function curve of the brain CTP will be lost, resulting in the loss of part of the data. In this embodiment, in order to solve the problem of data loss, the overlapping region is set for the CTA scanning region and the CTP scanning region when the region of interest is set. The CTA scanning data and the CTP scanning data of the adjusted region of interest are collected to provide data support for the supplement of the arterial input function curve.

[0040] In this embodiment, the process of the combined scanning of the brain CTP and the head and neck CTA is also taken as an example, Figure 4 is a schematic diagram of the overlapping region according to the embodiment of the present application, as Figure 4 shown, P3 is the third region of interest arranged at the subject, and the combined scanning of CTP and CTA is performed in the third region of interest to obtain the brain CTP scanning data and the brain CTA scanning data of the subject. P4 is the fourth region of interest arranged at the neck of the subject. The fourth region of interest can be divided into two parts. One is the P5 region, which is the overlapping region of the P3 region and the P4 region, and the combined scanning of CTP and CTA is performed in the region. This process can also be referred to as overlapping scanning. The P6 region performs CTA scanning to obtain the neck CTA scanning data of the subject. Based on the division of the scanning region, CTA scanning is performed in both the P3 region and the P4 region to obtain the head and neck CTA scanning data of the subject, and CTP scanning is performed in the P3 region to obtain the brain CTP scanning data of the subject, thereby ensuring the integrity of the scanning data.

[0041] Figure 5 This is a schematic diagram of scanning data acquisition according to an embodiment of this application, such as... Figure 5 As shown, following the chronological order, the third region of interest, P3, is scanned first. Once the intracervical cavity is filled with contrast agent, the fourth region of interest, P4, is scanned, resulting in CTP scan data for the overlapping region P5 and CTA scan data for region P6. The scan data from regions P3 and P4 are then integrated using stitching technology. The presence of overlapping regions improves stitching efficiency and accuracy. Through data integration, complete head and neck CTA scan data of the examined target is finally obtained.

[0042] Step S202: Determine the initial arterial input function curve based on the CTP scan data.

[0043] For example, data is acquired from the region of interest in a CTP scan to obtain CTP scan data. An initial arterial input function curve is determined based on the CTP scan data. Preferably, the arterial input function of the healthy side middle hemisphere or anterior cerebral artery is used for CTP calculation. The healthy side refers to the side opposite to the affected side; specifically, in a cranial CTP scan, the healthy side middle hemisphere refers to the middle artery of the healthy side of the brain. The arterial input function, abbreviated as AIF, is used in the calculation of the arterial input function AIF. During the calculation of the arterial input function AIF, T... max The parameters are highly dependent on the location of the artery. If CTP data from the carotid artery is taken, T... max The parameters will increase significantly, and the absolute quantification will be affected. Furthermore, because the arterial input function curve obtained based on the current CTP scan data lacks information during the rising phase, it is called the initial arterial input function curve.

[0044] Step S203: The initial arterial input function curve is supplemented based on the overlapping region data to obtain the target arterial input function curve, wherein the overlapping region data is the CTA scan data of the overlapping region.

[0045] For example, such as Figure 4 As shown, in region P5, which is the overlapping region, both CTA and CTP scan data of the target were acquired. However, since CTP calculations generally use data from the healthy middle cervical artery, and this overlapping region is located in the carotid artery region, the CTP scan data from this overlapping region cannot be directly applied to the calculation of the arterial input function curve. Therefore, the CTA scan data from the overlapping region is used as the overlapping region data for calculation. The overlapping region data can be obtained by performing an overlapping axial scan on the overlapping region, i.e., by performing an axial scan or helical scan on the overlapping region.

[0046] Axial scan, also called sequential scan, step-and-shoot scan or step-and- acquire mode. In axial scan mode, the scan bed is first controlled to step to the position where imaging is needed, the scan bed stops moving; the tube and the detector rotate to acquire data; then the tube and the detector stop working, the scan bed moves a preset distance and then stops, the tube and the detector rotate again and acquire data; the above process is repeated until the scan is completed.

