Methods, devices, and storage media for determining tissue window evaluation values ​​in stroke

Abstract

This invention discloses a method, apparatus, and storage medium for determining the evaluation value of a tissue window in stroke, relating to the field of medical imaging technology, and primarily aimed at improving the evaluation accuracy of the tissue window in stroke. The method includes: based on the DWI and ADC sequences of the examined object, obtaining the first target evaluation region R in the DWI sequence. dwi Based on the FLAIR sequence of the object under inspection and the first target evaluation region R dwi Obtain the second target evaluation region R infarction The second objective evaluation region R infarction To compare with the first target evaluation region R dwi High-signal regions of overlapping FLAIR sequences; based on the first target evaluation region R dwi The corresponding first evaluation volume and second target evaluation region R infarction The corresponding second evaluation volume is used to calculate the stroke tissue window evaluation value of the examined subject.

Description

Technical Field

[0001] This invention relates to the field of medical imaging technology, and in particular to a method, apparatus, and storage medium for determining the evaluation value of tissue windows in stroke patients. Background Technology

[0002] Stroke is currently one of the leading causes of death and disability worldwide, and it has become the leading cause of death and disability among Chinese adults. Stroke is caused by the rupture or blockage of blood vessels in the brain, leading to hypoxia and ischemia, resulting in localized brain tissue necrosis. This necrosis is usually irreversible. One of the key decision-making factors in the specific treatment of acute stroke patients is the size of the core infarct area. It is generally believed that patients with large core infarcts have already suffered irreversible necrosis of most of their ischemic brain tissue, and may not benefit from effective treatment. However, with continuous technological innovation, experts in the field have conducted in-depth analysis of the evaluation model of the core infarct area. In clinical practice, it has been found that not all cells in the preoperative core infarct area are irreversibly necrotic; there exists a portion of "severely ischemic tissue with uncertain viability," namely the ischemic penumbra, which can be restored through treatment. Timely restoration of blood flow can fully restore the physiological function of the ischemic penumbra. When treating stroke patients, if there is no salvageable ischemic penumbra, opening the blood vessel is not only ineffective but may also increase the risk of hyperperfusion. Therefore, the purpose of introducing imaging standards is to identify tissue regions that may benefit from thrombolytic therapy, i.e., tissue windows. Consequently, tissue window evaluation has gradually entered the field of stroke.

[0003] Currently, tissue windows are typically evaluated by determining the degree of mismatch between DWI (Diffusion Weighted Imaging) and PWI (Perfusion Weighted Imaging). However, PWI scanning is complex, and due to varying physician skill levels, errors can occur. Furthermore, PWI imaging is heavily reliant on the blood-brain barrier; if the blood-brain barrier is damaged or incomplete, the contrast agent's movement within the brain tissue is affected, leading to PWI imaging errors. Additionally, existing techniques using DWI to identify the core infarct area may suffer from negative reversion errors. Therefore, the aforementioned methods cannot accurately determine the degree of mismatch between DWI and PWI, resulting in low accuracy in tissue window evaluation. Summary of the Invention

[0004] This invention provides a method, apparatus, and storage medium for determining the evaluation value of the tissue window in stroke, which mainly improves the evaluation accuracy of the tissue window in stroke.

[0005] According to a first aspect of the present invention, a method for determining the tissue window evaluation value of stroke is provided, comprising:

[0006] Based on the DWI and ADC sequences of the object under inspection, the first target evaluation region R in the DWI sequence is obtained. dwi The first target evaluation region R dwi The high-signal region of the DWI sequence that overlaps with the low-signal region of the ADC sequence.

[0007] Based on the FLAIR sequence of the inspected object and the first target evaluation region R dwi Obtain the second target evaluation region R infarction The second target evaluation region R infarction To be consistent with the first target evaluation region R dwi High-signal regions of the FLAIR sequences that overlap in position;

[0008] Based on the first target evaluation region R dwi The corresponding first evaluation volume and the second target evaluation region R infarction The corresponding second evaluation volume is used to calculate the stroke tissue window evaluation value of the examined object.

[0009] According to a second aspect of the present invention, an apparatus for determining the tissue window evaluation value of stroke is provided, comprising:

[0010] The first acquisition unit is used to acquire the first target evaluation region R in the DWI sequence based on the DWI sequence and ADC sequence of the object being inspected. dwi The first target evaluation region R dwi The high-signal region of the DWI sequence that overlaps with the low-signal region of the ADC sequence.

[0011] The second acquisition unit is used to acquire data based on the FLAIR sequence of the object being inspected and the first target evaluation region R. dwi Obtain the second target evaluation region R infarction The second target evaluation region R infarction To be consistent with the first target evaluation region R dwi High-signal regions of the FLAIR sequences that overlap in position;

[0012] The calculation unit is used to calculate based on the first target evaluation region R. dwi The corresponding first evaluation volume and the second target evaluation region R infarction The corresponding second evaluation volume is used to calculate the stroke tissue window evaluation value of the examined object.

[0013] According to a third aspect of the present invention, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, performs the following steps:

[0014] Based on the DWI and ADC sequences of the object under inspection, the first target evaluation region R in the DWI sequence is obtained. dwi The first target evaluation region R dwi The high-signal region of the DWI sequence that overlaps with the low-signal region of the ADC sequence.

[0015] Based on the FLAIR sequence of the inspected object and the first target evaluation region R dwi Obtain the second target evaluation region R infarction The second target evaluation region R infarction To be consistent with the first target evaluation region R dwi High-signal regions of the FLAIR sequences that overlap in position;

[0016] Based on the first evaluation volume corresponding to the first target evaluation region Rdwi and the second target evaluation region R infarction The corresponding second evaluation volume is used to calculate the stroke tissue window evaluation value of the examined object.

