KR101611489B1 - Method for estimating onset time of infarct region based on brain image - Google Patents

Abstract

The present disclosure relates to a method and apparatus for obtaining brain images; Extracting a quantitative value set that varies according to an onset time of an infarct region from an infarct region included in brain images; And a quantitative value set corresponding to a time point of occurrence of the infarct region using a corresponding relation prepared to correspond a quantitative value set to the occurrence time of the infarct region. The present invention relates to an estimation method of an occurrence time point of an infarction area based on an image.

Description

[0001] METHOD FOR ESTIMATING ONSET TIME OF INFARCT REGION BASED ON BRAIN IMAGE [0002]

The present disclosure relates to a method of estimating the time of occurrence of an infarct area based on brain images as a whole, and more particularly, to a method of estimating an occurrence time of an infarct area using a quantitative value set extracted from brain images, And a method for estimating the time of occurrence of an infarct area based on the estimated brain image.

Herein, the background art relating to the present disclosure is provided, and these are not necessarily meant to be known arts.

Methods for evaluating the status of lesions using medical images are widely used. For example, in patients with acute stroke, such as cerebral infarct or cerebral hemorrhage, the infarct area is assessed by imaging the brain with MRI.

Cho AH, Sohn SI, Han MK, et al. Safety and efficacy of MRI-based thrombolysis in unclear-onset stroke. A preliminary report. Cerebrovasc Dis 2008; 25: 572-579, discloses a method for estimating the time of occurrence of an infarct region with FLAIR (fluid-attenuated inversion recovery) images of an MRI for a patient whose time of onset of cerebral infarction is unclear. In this paper, MRI-based conditions used as the basis for the time of occurrence include positive perfusion-diffusion mismatch (PWI-DWI mismatch) conditions and absence of well-developed fluid-attenuated inversion recovery changes of acute diffusion lesions (FLAIR CHANGE) do.

However, in this paper, conditions such as PWI-DWI mismatch and FLAIR CHANGE are indices by the physician's visual inspection and do not provide a quantitative and objective criterion.

On the other hand, the paper by Shlee S. Song, Lawrence L. Latour, Carsten H. Ritter, et al. A Pragmatic Approach Using Magnetic Resonance Imaging to Treat Ischemic Strokes of Unknown Onset Time in a Thrombolytic Trial. the American Heart Association's online June 12, 2012. The reader-measured signal intensity ratio (SIR), which measures the intensity (eg intensity) of FLAIR images corresponding to diffusion-positive regions, .

However, in the SIR method, since the SIR value is determined by a method in which a plurality of image readers score, there is still a limit to be an objective index. In addition, since it is a single parameter method in which the time of occurrence of cerebral infarction is judged only by the brightness of the FLAIR image, reliability is limited. This is because the extent of the infarct area in the FLAIR image from the time point to the bright area depends on the location of the infarct area, the volume, the patient's age, sex, and so on. Also, in FLAIR images that change with time, although the brightness has a spectrum distribution, it is objectivity and reliability to determine the occurrence time point with a specific SIR value as a boundary.

This will be described later in the Specification for Implementation of the Invention.

SUMMARY OF THE INVENTION Herein, a general summary of the present disclosure is provided, which should not be construed as limiting the scope of the present disclosure. of its features).

According to one aspect of the present disclosure, brain images are acquired; Extracting a quantitative value set that varies according to an onset time of an infarct region from an infarct region included in brain images; And a quantitative value set corresponding to a time point of occurrence of the infarct region using a corresponding relation prepared to correspond a quantitative value set to the occurrence time of the infarct region. A method of estimating the occurrence time of an infarction area based on an image is provided.

This will be described later in the Specification for Implementation of the Invention.