[0047] Helical scan, also called spiral scan, because the scan trajectory is a spiral line, in the scanning process, the X-ray tube continuously exposes around the gantry, at the same time of exposure, the scan bed moves at a constant speed synchronously, and the detector acquires data at the same time. When facing a case where the size of the region of interest is large, helical scan can be used to complete data acquisition.

[0048] Taking head and neck CTA combined with brain CTP scan as an example, the overlap region data is the CTA scan data of the overlap region, which is because in the process of combined scanning of CTP and CTA, the time-density curve of the neck region of interest about CT value will reach the peak period earlier than that of the brain region of interest, that is, the CTA scan of the neck region of interest needs to be performed first, therefore, the CTA scan data about the neck region of interest at this time is taken as the overlap region data; another important reason is that the CTP calculation generally selects the data of the healthy middle cerebral artery for calculation, and the overlap region includes part of the internal carotid artery, directly applying the CTP scan data of this part will make the parameter T max of the arterial input function of the intracranial artery obtained by calculation larger, and the absolute quantification will be affected, therefore, the CTA scan data of the overlap region is selected as the overlap region data.

[0049] In addition, an important property of the overlap region data is that the data can ensure the alignment of the sampling points of the CTA data and the CTP data. In the scanning scheme in the related art, due to the limitation of the hardware device, the information of the rising period of the arterial input function curve of the intracranial artery of CTP is sacrificed, the time interval reaches 4 seconds or more, while the CTP calculation generally requires a time interval of 2 seconds or more. By using the scanning mode of the overlap region in the embodiment, the CTA data can be acquired with a proper time interval, and the alignment can be performed through the registration of the CTA scan data and the CTP scan data. For example: when performing head and neck CTA scan, the scanning mode with two overlapping axial scans is used to complete, that is, the curve of the internal carotid artery with a time interval of about 2 seconds is obtained. Combined with the CTP scan data, the prior knowledge that the contrast agent change of the intracranial artery is affected by the contrast agent change of the neck artery is used to supplement the information of the rising period of the intracranial artery.

[0050] Through the above steps, the arterial input function curve acquisition method proposed in this embodiment adjusts the position of the region of interest, so that the region of interest of the CTA scan and the region of interest of the CTP scan contain an overlapping region. With the aid of the overlapping region data, the missing part of the arterial input function curve is supplemented by using the prior knowledge that there is a correlation between the contrast agent concentration changes between the overlapping region and the missing region of interest, thereby solving the problem of low accuracy of the arterial input function curve, and improving the integrity and accuracy of the arterial input function curve.

[0051] In one of the embodiments, Figure 6 is a schematic diagram of an unsupervised arterial input function curve supplement process according to an embodiment of the present application, as Figure 6 shown, the process includes:

[0052] Step S601, determining a reference arterial curve based on the overlapping region data;

[0053] Step S602, determining the correlation relationship parameters of the reference arterial curve and the initial arterial input function curve;

[0054] Step S603, determining a supplement curve based on the correlation relationship parameters and the reference arterial curve;

[0055] Step S604, supplementing the initial arterial input function curve based on the supplement curve to obtain a target arterial input function curve.

[0056] Exemplarily, after the CTP scan data and the CTA scan data are acquired, the CTP scan data and the CTA scan data are aligned using a registration algorithm, and the best internal carotid artery curve of the overlapping region is automatically or manually acquired from the CTA data as a reference arterial curve, wherein the best internal carotid artery curve refers to a standard flow-in and flow-out curve, i.e., a curve with four stages of a plateau period, a rising period, a falling period and a stable period, and with a high peak and smooth curve. The best curve of the healthy side middle artery is automatically or manually acquired from the CTP data as an initial arterial input function curve. Since the contrast agent flow into the intracranial artery comes from the contrast agent flow into the internal carotid artery, there is a strong connection between them. Due to the difference in position and the diffusion effect, there are changes in delay and amplitude, so a calculation model between the supplement curve and the reference arterial curve can be established, which includes a correlation relationship coefficient. The correlation relationship coefficient is calculated by a mathematical method, and then the supplement curve is determined. Finally, the initial arterial input function curve is supplemented according to the supplement curve to obtain a target arterial input function curve.