[0017] The present invention provides a method, apparatus, storage medium, and computer device for determining the evaluation value of a tissue window in stroke. Compared with the current method of evaluating tissue windows by determining the matching degree between DWI and PWI, the present invention obtains the first target evaluation region R in the DWI sequence based on the DWI sequence and ADC sequence of the examined object. dwi The first target evaluation region R dwi The high-signal region of the DWI sequence overlaps with the low-signal region of the ADC sequence; and is based on the FLAIR sequence of the object under inspection and the first target evaluation region R. dwi Obtain the second target evaluation region R infarction The second target evaluation region R infarction To be consistent with the first target evaluation region R dwi The high-signal regions of the FLAIR sequences that overlap in position; ultimately based on the first target evaluation region R dwi The corresponding first evaluation volume and the second target evaluation region R infarction The corresponding second evaluation volume is used to calculate the stroke tissue window evaluation value of the examined subject, thereby determining the first target evaluation region R in the DWI sequence. dwi The second target evaluation region R of the FLAIR sequence infarction Then calculate the first target evaluation region R. dwiThe corresponding first evaluation volume and second target evaluation region R infarction The corresponding second evaluation volume, and finally based on the first and second evaluation volumes, determines the evaluation value of the stroke tissue window of the examined object. This avoids the situation in the prior art where PWI imaging errors are caused by damage or incompleteness of the blood-brain barrier. At the same time, it also avoids the error of using DWI to determine the core infarct area, which leads to the inability to accurately obtain the degree of mismatch between DWI and PWI, thereby improving the evaluation accuracy of the tissue window. Attached Figure Description

[0018] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings:

[0019] Figure 1 A flowchart of a method for determining the tissue window evaluation value in stroke, provided by an embodiment of the present invention, is shown.

[0020] Figure 2 This invention provides a flowchart of another method for determining the tissue window evaluation value in stroke, according to an embodiment of the present invention.

[0021] Figure 3 This diagram illustrates the structure of a device for determining the tissue window evaluation value of stroke according to an embodiment of the present invention.

[0022] Figure 4 A schematic diagram of the physical structure of a computer device provided in an embodiment of the present invention is shown. Detailed Implementation

[0023] The present invention will be described in detail below with reference to the accompanying drawings and embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in the present application can be combined with each other.

[0024] Currently, the method of determining the degree of matching between DWI and PWI to evaluate tissue windows results in the inability to accurately obtain the degree of mismatch between DWI and PWI, thus leading to low accuracy in tissue window evaluation.

[0025] To address the aforementioned problems, embodiments of the present invention provide a method for determining the tissue window evaluation value in stroke, such as... Figure 1 As shown, the method includes:

[0026] 101. Based on the DWI and ADC sequences of the object under inspection, obtain the first target evaluation region R in the DWI sequence. dwi The first target evaluation region R dwi The high-signal region of the DWI sequence that overlaps with the low-signal region of the ADC sequence.

[0027] The specific subjects examined can be stroke patients. DWI sequence is a diffusion-weighted imaging sequence, and ADC sequence is an apparent diffusion coefficient sequence. DWI and ADC sequences can be obtained by scanning the head of the subject. DWI sequence can show the degree of diffusion of water molecules in tissues. When infarction occurs, the diffusion of water molecules is restricted, and the corresponding area of ​​DWI will show a high signal. ADC also represents the degree of diffusion of water molecules. When the diffusion of water molecules is restricted, the ADC value shows a low signal.

[0028] Specifically, the diffusion of water molecules within the subject's body is simulated and imaged to obtain a diffusion-weighted imaging (DWI) sequence. Simultaneously, based on the grayscale values ​​of each pixel in the DWI sequence, the apparent diffusion coefficient (ADC) sequence corresponding to the DWI sequence is calculated. Then, the brain tissue regions in the DWI sequence are segmented, and based on the brain tissue regions of the subject and the grayscale values ​​of each pixel in the ADC sequence, the low-signal region R corresponding to the ADC sequence is determined. adc Meanwhile, based on the grayscale values ​​of each pixel in the DWI sequence, the high-signal region R corresponding to the DWI sequence is determined. d Ultimately based on the low signal region R adc and high signal region R d Determine the first target evaluation region R in the DWI sequence. dwi .

[0029] In another embodiment of the present invention, a deep learning network can be constructed, using sample DWI and ADC sequences as input, to label the first target evaluation region R. dwi Using the sample DWI sequences as labels, the deep learning network is trained to obtain the first target evaluation region R. dwi Extract the model, and then use the model to determine the first target evaluation region R in the DWI sequence. dwi The first target evaluation region R dwi The training process of the extraction model can employ relevant techniques, which will not be elaborated upon here. The first target evaluation region R... dwi The extracted model can specifically be a network such as U-Net or Deepmedic.

[0030] 102. Based on the FLAIR sequence of the inspected object and the first target evaluation region R dwi Obtain the second target evaluation region R infarction The second objective evaluation region R infarction To compare with the first target evaluation region R dwi High-signal regions in overlapping FLAIR sequences.

[0031] Among them, the FLAIR sequence is a liquid attenuation inversion recovery (IR) sequence for magnetic resonance imaging. The appearance of high signal on the FLAIR sequence indicates that intravascular plasma has leaked into the intracellular and extracellular space (vasogenic edema), the total water content of ischemic tissue has increased, and the tissue has softened and glialized. In this case, it can be basically determined that the tissue has suffered irreversible necrosis and has become the core infarct area.