FIG. 1 is a view for explaining an example of a method of estimating the time of occurrence of an infarction area based on a brain image according to the present disclosure;
FIG. 2 is a view for explaining an example of a method of distinguishing a case where a time point of occurrence of an infarction area is clear and an unclear case;
3 is a diagram showing brain images generated by MRI,
4 is a view for explaining an example of a change in brightness of an infarction area in a DWI image according to a time point of occurrence of an infarction area,
5 is a view for explaining a pattern in which quantitative values extracted from brain images vary with time;
6 is a view for explaining an example of a correspondence relationship generation method,
7 is a view for explaining an example of the relationship between the quantitative value set and the occurrence time of the infarction area,
8 is a view for explaining an example of a corresponding relationship generated by a regression analysis;

The present disclosure will now be described in detail with reference to the accompanying drawings.

1 is a view for explaining an example of a method of estimating the time of occurrence of an infarction area based on a brain image according to the present disclosure.

In the method of estimating the time of occurrence of the infarct area based on the brain image, brain images are obtained (10). A set of quantitative values that varies with the onset time of the infarct region from the infarct region included in the acquired brain images (DWI, ADC, PWI, FLAIR, T1, T2, (30). Thereafter, the quantitative value set corresponds to the occurrence time of the infarct region using the corresponding relation prepared to correspond the quantitative value set to the occurrence time of the infarct region (40).

After the brain images are acquired, the infarct region can be segmented using at least one of the obtained brain images before the quantitative value set is extracted (13). The peripheral region of the infarct region can be divided using at least one of the obtained brain images (15). In addition, brain images including at least one brain image in which the infarct region or the surrounding region is divided may be registered (20).

The quantitative value set may include at least one information selected from the group consisting of size information of the infarct region or the surrounding region, position information, and brightness information from the matched brain images.

2 is a view for explaining an example of a method of distinguishing a case where the occurrence time of the infarction area is clear and a case where it is not clear.

The method of estimating the time of occurrence of the infarct region based on the brain image according to the present example discloses a method of estimating the time of occurrence of the infarct region based on the brain image applied to both the known case and the unknown case of the occurrence time of the infarct region . Estimation of the time of occurrence is made using a previously established correspondence. The correspondence relation can be prepared by classifying the quantitative value set by the occurrence time of the infarct region using a classifier.

For example, the correspondence relation includes at least one of a multiple regression method, a support vector regression method, and a curve fitting method, which is a set of quantitative values accumulated from brain images in which the occurrence time of the infarct region is known To correspond to the occurrence time point of the infarction area.

2 shows an example of a method for determining the occurrence time of an infarction area (clear onset time) and determining an unclear onset time (Unclear Onset Time).

For example, if an acute stroke occurs in the brain, check that the Last-Known normal time matches the first-found abnormal time (101) ). If they match, they are classified as clear onset (103). If the Last-Known normal time does not match the First-found abnormal time, it is checked whether the emergency room ER arrived within 4.5 hours from the Last-Known normal time (105). Within 4.5 hours, it is the case of unclear onset, but may be considered for thrombolysis (107). After 4.5 hours, it corresponds to an unclean onset that is not considered for thrombolysis (109).

The method of estimating the time of occurrence of the infarction area based on the brain image according to the present disclosure does not specify a method of treatment or diagnosis, such as whether thrombolysis is performed or not, and discloses a method of estimating the time of occurrence of the infarction area only.

According to the method for estimating the time of occurrence of the infarct area based on the brain image according to the present disclosure, the time of occurrence of the infarct area is estimated based on the quantitative value set extracted from the brain images. Thus, the method of estimating the time of onset of an infarct area based on brain imaging is a method that does not rely on uncertain factors such as elapsed time from Last-Known normal time.

For example, a mapping relationship is provided by a classifier that is trained or learned to map the quantitative value set to the time of the infarct region.

Hereinafter, each step of the method of estimating the occurrence time of the infarct area based on the brain image will be described in detail.

3 is a diagram showing brain images generated by MRI.

First, brain images are acquired (10 in FIG. 1).

For example, brain images are generated by MRI. Brain images include DWI, ADC, PWI, FLAIR, T1, and T2 images. In this example, the brain image is not limited to MRI but may also include a CT image or other medical images.