[0057] In one of the embodiments, the determining the correlation parameter between the reference arterial curve and the initial arterial input function curve comprises: constructing a weight array based on the reference arterial curve and the initial arterial input function curve, the data quantity of the weight array being equal to the data quantity of the reference arterial curve; performing linear interpolation on the reference arterial curve, the initial arterial input function curve and the weight array to obtain a standard reference arterial curve, a standard initial arterial input function curve and a standard weight array, the time intervals of the data of the standard reference arterial curve, the standard initial arterial input function curve and the standard weight array being equal; and processing the standard reference arterial curve, the standard initial arterial input function curve and the standard weight array by using a numerical optimization method to determine the correlation parameter, wherein the numerical optimization method comprises a nonlinear least square method.

[0058] Specifically, the reference arterial curve is denoted as C ref , the time array corresponding to the reference arterial curve is taken as a first time array and denoted as T ref , the initial arterial input function curve is denoted as C aif , and the time array corresponding to the initial arterial input function curve is taken as a second time array and denoted as T aif . The second time array T aif is one time point less than the first time array T ref , the index position is denoted as I del , and the values of the other time points are equal.

[0059] Based on the above data, the unsupervised arterial input function curve supplement method specifically comprises:

[0060] First, a weight array W ref of the same length as the reference arterial curve is constructed. The time point position and quantity of the weight array W ref are the same as those of the first time array T ref , and the weight array W aif is one time point more than the second time array T aif . According to the weight array, the weight value of the position of the second time array T ref which is one time point less than the first time array T ref is assigned a small value, and the weight values of the other time point positions are respectively assigned large values of the same value. The large value and the small value refer to the relative size of the values, for example, the small value is set to 0.0001 and the large value is set to 1. In this embodiment, the size of the value is not specifically limited and can be adjusted according to the actual needs of calculation and the accuracy of the device.

[0061] Then, linear interpolation is used to obtain a standard reference arterial curve, a standard initial arterial input function curve and a standard weight array from the reference arterial curve C ref , the initial arterial input function curve C aif and the weight array W refLinear interpolation is performed, and the precision of the interpolation can be configured as needed, for example, to 1 second, to obtain the interpolated standard reference arterial curve C ref_1 , the standard initial arterial input function curve C aif_1 , and the standard weight array W ref_1 .

[0062] Since the intracranial arterial contrast agent inflow is from the internal carotid arterial contrast agent inflow, there is a strong relationship between the two. Due to the difference in position and diffusion, there is a delay and amplitude change, so the following model can be constructed:

[0063]

[0064] Where a, b, delay are parameters to be solved, representing the change parameters from the reference arterial curve to the complementary curve, i.e., the correlation parameter. is a convolution operation. The following optimization problem is solved iteratively using a nonlinear least squares method:

[0065]

[0066] The correlation parameters a, b, delay are obtained, and Linear interpolation is used again to interpolate back to the first time array T ref , to obtain

[0067] I del in is supplemented back to C aif , to construct a new C aif_n The arterial input function curve with the completed information is used for subsequent perfusion parameter calculation.

[0068] In one of the embodiments, Figure 7 is a supervised arterial input function curve completion process according to another embodiment of the present application, as shown in Figure 7 , the process includes:

[0069] Step S701, input the overlapping region data and the initial arterial input function curve into the trained neural network to obtain a complementary curve.

[0070] Step S702, based on the complementary curve, the initial arterial input function curve is completed to obtain a target arterial input function curve.