[0032] In an embodiment of the present invention, the first target evaluation region R in the DWI sequence is determined. dwi Subsequently, in order to determine the stroke tissue window evaluation value of the examined subject, it is first necessary to define the first target evaluation region R. dwi After warping, the image is mapped onto the FLAIR sequence to obtain the mask region R corresponding to the FLAIR sequence. flair Simultaneously, the high-signal region in the FLAIR sequence is identified, and the corresponding mask region R is determined. flair and high signal region R light After that, R flair Region as a mask and R light Find the intersection and define the region corresponding to the intersection as the signal anomaly region of the FLAIR sequence, i.e., the second target evaluation region R. infarction This allows for a more precise determination of the infarct area through the high-signal region corresponding to FLAIR signals, thus expanding the detectable range of the penumbra. It should be noted that in this embodiment of the invention, the FLAIR sequence can be input into a preset anomaly image recognition model for high-signal region extraction. Specifically, the preset anomaly image recognition model can be a network such as U-Net or Deepmedic. Inputting the FLAIR sequence into a network like U-Net or Deepmedic allows the network to output the high-signal region R in the FLAIR sequence. light .

[0033] In another embodiment of the present invention, a statistical method can be used to screen out pixels with low distribution on the histogram by drawing a grayscale histogram on the FLAIR sequence, thereby segmenting the FLAIR normal brain tissue region and the cerebrospinal fluid region, and calculating the mask region R. flair The signal intensity ratio (SIR) of the abnormal region is compared with that of the normal brain tissue region in FLAIR. Based on the SIR value, the abnormal signal regions corresponding to the FLAIR sequence are ultimately selected, which is the second target evaluation region R. infarction .

[0034] To eliminate the influence of uncontrollable factors such as patient movement during scanning, in one example, the FLAIR sequence is a sequence registered with the DWI sequence.

[0035] 103. Based on the first target evaluation region R dwi The corresponding first evaluation volume and second target evaluation region R infarction The corresponding second evaluation volume is used to calculate the stroke tissue window evaluation value of the examined subject.

[0036] In the embodiments of the present invention, when determining the first target evaluation region R dwi Second target evaluation region R infarction Then, calculate the first target evaluation region R. dwi The corresponding DWI volume, i.e., the first evaluation volume, and the calculation of the second target evaluation region R. infarction The corresponding FLA volume, i.e., the second evaluation volume, is used to calculate the stroke tissue window evaluation value of the examined subject using the first and second evaluation volumes. This determines the first target evaluation region R in the DWI sequence. dwi The second target evaluation region R of the FLAIR sequence infarction Then calculate the first target evaluation region R. dwi The corresponding first evaluation volume and second target evaluation region R infarction The corresponding second evaluation volume is used, and finally, based on the first and second evaluation volumes, the stroke tissue window evaluation value of the examined object is determined. This application embodiment utilizes the second target evaluation region in the FLAIR sequence to R... infarction Representing the core infarct area, FLAIR sequences avoid the potential for negative refraction errors that can occur when using DWI sequences to identify the core infarct area. Furthermore, FLAIR sequences are obtained by pulse-weighted MRI images and do not rely on specific instrument technology, making them more widely available than PWI. FLAIR sequences also do not require any contrast agents that pose risks to humans and are not affected by the blood-brain barrier, thus avoiding imaging errors caused by a damaged or incomplete blood-brain barrier, which is common in PWI and improves the accuracy of tissue window evaluation. In addition, FLAIR sequences avoid the problem of different populations having varying tolerances to ADC thresholds.

[0037] Furthermore, to better illustrate the process of determining the tissue window evaluation value for stroke described above, as a refinement and extension of the above embodiments, this invention provides another method for determining the tissue window evaluation value for stroke, such as... Figure 2 As shown, the method includes:

[0038] 201. Extracting normal brain tissue region R from DWI sequence brainmask .

[0039] Specifically, the active contouring method is used to segment DWI sequences, and an energy functional is constructed. Driven by the minimum value of the energy function, the contour curve gradually approaches the edge of the brain tissue region of the examined subject, ultimately segmenting the brain tissue region R of the examined subject. brainmask This method can eliminate the influence of other brain image sequences on the tissue window evaluation value of stroke, and improve the accuracy of tissue window evaluation value determination.

[0040] Prior to step 201, embodiments of the present invention may further include acquiring the DWI sequence and ADC sequence of the object to be examined.

[0041] For DWI sequences, an initial DWI sequence is typically obtained by scanning the brain of the subject. The pixel values ​​of this initial DWI sequence can range from tens of thousands. When saving it as a DICOM (Digital Imaging and Communications in Medicine) file, the entire DWI sequence is compressed. Different slopes and intercepts used during compression will result in different pixel values ​​read when accessing the DICOM file. Therefore, when acquiring a DWI sequence, if other slopes and intercepts are used, the initial pixel values ​​should be calculated first. These initial pixel values ​​should then be multiplied by a preset slope, and the product should be added to a preset intercept to obtain standard pixel values. The sequence composed of these standard pixel values ​​is the DWI sequence. Specifically, the preset slope can be set to 1, and the preset intercept can be set to 0. A typical DWI sequence can be a DWI-B sequence with a diffusion gradient factor of 1000. 1000 .