The brain images are generated using a combined pulse sequence according to the needs of each image. Thus, the brain images may have different modalities. However, ADC is an image generated by calculation based on DWI.

In order to acquire images by MRI, a large number of RF pulses must be applied. TR (repetition time) is a time interval for applying RF pulses of the same size, and is a repetition time. TE (time to echo) is the delay time or echo time. Since the signal can not be measured immediately after the RF pulse, it waits for a short time and measures the signal after a certain time. This short time is called TE.

The TR and TE can be adjusted by the inspector. By appropriately adjusting TR and TE, T1 or T2 images that can be applied to clinical applications can be obtained.

Hereinafter, brain images in which a quantitative value set can be extracted are briefly described.

DWI (diffusion weighted imaging) is an image that maps the diffusion motion of molecules, especially water molecules, in living tissue (see FIG. 3). The diffusion of water molecules in tissues is not free. DWI reflects the collision of water molecules with fibrous tissue or membranes. Thus, the diffusion pattern of water molecules indicates the normal or abnormal state of the tissue. In particular, DWI can well represent the normal or abnormal state of the fiber structure or gray matter of the white matter of the brain.

The apparent diffusion coefficient (ADC) is a function of temperature as a diffusion coefficient. Because the cell walls are present and the temperature is uneven in the body, the ADC can be calculated using DWI. DWI and ADC are reversed. In the infarcted area, the expansion of the cells reduces the diffusion of water outside the cells. Diffused areas show a decrease in signal intensity when the DWI is taken at B1000 and bright areas at the DWI image. On the other hand, regions with reduced diffusion are darker than normal in ADCs. Water, such as cerebrospinal fluid (CSF), is a free diffusion region with a bright ADC and dark DWI.

PWI (perfusion weighted imaging) is a perfusion image showing blood flow (see FIG. 3). For example, when Gadolinium is injected as a contrast agent and an MRI image is taken, parameters such as blood flow rate, blood flow velocity, MTT (mean transit time), and TTP (time to peak) can be obtained.

The FLAIR image is an image that nulls the signal from the fluid (see FIG. 3). For example, FLAIR images are used to inhibit the effects of cerebrospinal fluid on images when acquiring brain images with MRI. The FLAIR image shows the anatomy of the brain well. Depending on the organization, signals can be suppressed from specific tissues by well choosing the inversion time.

T1 and T2 images are images emphasizing T1 or T2 effects of specific tissues by regulating TR and TE. In the course of MRI, blocking the applied radio frequency (RF) pulses causes the protons of the tissue to re-orient themselves in the direction of the external magnetic field (B 0) (Z-axis direction) while releasing the absorbed energy to surrounding tissues. T1 is the time constant of the time during which the spindles of the proton are rearranged along the Z axis, which is the ordinate axis, i.e., the curve in which the magnetization in the Z axis is restored. T1 is the time constant of magnetization recovery and is called the longitudinal relaxation time or the spin-lattice relaxation time. On the other hand, when the RF pulse is interrupted, the XY component of the magnetization collapses. T2 is a time constant of the XY component decay curve of the magnetization, referred to as transverse relaxation time or spin-spin relaxation time. T1 and T2 are tissue specific values and have different values for water, solid, fat, and protein. Here, increasing the TR reduces the T1 effect. Conversely, if the TR is shortened, the T1 effect (contrast degree) is increased, that is, the T1 weighted image is obtained. If the TE is shortened, the T2 effect is reduced. If the TE is made long, the T2 effect is increased, that is, the T2 weighted image is obtained.

MRA images are images taken by MRI using contrast media.

The above-described brain images can be analyzed qualitatively.

For example, it is possible to judge the mismatch of DWI-PWI at the time and evaluate the degree of the peripheral region affected by the infarct region and the infarct region. Alternatively, a white area may appear in the DWI image, and a state in which the white area is not yet displayed in the FLAIR image (FLAIR negative) may be determined as an initial state of the infarcted cerebral infarction.