[0071] Through the supervised arterial input curve completion method in this embodiment, the pre-trained neural network is used for completion, and the curve completion speed is fast. In addition, as the training set data of the neural network is continuously improved, the accuracy of the neural network can be further improved, thereby improving the accuracy of the final target arterial input function curve.

[0072] In one of the embodiments, the coincident region data and the initial arterial input function curve are input into the trained neural network to obtain the supplemented curve, including: determining a reference arterial curve based on the coincident region data; extracting curve features of the reference arterial curve and the initial arterial input function curve; determining a feature vector based on the data at each time point of the reference arterial curve and the initial arterial input function curve and the curve features; inputting the feature vector into the trained neural network to obtain the supplemented curve.

[0073] Specifically, a supervised arterial input function curve completion method is provided by means of deep learning, which includes: constructing training data by normal whole brain CTP data, i.e. obtaining paired reference arterial curve C ref and corresponding arterial input function curve C aif_all , both of which have the same length N. C aif_all is subjected to CTA and CTP joint scanning and simulation to obtain arterial input function curve C aif , the simulation process is to delete the time points before the peak value time point to form a time interval greater than 4 seconds, and the deleted time points can be 1 or 2, which can be set according to the actual situation. Some features are extracted from the reference arterial curve and the arterial input function curve, such as contrast agent arrival time difference BAT diff , peak time difference TTP diff and other curve features.

[0074] Optionally, each time point of C ref is combined with the extracted features to construct a feature vector at one time point, or C aif is linearly interpolated and combined with each time point of C ref to construct a feature vector at one time point, or the extracted features are combined, and the feature vector related to the time sequence is denoted as F i , i = 1...N.

[0075] The feature vector is used as input and C aif_all is used as a supervised target to train a network model using a time sequence related deep learning network such as a recurrent neural network or a Transformer related network, which is not specifically limited here. The sequence output by the network is

[0076] The loss function is a similarity measure between and C aif_all , and optionally:

[0077]

[0078] wherein, is the weight information, and optionally, since the target is to obtain the missing information, the weight ratio can be increased at the missing index.

[0079] After obtaining the completion model, the real acquired CTA and CTP combined scanning data are processed and input into the network to obtain

[0080] The I del The numerical value is supplemented back to C aif , a new C aif_n The arterial input function curve as the completion information is used for subsequent perfusion parameter calculation.

[0081] In one of the embodiments, the acquiring CTP scanning data and CTA scanning data of the region of interest comprises: positioning the region of interest to determine the position of the region of interest; performing first dose scanning on the region of interest, and determining CTP scanning parameters and CTA scanning parameters based on the first dose scanning result, wherein the CTA parameters comprise a trigger time and a scanning start position; performing second dose CTP scanning and CTA scanning on the region of interest based on the CTP scanning parameters and the CTA scanning parameters; when the scanning trigger time of the CTA scanning is reached, controlling the scanning bed to move to a preset position, and performing second dose CTA scanning on the region of interest based on the CTA scanning parameters, with the scanning start position as the starting point; and obtaining the CTP scanning data and CTA scanning data of the region of interest based on the second dose CTP scanning result and the CTA scanning result.

[0082] For example, in the case where the subject does not inject contrast agent, the subject is scanned to obtain its anatomical data, and the regions of interest (ROIs) of the internal carotid arteries on both sides of the slice are positioned automatically or manually on the anatomical data. The ROIs can be positioned automatically by using deep learning. After the technician determines the current state of the patient, the patient is injected with a small amount of agent, and the ROIs are tracked in the single-slice scan, i.e., the first-dose scan, to monitor the concentration of the contrast agent and determine the arterial input function in the ROIs. After a period of time, for example, after three monitoring periods, when the arterial input detection curves in the ROIs all cross the peak value, the ROI tracking is stopped, and the period of time is taken as the scanning period. According to the arterial input function detected by the ROIs, the optimal CTP scanning time and sampling form and the CTA trigger time are automatically planned by using a rule or a machine learning method. Then, the technician confirms the state of the subject again, injects the subject with a normal dose of contrast agent, and performs a second-dose scan. The starting position of the second-dose scan is located in the scanning range of the CTP scan. The second-dose scan process includes: first, performing the CTP scan, and when the CTA trigger time point is reached, moving the scanning bed to a specified position in the overlap region. Performing the head and neck CTA scan, using overlapping axial scanning, and then performing subsequent CTP scanning until the automatic planning scan is completed, to obtain the CTP and CTA data. Based on the CTP and CTA data obtained by scanning, the images are reconstructed to obtain the CTP and CTA images of the scanned part of the patient.