[0042] For ADC sequences, the diffusion degree of the DWI sequence is controlled by the diffusion gradient factor b. Sequences are obtained when b=0 and b=1000, respectively, based on scanning and taking standard pixel values; these are known as DWI-B0 and DWI-B. 1000 The sequence is used to determine the grayscale value S0 corresponding to each pixel in the DWI-B0 sequence, and to determine the DWI-B sequence. 1000 The grayscale value S1 corresponding to the corresponding pixel in the sequence is obtained, and then the ADC sequence is calculated according to the formula ADC=ln(S0 / S1) / 1000.

[0043] 202. Determine the ADC threshold, and based on the ADC threshold, extract regions from the ADC sequence whose ADC values ​​are less than or equal to the ADC threshold and whose brain tissue regions R are similar to normal brain tissue. brainmask Low-signal regions of ADC sequences with overlapping positions.

[0044] The ADC threshold is a critical value set according to the actual situation to determine the high signal region and the low signal region in the ADC sequence. The region where the pixel is less than or equal to the critical value is the low signal region.

[0045] In this embodiment of the invention, the ADC image is first processed to obtain a grayscale ADC image, and the grayscale value of each pixel in the grayscale ADC image is determined. Based on each grayscale value, the corresponding region that is less than or equal to the ADC threshold is determined. Then, the corresponding region and the brain tissue region R are determined. brainmask The second intersecting region (positional overlap) is identified, and this second intersecting region is determined as the low-signal region R corresponding to the ADC sequence. adc .

[0046] 203. Extract the high signal region of the DWI sequence from the DWI sequence where the pixel value is greater than or equal to the first pixel threshold.

[0047] Wherein, the first pixel threshold is a critical value set according to actual conditions for determining high-signal regions and low-signal regions in the DWI sequence. In this embodiment of the invention, in order to determine the high-signal region R corresponding to the DWI sequence... d First, it is necessary to determine the first pixel threshold. Specific methods include: calculating the mean gray value (Gray) corresponding to each gray value based on the gray values ​​of each pixel in the DWI sequence. mean and grayscale standard deviation Gray std Based on grayscale mean Gray mean and grayscale standard deviation Gray std Calculate the abnormal grayscale evaluation value corresponding to the DWI sequence and use it as the first pixel threshold.

[0048] Specifically, firstly, the DWI image is processed to obtain a grayscale DWI image. Then, the grayscale value of each pixel in the ADC sequence corresponding to the grayscale DWI image is determined, and the mean grayscale value (Gray) of each grayscale value is calculated. mean and grayscale standard deviation Gray std Then based on the grayscale mean Gray mean and grayscale standard deviation Gray std Calculate the abnormal grayscale evaluation value corresponding to the DWI sequence, i.e., the first pixel threshold. The specific calculation formula is as follows:

[0049] Gray etv =max(Gray mean +2*Gray std 1.25*Gray mean )

[0050] Among them, Grayetv This represents the first pixel threshold, which is the average grayscale value (Gray). mean and grayscale standard deviation Gray std Substituting into the above formula, the first pixel threshold corresponding to the DWI sequence can be calculated. Then, among the pixels in the DWI sequence, the target pixels with gray values ​​greater than or equal to the first pixel threshold are identified, and the region corresponding to the target pixel is determined as the high signal region R of the DWI sequence. d .

[0051] 204. Based on the low-signal region of the ADC sequence, extract the first target evaluation region R from the high-signal region of the DWI sequence that overlaps with the low-signal region of the ADC sequence. dwi .

[0052] In an embodiment of the present invention, the low-signal region R of the ADC sequence is determined. adc and the high signal region R of the DWI sequence d Then, it is necessary to analyze the high-signal region R of the DWI sequence. d The included connected regions are used to determine the low-signal region R of the ADC sequence. adc Intersecting connected regions are considered as anomalous regions of the DWI signal, i.e., the first target evaluation region R in the DWI sequence. dwi For example, the high-signal region R in a DWI sequence d It contains a connected region R d1 Connected region R d2 Connected region R d3 Among the three connected components mentioned above, only the connected region R is determined. d2 The low signal region R of the ADC sequence adc If there is an intersection, then the connected region R is considered as one of the connected regions. d2 The first target evaluation region R in the DWI sequence was identified. dwi .

[0053] Furthermore, embodiments of the present invention can also pre-train and construct a first target evaluation region Rdwi extraction model, and input the DWI sequence and ADC sequence into the pre-trained first target evaluation region Rdwi. dwi Extract the model and obtain the first target evaluation region R. dwi Extract the first target evaluation region R from the model output. dwi Specifically, using sample DWI and ADC sequences as input, the first target evaluation region R is labeled. dwi Using the sample DWI sequences as labels, the deep learning network is trained to obtain the first target evaluation region R. dwi Extract the model, and then use the model to determine the first target evaluation region R in the DWI sequence. dwi The first target evaluation region Rdwi The training process of the extraction model can employ relevant techniques, which will not be elaborated upon here. The first target evaluation region R... dwi The extraction model can specifically be networks such as U-Net and Deepmedic.

[0054] 205. Extract the high signal region of the FLAIR sequence from the FLAIR sequence where the pixel value is greater than or equal to the second pixel threshold.

[0055] The second pixel threshold is a critical value used to distinguish between high-signal and low-signal regions in the FLAIR sequence. In this embodiment, target pixels with values ​​greater than or equal to the second pixel threshold can be identified from the pixel values ​​of each pixel in the FLAIR sequence, and the region corresponding to the target pixel is defined as the high-signal region of the FLAIR sequence. Alternatively, the FLAIR sequence can be input into a preset anomaly image recognition model to extract the high-signal region. This preset anomaly image recognition model is a machine learning model used for anomaly image recognition. It can be a supervised neural network model, and each layer of the preset image recognition model processes the FLAIR sequence accordingly to identify the high-signal region R in the FLAIR sequence. light .