However, such qualitative judgment is less accurate and reliable. Unlike this qualitative determination method, the method of estimating the occurrence time of the infarct region according to the present disclosure is characterized by extracting a quantitative value set from brain images and estimating the time of occurrence based on a quantitative value set.

Referring to FIG. 1 again, after acquiring brain images, an infarct region may be segmented in at least one brain image of acquired brain images (13). In addition, the penumbra of the infarct region can be divided in at least one of the brain images (15). The peripheral region is an area surrounding the infarction area or surrounding the infarction area, which is affected by the infarction and has a problem in the supply of blood flow.

For example, in the early stages, infarcts are well visible in DWI images. Therefore, infarct regions can be segmented using DWI or ADC images (13). In addition, since the surrounding region is well recognized by the PWI image, the surrounding region can be divided using the PWI image (15).

Next, brain images including at least one brain image in which the infarct region or the peripheral region is divided are registered (20).

For example, anatomical images such as FLAIR, T1, T2 images, DWI images with split infarct regions, ADC images, PWI images, and remaining brain images are matched. Also, brain images by CT may be matched together. In this example, it is explained that a plurality of brain images are matched, but it is not necessarily required that all of the brain images described above are matched. Two or more brain images, preferably having different modalities, are matched according to the purpose of viewing the brain image.

In this example, the remaining brain images are matched based on the DWI image in which the infarct region is divided. For matching, a rigid registration method can be used. The brain has little motion other than the heart or lung, but in some cases a non-rigid registration method may be used.

As the basis of the matching, an atlas having template information of a brain image can be used. The Atlas is a model of brain imaging from the brain images of many individuals.

Subsequently, a quantitative value set is extracted from the infarct region of the matched brain images (30). For example, the quantitative value set refers to the information of the infarct region correlated with the time of the infarct region. Preferably, a quantitative value set that varies depending on the time of occurrence of the infarct region is extracted from the matched brain images. The quantitative value set may include at least one selected from the group consisting of brightness information, size information, and position information of the infarct region.

For example, the volume of the infarct volume (V _I ), the infarct location (V _IL ), the volume of the penumbra volume (V _P ) surrounding the infarct area, The average brightness (DI _DWI , DI _ADC , DI _FLAIR , DI _T1 , and DI _T2 ) of the infarct region in each brain image and the average brightness of the penumbra position (V _PL ) For each brain image, two or more quantitative values of the average brightness (DP _DWI , DP _ADC , DP _FLAIR , DP _T1 , and DP _T2 ) in the surrounding region can be obtained. The average brightness of the infarct area (DI _DWI , DI _ADC , DI _FLAIR , DI _T1 , and DI _T2 ) in each brain image is closely related to the time of infarct region, so it is preferable to include brightness information.

4 is a view for explaining an example of a change in brightness of the infarct region in the DWI image according to the occurrence time of the infarct region.

The brightness of the ADC is shown as the inverse of the DWI brightness. Therefore, the method shown in FIG. 4 can be a method of roughly guessing the time of occurrence of a stroke with a single parameter. However, as described above, estimating the time of occurrence in a single parameter manner is poor in accuracy and reliability. The extent of the change in the infarct area in the DWI image over time depends on the location of the infarct area, the volume, the age of the patient, the sex, In addition, it is inappropriate to determine the point of occurrence from a specific DWI brightness value because the brightness change in the DWI image varies with time.

5 is a view for explaining a pattern in which quantitative values extracted from brain images change with time.

In this example, a multi-parameter is used to more accurately and more reasonably estimate the time of occurrence of the infarct region using brain images. MRI can generate various images for clinical purposes with various combinations of pulse sequences and appropriate combination of TR and TE as described above.