[0083] In one of the embodiments, after the initial arterial input function curve is supplemented based on the overlap region data to obtain the target arterial input function curve, the method further includes: determining the perfusion parameters based on the target arterial input function curve.

[0084] It should be noted that the steps shown in the above process or the flowchart of 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 can be executed in an order different from that shown here.

[0085] In this embodiment, an input function determination device is also provided, which is used to implement the above-described embodiments and preferred embodiments, and will not be described again. The terms "module", "unit", "sub-unit", etc. used below can be a combination of software and / or hardware that implements a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware or a combination of software and hardware is also possible and contemplated.

[0086] Figure 8 is a structural block diagram of the arterial input function curve determination device of the present embodiment, asFigure 8 The device comprises:

[0087] The acquisition module 10 is configured to acquire CTP scan data and CTA scan data of a region of interest, the CTP scan data and the CTA scan data being obtained through joint CTP and CTA scanning, and the CTP scan region and the CTA scan region having an overlapping region;

[0088] The initial curve determination module 20 is configured to determine an initial arterial input function curve based on the CTP scan data.

[0089] The supplement module 30 is configured to supplement the initial arterial input function curve based on overlapping region data to obtain a target arterial input function curve, the overlapping region data being CTA scan data of the overlapping region.

[0090] The supplement module 30 is further configured to determine a reference arterial curve based on the overlapping region data, determine a correlation parameter of the reference arterial curve and the initial arterial input function curve, determine a supplement curve based on the correlation parameter and the reference arterial curve, and supplement the initial arterial input function curve based on the supplement curve to obtain the target arterial input function curve.

[0091] The supplement module 30 is further configured to construct a weight array based on the reference arterial curve and the initial arterial input function curve, the weight array having a same number of data as the reference arterial curve, perform linear interpolation on the reference arterial curve, the initial arterial input function curve, and the weight array to obtain a standard reference arterial curve, a standard initial arterial input function curve, and a standard weight array, the standard reference arterial curve, the standard initial arterial input function curve, and the standard weight array having a same time interval of data, and process the standard reference arterial curve, the standard initial arterial input function curve, and the standard weight array using a nonlinear least square method to determine the correlation parameter.

[0092] The supplement module 30 is further configured to input the overlapping region data and the initial arterial input function curve into a trained neural network to obtain the supplement curve, and supplement the initial arterial input function curve based on the supplement curve to obtain the target arterial input function curve.

[0093] The supplement module 30 is further configured to determine a reference arterial curve based on the overlapping region data, extract curve features of the reference arterial curve and the initial arterial input function curve, determine a feature vector based on data at each time point of the reference arterial curve and the initial arterial input function curve and the curve features, and input the feature vector into a trained neural network to obtain the supplement curve.

[0094] The determination apparatus of the arterial input function curve is further configured to perform a first dose scan on the region of interest, determine CTP scan parameters and CTA scan parameters based on the first dose scan result, the CTA scan parameters including a scan trigger time and a scan start position, perform a second dose CTP scan and a CTA scan on the region of interest based on the CTP scan parameters and the CTA scan parameters, control the scan bed to move to a preset position when the scan trigger time of the CTA scan is reached, and perform a second dose CTA scan on the region of interest based on the CTA scan parameters from the scan start position.