[0056] 206. The first target evaluation area R dwi Mapping to the FLAIR sequence, we obtain the evaluation region R corresponding to the first target. dwi The mask region R of the FLAIR sequence corresponding to the position flair .

[0057] Specifically, a pre-defined affine registration algorithm is used to determine the deformation relation matrix for converting DWI sequences to FLAIR sequences; based on the deformation relation matrix, the first target evaluation region R in the DWI sequence is... dwi Mapping to the FLAIR sequence yields the mask region R corresponding to the FLAIR sequence. flair .

[0058] Specifically, the FLAIR sequence is used as a fixed image, and the first target evaluation region R in the DWI sequence is... dwi As a floating image (movingImage), a pre-defined affine registration algorithm is used to determine the first target evaluation region R. dwi The deformation relation matrix is ​​mapped to the FLAIR sequence. Finally, based on the deformation relation matrix, the first target evaluation region R in the DWI sequence is... dwi Mapping to the FLAIR sequence yields the mask region R corresponding to the FLAIR sequence. flair .

[0059] 207. Mask the FLAIR sequence region R flair As a mask, the second target evaluation region R is extracted from the high-signal region of the FLAIR sequence. infarction .

[0060] Among them, the high-signal region R of the mask region Rflair and FLAIR sequence was determined. light The first intersecting region in the sequence is identified, and this first intersecting region is defined as the second target evaluation region R corresponding to the FLAIR sequence. infarction And extract the second target evaluation region R. infarction .

[0061] 208. Determine the first target evaluation region R dwi The corresponding first evaluation volume and second target evaluation region R infarction The corresponding second evaluation volume is obtained by subtracting the first evaluation volume from the second evaluation volume to obtain the first stroke tissue window evaluation value of the examined object; the first evaluation volume is then divided by the second evaluation volume to obtain the second stroke tissue window evaluation value of the examined object.

[0062] Specifically, the first target evaluation region R is determined based on pixel size and pixel count. dwi The corresponding first evaluation volume and second target evaluation region R infarction The corresponding second evaluation volume.

[0063] Determine the first target evaluation region R dwi The corresponding first evaluation volume and second target evaluation region R infarction After determining the corresponding second evaluation volume, the stroke tissue window evaluation value for the examined object can be determined according to the following formula:

[0064] mismatch Volume=DWI Volume-FLA Volume

[0065]

[0066] Wherein, DWI Volume represents the first evaluation volume, FLA Volume represents the second evaluation volume, mismatchVolume represents the first stroke tissue window evaluation value, and mismatch ratio represents the second stroke tissue window evaluation value. Thus, the first stroke tissue window evaluation value and the first stroke tissue window evaluation value corresponding to the examined object can be calculated according to the above formula.

[0067] The first and second stroke tissue window evaluation values ​​can be used to evaluate the stroke tissue window independently, or they can be combined to evaluate the stroke tissue window.

[0068] Furthermore, after determining the first stroke tissue window evaluation value and the second stroke tissue window evaluation value corresponding to the examination subject, the method further includes: if the first stroke tissue window evaluation value is less than a first preset evaluation threshold, and / or the second stroke tissue window evaluation value is less than a second preset evaluation threshold, then it is determined that the examination subject's brain tissue is severely damaged and there is an unsalvageable ischemic penumbra area; if the first stroke tissue window evaluation value is greater than or equal to the first preset evaluation threshold, and / or the second stroke tissue window evaluation value is greater than or equal to the second preset evaluation threshold, then it is determined that the examination subject's brain tissue can restore the physiological function of the ischemic penumbra through blood reperfusion.

[0069] Wherein, the first evaluation volume is greater than or equal to the second evaluation volume, the first preset evaluation threshold can be set to 0.04, the second preset evaluation threshold can be set to 1.04, and the first target evaluation region R is... dwi Defined as a low-perfusion area, the second target evaluation area R infarction Defined as the core infarct region, the first target evaluation region R dwi Except for the second target evaluation region R infarction The area outside of this zone is defined as the ischemic penumbra.

[0070] Specifically, the abnormal DWI signal area, i.e., the first target evaluation area R, is... dwi The hypoperfusion area in the brain tissue of the subject was identified as the region with abnormal FLAIR signal, which was then designated as the second target evaluation region R. infarction The core infarct area in the brain tissue of the subject is identified. In this embodiment of the invention, if the evaluation value of the first stroke tissue window is less than the first preset evaluation threshold, or the evaluation value of the second stroke tissue window is less than the second preset evaluation threshold, then the abnormal DWI signal area closely matches the abnormal FLAIR signal area. This indicates severe brain tissue damage in the subject, with almost no salvageable ischemic penumbra. The second target evaluation area R... infarction This refers to the core infarct area of ​​the examinee's brain. If the first stroke tissue window evaluation value is greater than or equal to the first preset evaluation threshold, or the second stroke tissue window evaluation value is greater than or equal to the second preset evaluation threshold, then it is determined that the examinee can restore the physiological function of the ischemic penumbra through blood reperfusion, possessing good therapeutic value. The first target evaluation area R... dwi Remove the second target evaluation region R infarction Perfusion therapy is performed on areas other than the ischemic penumbra.

[0071] Furthermore, to provide doctors with reference information for treating the examined patient, the method also includes: evaluating the first stroke tissue window value, the second stroke tissue window value, and the first target evaluation region R. dwi Corresponding volume and location information, second target evaluation area R infarction The system displays at least one of the following: corresponding volume and location information, volume and location information of the ischemic penumbra region, and treatment information.