FIG. 5 illustrates that various quantitative parameters (QV1, QV2, QV3, QV4) vary with time from these various brain images. The change pattern of the quantitative parameters (QV1, QV2, QV3, QV4) in FIG. 5 is shown in a plausible pattern for the sake of convenience of explanation, and does not mean actually measured. In Fig. 5, QV1 is approximately similar to the pattern of DWI, and QV3 is similar to the pattern of ACD. The quantitative values (V _{I ,} V _IL V _P , V _PL , M, DI _DWI , DI _ADC , DI _FLAIR , DI _T1 , DI _{T2 ,} DP _DWI , DP _ADC , DP _FLAIR , DP _T1 , DP _T2 ) , It is possible to obtain patterns similar to or different from those shown in Fig. That is, the quantitative values varying depending on the occurrence time of the infarct region can be obtained.

For example, multi-modality brain images can be obtained from patients (clear-onset) in which the time of onset of the infarct zone is known. Quantitative value sets can be accumulated from these multimodal brain images. Animal experiments can also be used to accumulate a quantitative set of values for the same individual that varies over time from the occurrence of the infarct zone.

If a graph such as that shown in FIG. 5 is generated from the accumulated point-of-quantitative value set data, if the point of occurrence can not be determined from any one of the quantitative values, the point of occurrence is roughly specified with other quantitative values can do.

For example, the occurrence time point of the infarction area can be estimated to be approximately the interval between the first occurrence point and the second occurrence point by the quantitative value QV1. Further, the occurrence time point of the infarction area can be estimated to be approximately the interval between the third occurrence point and the fourth occurrence point by the quantitative value QV4. As described above, the occurrence time point can be more accurately estimated by estimating the occurrence time point of the infarction area using a plurality of quantitative values (multi-parameters) and obtaining the intersection of the estimated occurrence time points.

6 is a diagram for explaining an example of a correspondence relationship generation method. 7 is a diagram for explaining an example of the relationship between the quantitative value set and the occurrence time point of the infarction area.

In this example, a method of specifying the generation time point by the multi-parameter described in FIG. 5 is further generalized, and a corresponding relationship is used in which a quantitative value set is associated with the time of the infarct region.

Correspondence is provided, for example, by a classifier 300 trained or learned to map the quantitative value set to the time of the infarct region. As shown in FIG. 6, the extracted quantitative value set corresponds to the point in time of occurrence by the corresponding relation.

As an example of a classifier, a correspondence can be created using a statistical method such as multiple regression or a Support Vector Regression method. As another example, the correspondence relationship may be provided by curve fitting the relationship between the quantitative value set and the time of occurrence.

Accuracy and reliability of the correspondence relationship can be improved by training and learning by using the accumulated time point-quantitative value set data to associate the quantitative value set with the occurrence time of the infarct region.

Here, training and learning can mean that the corresponding relationship is corrected, supplemented or corrected as the point in time at which the quantitative value set data accumulates.

For example, it is possible to evaluate how accurately the correspondence relationship can be estimated by repeating the process of associating a set of quantitative values extracted from brain images with unknown occurrence time of the infarct region with the generation time.

In addition, the validity of the corresponding relationship can be evaluated by associating a quantitative value set with a known occurrence time of the infarction region with the generation time by the corresponding relationship.

Such a mapping is the ones described above, a quantitative value set shown in Figure 7; generate (P11, P12, P13, P14, P15, P16 Yes DI _DWI, V _I, DI _FLAIR, DI _T1, M, V _P) time (P1) in the multidimensional space corresponding to the point (onset1). This correspondence can be made for a plurality of times such as the occurrence time onset2.

Fig. 8 is a diagram for explaining an example of the correspondence relationship generated by the regression analysis.

As an example using a classifier, a multiple regression equation can be made by multiple regression analysis. As described above, it is possible to perform multiple regression analysis on the relationship between the quantitative value set and the time of occurrence by using the accumulated occurrence point-quantitative value set data. As a result, a regression equation (visualized as 400 in FIG. 8) may be provided that maps the quantitative value set to the time of occurrence.