[0095] The supplement module 30 is further configured to determine the perfusion parameters based on the target arterial input function curve.

[0096] It should be noted that each of the above modules can be a functional module or a program module, and can be implemented by software or hardware. For the modules implemented by hardware, each of the above modules can be located in the same processor, or each of the above modules can be located in different processors in any combination.

[0097] In this embodiment, a computer device is also provided, including a memory and a processor, the memory storing a computer program, and the processor being configured to execute the computer program to perform the steps in any of the above method embodiments.

[0098] Optionally, the computer device can further include a transmission device and an input / output device, wherein the transmission device is connected to the processor, and the input / output device is connected to the processor.

[0099] Optionally, in this embodiment, the processor can be configured to execute the following steps through the computer program:

[0100] S1, acquiring CTP scan data and CTA scan data of a region of interest by an imaging device, and the CTP scan region and the CTA scan region having an overlapping region.

[0101] S2, determining an initial arterial input function curve based on the CTP scan data.

[0102] S3, supplementing the initial arterial input function curve based on the overlapping region data to obtain a target arterial input function curve, the overlapping region being CTA scan data of the overlapping region.

[0103] It should be noted that the specific examples in the present embodiment can refer to the examples described in the above embodiments and optional implementation manners, which will not be described herein again.

[0104] In addition, in combination with the determination method of the arterial input function curve provided in the above embodiments, a storage medium can also be provided in the present embodiment to realize. The storage medium has a computer program stored thereon; the computer program is executed by a processor to realize the determination method of the arterial input function curve in any one of the above embodiments.

[0105] It should be noted that the user information (including but not limited to user equipment information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the present application are all information and data authorized by the user or authorized by all parties.

[0106] It should be understood that the specific embodiments described herein are only used to explain this application, but not to limit it. According to the embodiments provided in the present application, all other embodiments obtained by those of ordinary skill in the art without creative labor are within the scope of protection of the present application.

[0107] Obviously, the drawings are only some examples or embodiments of the present application, and those of ordinary skill in the art can also apply the present application to other similar situations according to the drawings without creative labor. In addition, it can be understood that although the work done in the development process may be complex and long, but for those of ordinary skill in the art, some design, manufacture or production changes according to the technical content disclosed in the present application are only routine technical means, and should not be regarded as insufficient disclosure of the present application.

[0108] The term "embodiment" in the present application means that the specific features, structures or characteristics described in combination with the embodiments can be included in at least one embodiment of the present application. The appearance of this phrase in various places in the specification does not necessarily mean the same embodiment, nor does it mean independence or alternative to other embodiments. Those of ordinary skill in the art can clearly or implicitly understand that the embodiments described in the present application can be combined with other embodiments without conflict.

[0109] The above described embodiments only express several implementation manners of the present application, and the description is more specific and detailed, but it should not be understood as a limitation on the scope of patent protection. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, a number of modifications and improvements can be made, which are within the scope of protection of the present application. Therefore, the protection scope of the present application should be subject to the appended claims.

Claims

1. A method for determining an arterial input function curve, characterized in that, The method includes: Acquire CTP scan data and CTA scan data of the imaging equipment targeting the region of interest; the region of interest includes the brain region of interest and the neck region of interest of the examined target; the brain region of interest and the neck region of interest have overlapping areas; the overlapping area is located in the carotid artery region; the CTP scan data includes brain CTP scan data; The initial arterial input function curve is determined based on the CTP scan data; The initial arterial input function curve is supplemented based on the overlapping region data to obtain the target arterial input function curve, wherein the overlapping region data is the CTA scan data of the overlapping region; The process of supplementing the initial arterial input function curve based on overlapping region data to obtain the target arterial input function curve includes: Based on the overlapping region data and the initial arterial input function curve, a supplementary curve is determined; Based on the supplemented curve, the initial arterial input function curve is supplemented to obtain the target arterial input function curve; The step of determining the supplementary curve based on the overlapping region data and the initial arterial input function curve includes: A reference arterial curve is determined based on the data from the overlapping region. Determine the correlation parameters between the reference arterial curve and the initial arterial input function curve; The supplementary curve is determined based on the correlation parameters and the reference arterial curve; Alternatively, determining the supplementary curve based on the overlapping region data and the initial arterial input function curve includes: The overlapping region data and the initial arterial input function curve are input into a trained neural network to obtain the supplementary curve.