[0072] According to another method for determining the evaluation value of the tissue window in stroke provided by the present invention, compared with the current method of evaluating the tissue window by determining the matching degree between DWI and PWI, the present invention obtains the first target evaluation region R in the DWI sequence based on the DWI sequence and ADC sequence of the examined object. dwi The first target evaluation region R dwi The high-signal region of the DWI sequence overlaps with the low-signal region of the ADC sequence; and is based on the FLAIR sequence of the object under examination and the first target evaluation region R. dwi Obtain the second target evaluation region R infarction The second objective evaluation region R infarction To compare with the first target evaluation region R dwi High-signal regions in overlapping FLAIR sequences; ultimately, based on the first target evaluation region R... dwi The corresponding first evaluation volume and second target evaluation region R infarction The corresponding second evaluation volume is used to calculate the stroke tissue window evaluation value of the examined subject, thereby determining the first target evaluation region R in the DWI sequence. dwi The second target evaluation region R of the FLAIR sequence infarction Then calculate the first target evaluation region R. dwi The corresponding first evaluation volume and second target evaluation region R infarction The corresponding second evaluation volume, and finally based on the first and second evaluation volumes, determines the evaluation value of the stroke tissue window of the examined object. This avoids the situation in the prior art where PWI imaging errors are caused by damage or incompleteness of the blood-brain barrier, which leads to the inability to accurately obtain the degree of mismatch between DWI and PWI, thereby improving the evaluation accuracy of the tissue window.

[0073] Furthermore, as Figure 1 In specific implementation, embodiments of the present invention provide a device for determining the evaluation value of a tissue window in stroke, such as... Figure 3 As shown, the device includes: a first acquisition unit 31, a second acquisition unit 32, and a calculation unit 33.

[0074] The first acquisition unit 31 can be used to acquire the first target evaluation region R in the DWI sequence based on the DWI sequence and ADC sequence of the object being inspected. dwi The first target evaluation region R dwi The high-signal region of the DWI sequence that overlaps with the low-signal region of the ADC sequence.

[0075] The second acquisition unit 32 can be used to acquire data based on the FLAIR sequence of the object being inspected and the first target evaluation region R. dwi Obtain the second target evaluation region R infarction The second objective evaluation region R infarction To compare with the first target evaluation region R dwi High-signal regions in overlapping FLAIR sequences.

[0076] Calculation unit 33 can be used to evaluate the first target region R. dwi The corresponding first evaluation volume and second target evaluation region R infarction The corresponding second evaluation volume is used to calculate the stroke tissue window evaluation value of the examined subject.

[0077] In specific application scenarios, in order to obtain the first target evaluation region R in the DWI sequence dwi The first acquisition unit 31 includes a first extraction module 311 and a first acquisition module 312.

[0078] The first extraction module 311 can be used to extract normal brain tissue regions R from DWI sequences. brainmask .

[0079] The first extraction module 311 can be specifically used to determine the ADC threshold and, based on the ADC threshold, extract from the ADC sequence regions whose ADC values ​​are less than or equal to the ADC threshold and are consistent with normal brain tissue regions R. brainmask Low-signal regions of ADC sequences with overlapping positions.

[0080] The first extraction module 311 can also be used to extract high signal regions of the DWI sequence from the DWI sequence whose pixel values ​​are greater than or equal to the first pixel threshold.

[0081] The first acquisition module 312 can be used to extract a first target evaluation region R that overlaps with the low-signal region of the ADC sequence from the high-signal region of the DWI sequence based on the low-signal region of the ADC sequence. dwi Alternatively, input the DWI sequence and ADC sequence into the pre-trained first target evaluation region R. dwi Extract the model and obtain the first target evaluation region R. dwi Extract the first target evaluation region R from the model output. dwi .

[0082] In specific application scenarios, in order to determine the first pixel threshold, the first extraction module 311 can be used to calculate the gray mean value Gray corresponding to each gray value based on the gray value corresponding to each pixel in the DWI sequence. mean and grayscale standard deviation Gray std Based on grayscale mean Gray mean and grayscale standard deviation Gray std Calculate the abnormal grayscale evaluation value corresponding to the DWI sequence and use it as the first pixel threshold.

[0083] In specific application scenarios, in order to determine the second target evaluation region R corresponding to the FLAIR sequence infarction The second acquisition unit 32 includes a second extraction module 321 and a second acquisition module 322.

[0084] The second extraction module 321 can be used to extract high signal regions of the FLAIR sequence from the FLAIR sequence whose pixel values ​​are greater than or equal to the second pixel threshold.

[0085] The second acquisition module 322 can be used to obtain the first target evaluation region R dwi Mapping to the FLAIR sequence, we obtain the evaluation region R corresponding to the first target. dwi The mask region R of the FLAIR sequence corresponding to the position flair .

[0086] The second extraction module 321 can specifically be used to extract the mask region R of the FLAIR sequence. flair As a mask, the second target evaluation region R is extracted from the high-signal region of the FLAIR sequence. infarction .

[0087] In specific application scenarios, in order to determine the mask region R of the FLAIR sequence flair The second acquisition module 322 includes a determination submodule and a mapping submodule.

[0088] The determination submodule can be used to determine the deformation relationship matrix for converting DWI sequences to FLAIR sequences using a preset affine registration algorithm.

[0089] The mapping submodule can be used to map the first target evaluation region R in the DWI sequence based on the deformation relation matrix. dwi Mapping this onto the FLAIR sequence yields the mask region R of the FLAIR sequence. flair .

