JPH06181913A - Mri cinerama image processing device - Google Patents

Abstract

PURPOSE:To provide an MRI Cinerama image processing device capable of handling the density constant at the portion not changed with the signal intensity. CONSTITUTION:An MRI Cinerama image processing device receiving the raw image data of multiple images to generate the Cinerama image is provided with an region of interest selecting means 11 selecting the noteworthy region of interest of the portion having a particularly large fluctuation of the picture element density, a standard pre-processing means 12 determining the picture element signal intensity of the raw image data corresponding to the image display highest level and the image display lowest level for processing the Cinerama image from the signal intensity of the picture element in the region of interest and the picture element outside the region of interest on the raw image data of multiple images as the standard value, and an evaluating means 13 evaluating the raw image data of multiple images based on the standard value determined by the standard pre-processing means 12 to generate the Cinerama image data.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an improvement of an MRI cine image processing apparatus for capturing a synchronous image by collecting an ecological signal of the heart or the like as a trigger pulse in the MRI apparatus.

[0002]

2. Description of the Related Art An X-ray CT image is an image displayed by using an absolute value determined by a CT value according to a tissue. That is, by using the X-ray transmittance of other soft tissues as the CT value with water, air, and bone as the reference, the image gradation expression is used based on the raw image data that is almost close to the absolute value.

On the other hand, the MRI image is displayed by using the relative values standardized by setting the pixel having the highest signal intensity and the pixel having the lowest signal intensity in one image as 100% and 0%, respectively. That is, there is no reference signal intensity or sample substance in the raw image data of the MRI image, and the signal intensity depends on the device (especially the finished state of the magnet) and the pulse sequence, and the detection depends on the device external noise and the device internal noise. The lightness and darkness of the image are expressed relative to each other with the limit difference as the signal intensity. For example, in the raw image data obtained by the apparatus, the value of the pixel showing the maximum signal intensity is standardized to “1” and the value of the minimum signal intensity is standardized to “0”, and other pixels are interpolated or a weighting function is set. The multiplied value is used as the density value.

Therefore, if there is a portion accompanied by an extreme signal change (time change) in the same cross section of the same sample imaged under the same imaging condition, the entire image is obtained by using the signal intensity of that pixel. The imaging time to be standardized and the imaging time using other pixels are generated.

[0005]

Therefore, in the case of a synchronous image in which the imaging time is determined by a biological signal, for an imaging region that strongly indicates the state of the fluid (velocity, turbulence, etc.) such as the heart, However, it was not possible to express a series of images of the same cross-section that follow the time change in the same tissue and the same region with the same density. Therefore, the MRI cine image is inferior from the viewpoint of stability as compared with the cine image using the X-ray CT image.

FIG. 7 is an explanatory view showing an example of a cross section of a living body. Here, centering on the heart cross section 1, the lung 2, the skin fat layer 3
There is a bone 4 between the lung 2 and the skin fat layer 3. Also, arms 5 and 6 are present at both ends.

8 to 10 are explanatory views showing similar cross sections at each time. In MRI, a cross section of a target sample is excited by a high frequency magnetic field to create a resonance state. After that, some magnetic field modulation is applied. Then, the information including the history of modulation is measured as a high-frequency magnetic field signal and taken into the device. Here, a finite time elapses from the high frequency excitation to the signal measurement. Therefore, a change in intensity occurs in the signals obtained from the cross section of the stationary sample and the cross section containing the fluid. 8 to 10, the skin fat layer 3 and the bone 4 are shown.
Is a stationary sample and the blood part in the heart 1 is a sample containing flow.

By the way, the flow velocity and the signal intensity cannot maintain a linear relationship except for measurement under limited conditions. Generally, it is a multi-valued function. In other words, the cross-sectional signal strength of the blood part changes greatly in a cycle that matches the heartbeat.

FIG. 8 shows a state in which the signal strength of the heart 1 is the lowest. Here, pixels other than the heart 1 are standardized as the maximum signal value. FIG. 9 shows the state where the signal intensity of the heart 1 portion is the maximum, and the pixel of this heart portion is selected as the maximum signal value and standardized. In FIG. 10, the heart 1
Blood flow is larger than that in FIG. 9, and the pixels in other portions are relatively low.