2. The method according to claim 1, characterized in that, The parameters for determining the correlation between the reference arterial curve and the initial arterial input function curve include: A weight array is constructed based on the reference arterial curve and the initial arterial input function curve; Linear interpolation is performed on the reference arterial curve, the initial arterial input function curve, and the weight array to obtain the standard reference arterial curve, the standard initial arterial input function curve, and the standard weight array. The correlation parameters are determined based on the standard reference arterial curve, the standard initial arterial input function curve, and the standard weight array.

3. The method according to claim 1, characterized in that, The step of inputting the overlapping region data and the initial arterial input function curve into a trained neural network to obtain the supplementary curve includes: A reference arterial curve is determined based on the data from the overlapping region. Extract the curve features of the reference arterial curve and the initial arterial input function curve; A feature vector is determined based on the data at each time point of the reference arterial curve and the initial arterial input function curve, as well as the curve features. The feature vector is input into a trained neural network to obtain the supplementary curve.

4. The method according to any one of claims 1-3, characterized in that, The acquisition of CTP scan data and CTA scan data of the imaging device for the region of interest includes: A first dose scan is performed on the region of interest, and CTP scan parameters and CTA scan parameters are determined based on the first dose scan results. The CTA scan parameters include the scan trigger time and the scan start position. Based on the CTP and CTA scanning parameters, a second dose of CTP and CTA scanning is performed on the region of interest. When the scan trigger time of the CTA scan is reached, the scan bed is controlled to move to a preset position, and a second dose of CTA scan is performed on the region of interest based on the CTA scan parameters, with the scan start position as the starting point. Based on the CTP scan results and CTA scan results of the second dose, the CTP scan data and CTA scan data of the region of interest are obtained.

5. The method according to claim 1, characterized in that, After supplementing the initial arterial input function curve based on overlapping region data to obtain the target arterial input function curve, the process further includes: Perfusion parameters are determined based on the target arterial input function curve.

6. A device for determining an arterial input function curve, characterized in that, The device includes: The acquisition module is used to acquire CTP scan data and CTA scan data of the imaging device for the region of interest, wherein the CTP scan area and the CTA scan area overlap; the region of interest includes the brain region of interest and the neck region of interest of the examined target; the brain region of interest and the neck region of interest overlap; the overlapping area is located in the carotid artery region; the CTP scan data includes brain CTP scan data; An initial curve determination module is used to determine an initial arterial input function curve based on the CTP scan data; The completion module is used to complete the initial arterial input function curve based on the overlapping region data to obtain the target arterial input function curve, wherein the overlapping region data is the CTA scan data of the overlapping region; The supplementation module is further configured to determine a supplementation curve based on the overlapping region data and the initial arterial input function curve; and to supplement the initial arterial input function curve based on the supplementation curve to obtain the target arterial input function curve. The step of determining the supplementary curve based on the overlapping region data and the initial arterial input function curve includes: A reference arterial curve is determined based on the data from the overlapping region. Determine the correlation parameters between the reference arterial curve and the initial arterial input function curve; The supplementary curve is determined based on the correlation parameters and the reference arterial curve; Alternatively, determining the supplementary curve based on the overlapping region data and the initial arterial input function curve includes: The overlapping region data and the initial arterial input function curve are input into a trained neural network to obtain the supplementary curve.

7. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 5.

Images