For example, the multiple regression equation,

And can be defined as: Onset time = f (V _I , V _IL , V _P , V _PL , M, DI _DWI , DI _ADC , DI _FLAIR , DI _T1 , DI _T2 ,

In FIGS. 1 to 8, the method of estimating the occurrence time of the infarction area based on the brain image may be performed automatically by an application tailored to each process or may be performed together with a user interface.

Various embodiments of the present disclosure will be described below.

(1) The correspondence relationship is provided by classifying the quantitative value set by the occurrence time of the infarct region using a classifier, and estimating the time point of occurrence of the infarct region based on the brain image.

The method of estimating the time of onset of an infarct area based on brain imaging can be used to estimate the time associated with a lesion or tissue such as infarct. The time associated with the lesion or tissue may include the time of the lesion or tissue state change, the onset time of the cause of the lesion, and the like.

The method of estimating the time of occurrence of the infarct area based on the brain image can be applied to both humans and animals.

(2) segmenting the infarcted region using at least one of acquired brain images before the step of extracting the quantitative value set after the step of acquiring brain images; Dividing a peripheral region of the infarcted region using at least one of acquired brain images; And registering brain images including at least one brain image in which an infarction region or a peripheral region is divided.

(3) The correspondence relationship is defined as at least one of the group consisting of multiple regression method, support vector regression method, and curve fitting method, the quantitative value set accumulated from brain images of which the time of occurrence of the infarct region is known Wherein the classifier is configured to correspond to a time point of occurrence of the infarct region using the classifier.

(4) The step of extracting the quantitative value set includes: extracting at least one piece of information selected from the group consisting of the size information of the infarct region or the surrounding region, the position information, and the brightness information from the matched brain images A method for estimating the time of occurrence of an infarct region based on a brain image.

The quantitative value set may include some non-quantitative values. The quantitative value may be a value obtained by quantifying other characteristics (e.g., shape) of the infarct region in addition to size information, position information, and brightness information.

(5) At the stage where the quantitative value set corresponds to the occurrence time of the infarct area, the quantitative value set extracted from the brain images of which the infarct area is not known at the time corresponds to the occurrence time of the infarct area by the corresponding relationship, Wherein the correspondence relationship is corrected using the accumulated information by associating a quantitative value set in which the occurrence time point of the region is not known with the occurrence time point of the infarction region.

(6) The step of acquiring brain images includes: a process in which two or more selected from the group consisting of DWI image by MRI, ADC image, PWI image, FLAIR image, T1 image and T2 image are generated A method for estimating the time of occurrence of an infarct region based on a brain image.

The acquisition of brain images is a process in which brain images with different modalities are generated by MRI; And a process in which at least one brain image is generated by the CT.

(7) dividing the infarction region using the DWI image or the ADC image before the step of extracting the quantitative value set after the step of acquiring the brain images; Dividing a penumbra using a PWI image; And matching the brain images including the DWI image, the ADC image, or the PWI image divided into the peripheral region, with the infarct region being divided into the brain images.

(8) The step of extracting the quantitative value set consists of: the infarct volume (V _I ), the infarct location (V _IL ), and the volume of the surrounding area from the matched brain images. V _P), the location of the peripheral area (penumbra position; V _PL), the average brightness of the infarct area and the mismatch between the peripheral regions (M), infarct area in each brain imaging _{(DI DWI, DI ADC, DI} FLAIR, DI T1, in that it comprises; DI _T2), and an average brightness of the peripheral area in each brain image (DP _DWI, DP _ADC, acquiring two or more quantitative values selected from a quantitative value the group consisting of _{DP FLAIR, DP T1, DP T2} ) A method for estimating the time of occurrence of an infarct region based on a brain image.