[0090] In specific application scenarios, in order to calculate the evaluation value of the stroke tissue window of the examined object, the calculation unit 33 includes a subtraction module 331 and a division module 332.

[0091] The subtraction module 331 can be used to subtract the first evaluation volume from the second evaluation volume to obtain the first stroke tissue window evaluation value corresponding to the examined object.

[0092] The division module 332 can be used to divide the first evaluation volume by the second evaluation volume to obtain the second stroke tissue window evaluation value corresponding to the examined object.

[0093] In specific application scenarios, in order to evaluate the stroke tissue window based on the stroke tissue window evaluation value, the device also includes: a determination unit 34.

[0094] The determination unit 34 can be used to determine that the brain tissue of the examined object is severely damaged and there is an unsalvageable ischemic penumbra area if the evaluation value of the first stroke tissue window is less than the first preset evaluation threshold and / or the evaluation value of the second stroke tissue window is less than the second preset evaluation threshold.

[0095] The determination unit 34 can also be used to determine that if the evaluation value of the first stroke tissue window is greater than or equal to the first preset evaluation threshold, and / or the evaluation value of the second stroke tissue window is greater than or equal to the second preset evaluation threshold, then the brain tissue of the examined object can restore the physiological function of the ischemic penumbra through blood reperfusion.

[0096] In specific application scenarios, in order to display the evaluation information of the stroke tissue window to doctors, the device also includes a display unit 35.

[0097] The display unit 35 can be used to display the evaluation values ​​of the first stroke tissue window, the second stroke tissue window, and the first target evaluation area R. dwi Corresponding volume and location information, second target evaluation area R infarction The system displays at least one of the following: corresponding volume and location information, volume and location information of the ischemic penumbra region, and treatment information.

[0098] It should be noted that other corresponding descriptions of the functional modules involved in the device for determining the evaluation value of a stroke tissue window provided in this embodiment of the invention can be found in the following references. Figure 1 The corresponding description of the method shown will not be repeated here.

[0099] Based on the above, Figure 1 Accordingly, this embodiment of the invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, performs the following steps: based on the DWI sequence and ADC sequence of the object being inspected, obtain a first target evaluation region R in the DWI sequence. dwi The first target evaluation region R dwiThe high-signal region of the DWI sequence that overlaps with the low-signal region of the ADC sequence; based on the FLAIR sequence of the object under inspection and the first target evaluation region R. dwi Obtain the second target evaluation region R infarction The second objective evaluation region R infarction To compare with the first target evaluation region R dwi High-signal regions of overlapping FLAIR sequences; based on the first target evaluation region R dwi The corresponding first evaluation volume and second target evaluation region R infarction The corresponding second evaluation volume is used to calculate the stroke tissue window evaluation value of the examined subject.

[0100] Based on the above, Figure 1 The method shown and as Figure 3 The embodiment of the device shown in the invention also provides a physical structure diagram of a computer device, such as... Figure 4 As shown, the computer device includes: a processor 41, a memory 42, and a computer program stored in the memory 42 and executable on the processor. Both the memory 42 and the processor 41 are mounted on a bus 43. When the processor 41 executes the program, it performs the following steps: based on the DWI sequence and ADC sequence of the object being inspected, it obtains a first target evaluation region R in the DWI sequence. dwi The first target evaluation region R dwi The high-signal region of the DWI sequence that overlaps with the low-signal region of the ADC sequence; based on the FLAIR sequence of the object under inspection and the first target evaluation region R. dwi Obtain the second target evaluation region R infarction The second objective evaluation region R infarction To compare with the first target evaluation region R dwi High-signal regions of overlapping FLAIR sequences; based on the first target evaluation region R dwi The corresponding first evaluation volume and second target evaluation region R infarction The corresponding second evaluation volume is used to calculate the stroke tissue window evaluation value of the examined subject.

[0101] Through the technical solution of this invention, the first target evaluation region R in the DWI sequence is obtained based on the DWI sequence and ADC sequence of the object being inspected. dwi The first target evaluation region R dwi The high-signal region of the DWI sequence overlaps with the low-signal region of the ADC sequence; and is based on the FLAIR sequence of the object under examination and the first target evaluation region R. dwi Obtain the second target evaluation region R infarction The second objective evaluation region R infarction To compare with the first target evaluation region Rdwi High-signal regions in overlapping FLAIR sequences; ultimately, based on the first target evaluation region R... dwi The corresponding first evaluation volume and second target evaluation region R infarction The corresponding second evaluation volume is used to calculate the stroke tissue window evaluation value of the examined subject, thereby determining the first target evaluation region R in the DWI sequence. dwi The second target evaluation region R of the FLAIR sequence infarction Next, the evaluation region R of the first target is calculated. dwi The corresponding first evaluation volume and second target evaluation region R infarction The corresponding second evaluation volume, and finally based on the first and second evaluation volumes, determines the evaluation value of the stroke tissue window of the examined object. This avoids the situation in the prior art where PWI imaging errors are caused by damage or incompleteness of the blood-brain barrier. At the same time, it also avoids the error of using DWI to determine the core infarct area, which leads to the inability to accurately obtain the degree of mismatch between DWI and PWI, thereby improving the evaluation accuracy of the tissue window.

[0102] It is obvious to those skilled in the art that the modules or steps of the present invention described above can be implemented using general-purpose computing devices. They can be centralized on a single computing device or distributed across a network of multiple computing devices. Optionally, they can be implemented using computer-executable program code, thereby storing them in a storage device for execution by a computing device. In some cases, the steps shown or described can be performed in a different order than those presented herein, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. Thus, the present invention is not limited to any particular combination of hardware and software.