As described above, as a cine image which is displayed in time series in order to make it easy to recognize a time-varying portion, the cine image shown in FIG.
~ Using Fig. 10, there was a problem that it was perceived that the signal intensity of the bone or the skin fat layer, which should not change originally, was recognized as changing.

The present invention has been made in view of the above points, and an object of the present invention is to realize an MRI cine image processing apparatus capable of handling the signal intensity of a portion that does not change constantly.

[0012]

SUMMARY OF THE INVENTION The present invention which solves the above-mentioned problems is an MRI cine image processing apparatus which receives raw image data of a plurality of images and generates a cine image. The pixel signal intensity of the raw image data corresponding to the image display maximum level and the image display minimum level for cine image processing is standardized from the region selection means and the signal intensity of the pixels in the region of interest for the raw image data of a plurality of images. A standard pre-processing means for obtaining a value, and an evaluation means for generating cine image data by evaluating raw image data of a plurality of images based on the standard value obtained by the standard pre-processing means. It is a feature.

[0013]

The region of interest is set by the region of interest selecting means, the standard value is obtained from the signal intensities of the pixels in the region of interest for the raw image data of the plurality of images, and the standard value is used as the raw image data of the plurality of images. A cine image is generated by performing evaluation as a common standard value of. Since the cine images obtained in this way are evaluated based on a common standard, portions having the same signal intensity between the cine images are represented by the same density.

[0014]

Embodiments of the present invention will now be described in detail with reference to the drawings. FIG. 1 is a configuration diagram showing a schematic configuration of an MRI cine image processing apparatus according to an embodiment of the present invention. In this figure, the MRI cine image processing apparatus 10 has a region of interest (hereinafter,
ROI selection device 11 for selecting an instruction of “ROI”, a standard preprocessing circuit 12 for determining a standard value (maximum value and minimum value of signal intensity) through raw image data of a plurality of images, and a standard value. An evaluation circuit 13 for evaluating raw image data of a plurality of images to generate cine image data based on the control circuit 14, a control circuit 14 for controlling each part, and a display control circuit 15 for outputting the cine image to an external display device. .

The operation of the apparatus of this embodiment thus constructed is as follows. First, by a known method, data is collected by utilizing nuclear magnetic resonance with a trigger signal synchronized with the time phase of the living body, and this is converted into raw image data.
Then, such raw image data is converted into cine image data by the MRI cine image processing apparatus 10 of the present embodiment by the method described below, and cine display is performed.

Raw image data of each time phase (hereinafter referred to as time phase data) is two-dimensional image information, and I, J
It is composed of pixels of (I = 1 to m, J = 1 to n). The signal strength Pi, j of the pixel is the grayscale information. Therefore,
For each image Lq corresponding to a general MRI image, an image provisionally standardized with a maximum value Pmax = 1 and a minimum value Pmin = 0 is obtained by the first step processing (where q = 1 to 0). (In the process of cine display,
L _{1 to} L ₀ , L _{1 to} L ₀ , ... Are continuously displayed. ) Hereinafter, selection of ROI, pre-standard processing for obtaining a standardization coefficient, evaluation based on the standardization coefficient and each operation will be described.

[Selection of ROI] First, the display control unit 15 generates a provisionally standardized MRI image shown in FIG. 2 from the raw image data, and this is temporarily displayed on an external display device. The inspector selects the image of the temporarily displayed Lq and judges the state of the time phase change. One image Lq 'in which the judgment result is remarkably expressed on the image is displayed on the display device. If FIG. 3 is the selected image Lq ′, i =
A group of pixels Pi, j designated by k to k + 3 and j = 1 to l + 3 is judged to be a heart part where the temporal change is particularly remarkable.
In this way, the ROI is selected via the ROI selection device 11.