(9) The step of the quantitative value set corresponding to the occurrence time of the infarct region corresponds to the following: the occurrence time point of the infarct region corresponds to the first section for any one of the quantitative values included in the quantitative value set; Wherein the occurrence time point of the infarct region corresponds to the second interval for another quantitative value included in the quantitative value set; And calculating the intersection interval of the first section and the second section to narrow the corresponding section of the infarct area at the time of occurrence of the infarct area.

According to the method of estimating the time of occurrence of the infarct region based on the brain image according to the present disclosure, it is possible to estimate the onset time of the lesion more accurately and more rationally using quantitative multi-parameters based on brain images .

Claims (10)

Obtaining brain images;
Extracting a quantitative value set that varies according to an onset time of an infarct region from an infarct region included in brain images; And
Wherein the quantitative value set corresponds to the occurrence time of the infarct region using a corresponding relation prepared so as to correspond to the occurrence time of the infarct region. Based on the estimated time of occurrence of the infarct region. The method according to claim 1,
Wherein the correspondence relationship is provided by classifying the quantitative value set by the occurrence time of the infarct region using a classifier. The method according to claim 1,
After the step of acquiring the brain images, before the step of extracting the quantitative value set,
Segmenting the infarcted region using at least one of acquired brain images;
Dividing a peripheral region of the infarcted region using at least one of acquired brain images; And
And a step of registering brain images including at least one brain image in which the infarct region or the peripheral region is divided. The method according to claim 1,
The correspondence relationship includes at least one of a group consisting of a multiple regression method, a support vector regression method, and a curve fitting method, which is a set of quantitative values accumulated from brain images in which the occurrence time of an infarction area is known Wherein the classifier is used to correspond to a time point of occurrence of the infarction area. The method of claim 3,
The step of extracting a quantitative value set includes:
And extracting at least one piece of information selected from the group consisting of size information of the infarct region or the surrounding region, position information, and brightness information from the matched brain images. . The method of claim 4,
The quantitative value set extracted from the brain images in which the occurrence time point of the infarct region is not known corresponds to the occurrence time point of the infarct region by the corresponding relationship at the stage where the quantitative value set corresponds to the occurrence time point of the infarct region,
Wherein the corresponding relationship is corrected using information accumulated by associating a quantitative value set in which the occurrence time point of the infarction area is unknown with the occurrence time point of the infarction area, and the corresponding relationship is corrected. The method according to claim 1,
The steps in which brain images are acquired include:
And a process in which at least two selected from the group consisting of DWI image, ADC image, PWI image, FLAIR image, T1 image, and T2 image by MRI are generated are included. Estimation method. The method of claim 7,
After the step of acquiring the brain images, before the step of extracting the quantitative value set,
Dividing an infarction area using a DWI image or an ADC image;
Dividing a penumbra using a PWI image; And
A step of matching brain images including a DWI image, an ADC image, or a PWI image in which a peripheral region is divided, the infarcted region being divided; and a method of estimating a time point of occurrence of an infarction region based on a brain image. The method of claim 8,
The step of extracting a quantitative value set includes:
The infarct volume (V _I ), the infarct location (V _IL ), the penumbra volume (V _P ), and the penumbra position of the infarct volume from the matched brain images. V _PL), the average of the infarct region and a mismatch between the peripheral regions (M), the average brightness of the infarct area in each brain imaging _{(DI DWI, DI ADC, DI} FLAIR, DI T1, DI T2) and a peripheral region on each neuroimaging Obtaining quantitative values of two or more quantitative values selected from a group of quantitative values consisting of brightness (DP _DWI , DP _ADC , DP _FLAIR , DP _T1 , and DP _T2 ) . The method according to claim 1,
Wherein the quantitative value set corresponds to the time of occurrence of the infarction zone:
Wherein the occurrence time point of the infarct region corresponds to a first interval with respect to any one of the quantitative values included in the quantitative value set;
Wherein the occurrence time point of the infarct region corresponds to the second interval for another quantitative value included in the quantitative value set; And
And calculating the intersection interval of the first interval and the second interval to narrow the corresponding interval of the occurrence time of the infarction zone.

Images