[0103] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for determining the tissue window evaluation value in stroke, characterized in that, include: Based on the DWI sequence and the ADC sequence of the examination object, a first target evaluation region R is acquired in the DWI sequence dwi , the first target evaluation region R dwi is a high signal region of the DWI sequence overlapping with a low signal region position of the ADC sequence; Based on the FLAIR sequence of the inspected object and the first target evaluation region R dwi Obtain the second target evaluation region R infarction The second target evaluation region R infarction To be consistent with the first target evaluation region R dwi High-signal regions of the FLAIR sequences that overlap in position; based on the first target evaluation region R dwi a corresponding first evaluation volume and the second target evaluation region R infarction a corresponding second evaluation volume, to calculate a stroke tissue window evaluation value of the examination object; The first target evaluation region R dwi The first evaluation volume and the second target evaluation region R infarction The first evaluation volume and the second target evaluation region R The first evaluation volume and the second target evaluation region R 2. The method according to claim 1, characterized in that, The first target evaluation region R in the DWI sequence is obtained based on the DWI sequence and ADC sequence of the object being inspected. dwi ,include: Normal brain tissue region R was extracted from the DWI sequence. brainmask ; Determine the ADC threshold, and based on the ADC threshold, extract from the ADC sequence regions whose ADC values ​​are less than or equal to the ADC threshold and are related to the normal brain tissue region R. brainmask Low-signal regions of the ADC sequences that overlap in position; Extract the high-signal regions of the DWI sequence whose pixel values ​​are greater than or equal to a first pixel threshold. Based on the low-signal region of the ADC sequence, a first target evaluation region R that overlaps with the low-signal region of the ADC sequence is extracted from the high-signal region of the DWI sequence. dwi ;or, The DWI sequence and the ADC sequence are input into the pre-trained first target evaluation region R. dwi Extract the model and obtain the first target evaluation region R. dwi Extract the first target evaluation region R from the model output. dwi .

3. The method according to claim 2, characterized in that, Determining the first pixel threshold includes: Based on the gray values ​​corresponding to each pixel in the DWI sequence, calculate the mean gray value Gray corresponding to each gray value. mean and grayscale standard deviation Gray std ; Based on the grayscale mean Gray mean and the grayscale standard deviation Gray std Calculate the abnormal grayscale evaluation value corresponding to the DWI sequence and use it as the first pixel threshold.

4. The method according to claim 1, characterized in that, The FLAIR sequence of the inspected object and the first target evaluation region R dwi Obtain the second target evaluation region R infarction ,include: Extract the high-signal regions of the FLAIR sequence whose pixel values ​​are greater than or equal to a second pixel threshold from the FLAIR sequence; The first target evaluation region R dwi Mapped to the FLAIR sequence, the first target evaluation region R is obtained. dwi Mask regions R of the overlapping FLAIR sequences flair ; The mask region R of the FLAIR sequence flair As a mask, the second target evaluation region R is extracted from the high-signal region of the FLAIR sequence. infarction .

5. The method according to claim 4, characterized in that, The first target evaluation region R dwi Mapped to the FLAIR sequence, the first target evaluation region R is obtained. dwi Mask regions R of the overlapping FLAIR sequences flair ,include: Using a preset affine registration algorithm, the deformation relationship matrix for converting the DWI sequence to the FLAIR sequence is determined; Based on the deformation relationship matrix, the first target evaluation region R in the DWI sequence dwi Mapped onto the FLAIR sequence, the mask region R of the FLAIR sequence is obtained. flair .

6. The method according to claim 1, characterized in that, Based on the first target evaluation region R dwi The corresponding first evaluation volume and the second target evaluation region R infarction After calculating the stroke tissue window evaluation value of the examined subject for the corresponding second evaluation volume, the method further includes: The evaluation values ​​of the first stroke tissue window, the second stroke tissue window, and the first target evaluation region R are used. dwi The corresponding volume information and location information, the second target evaluation area R infarction The system displays at least one of the following: corresponding volume and location information, volume and location information of the ischemic penumbra region, and treatment information.

7. A device for determining the evaluation value of tissue window in stroke, characterized in that, include: The first acquisition unit is used to acquire the first target evaluation region R in the DWI sequence based on the DWI sequence and ADC sequence of the object being inspected. dwi The first target evaluation region R dwi The high-signal region of the DWI sequence that overlaps with the low-signal region of the ADC sequence. The second acquisition unit is used to acquire data based on the FLAIR sequence of the object being inspected and the first target evaluation region R. dwi Obtain the second target evaluation region R infarction The second target evaluation region R infarction To be consistent with the first target evaluation region R dwi High-signal regions of the FLAIR sequences that overlap in position; The calculation unit is used to calculate based on the first target evaluation region R. dwi The corresponding first evaluation volume and the second target evaluation region R infarction The corresponding second evaluation volume is used to calculate the stroke tissue window evaluation value of the examined object, wherein the evaluation value is based on the first target evaluation region R. dwi The corresponding first evaluation volume and the second target evaluation region R infarction The corresponding second evaluation volume is used to calculate the stroke tissue window evaluation value of the examined object, including: subtracting the first evaluation volume from the second evaluation volume to obtain the first stroke tissue window evaluation value corresponding to the examined object; and dividing the first evaluation volume from the second evaluation volume to obtain the second stroke tissue window evaluation value corresponding to the examined object.

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 6.

Images