[Standard preprocessing] A group of pixels Pi, j designated by i = k to k + 3, j = 1 to l + 3 is cut out from all the images of L _{1 to} L ₀ , and raw images corresponding to the pixels are cut out. From the image data, the one showing the maximum signal strength and the one showing the minimum signal strength are extracted by comparison. Then, the maximum value of the raw image data is stored as Pmax and the minimum value is stored as Pmin in a register or the like in the standard preprocessing circuit 12. Further, as shown in FIG. 4, the average value of the signal intensities of the pixels in the region other than the ROI thus cut out is obtained and similarly stored as Pave.

Further, the standard preprocessing circuit 12 obtains a histogram of all pixels as shown in FIG. In this histogram, the vertical axis represents the frequency of pixel appearance and the horizontal axis represents the signal intensity of the pixel, and the values are from "Pmin" to "Pmax". Then, the input variable Hnoise is the gradient of the hist space (pixel appearance frequency /
The signal strength of the pixel) is input in advance. From the signal intensity section on the origin side of the horizontal axis of the histogram, the next signal intensity section is evaluated using Hnoise. The value of the section to be evaluated is
The point of the section that exceeds “the pixel appearance frequency of the section on the origin side” + “Hnoise” × “the section interval of the signal intensity” is set as a temporary Pnoise. Image of the previous extraction area (i = k to k + 3, j
Similar processing is performed for a group of pixels Pi, j) designated by = 1 to 1 + 3. When the noise level obtained from the extraction area is lower than the previous Pnoise, the value obtained here is set as a new Pnoise. Then, a standard value Nv is set so that Pnoise is "0" and Pmax is "100". This makes it possible to remove a signal having an intensity equal to or lower than Pnoise as noise.

Using this Nv, the raw image data of L _{1 to} L ₀ are standardized, and the average value of their signal intensities is obtained. Here, if the average value of the signal intensity after standardization falls below the lower limit range f (Pave) of the average value that is predetermined as the lower limit signal intensity that can be easily visually recognized, it is obtained by normalizing the pixels. The standard value Nv is recalculated so that the average value of the signal strength matches f (Pave).

[Evaluation] Using the standard value Nv obtained by the standard preprocessing circuit 12 obtained as described above as the common standard value of the raw image data of a plurality of images, the evaluation circuit 13 sets L ₁ to Normalize each pixel of L ₀ . At this time, if the signal strength after normalization exceeds 1, 1
Treat as.

As described above, the common standard value is obtained from the raw image data of a plurality of images, and the standardization is performed according to this standard value. As shown, it becomes possible to handle a portion where the signal intensity does not change at the time of displaying a cine image, with a constant density.

In the above embodiment, the ROI is first evaluated and then evaluated again. However, when the ROI is required inside the apparatus, R is calculated.
No evaluation is required when selecting OI.

Further, in the above embodiments, the inspector has been described as selecting the ROI, but the ROI selecting device 11 may automatically perform this. For example, the ROI selection device 11 takes the sum of the signal intensities of the pixels for each of the images L _{1 to} L ₀ . Let these values be Σ ₁ to Σ ₀ , respectively. Then, the maximum value Σma of Σ _{1 to} Σ ₀
x, minimum value Σmin is calculated. Next, Σmax, minimum value Σmin
For the two images corresponding to, the corresponding pixel of the image containing the minimum value is subtracted from each pixel of the image containing the maximum value. A histogram is obtained using this result. Then, the minimum value that appears first from the low frequency side (origin) of the histogram is used as the determination value, and the region of pixels that exceed this determination value is automatically set as the ROI. When the ROI selection device 1 performs the above, the ROI is automatically selected.

[0025]

As described in detail above, according to the present invention, the standard value is obtained from the signal strength of the pixel in the region of interest for the raw image data of the plurality of images, and the standard value is calculated for the plurality of images. An MR that can treat the signal strength of the unchanged portion as a constant density by performing evaluation as a common standard value of raw image data to generate a cine image
An I-cine image processing apparatus can be realized.

[Brief description of drawings]

FIG. 1 is a configuration diagram showing a configuration of an apparatus according to an embodiment of the present invention.

FIG. 2 is an explanatory diagram for explaining the operation of the apparatus according to the embodiment of the present invention.

FIG. 3 is an explanatory diagram for explaining an operation of the apparatus according to the embodiment of the present invention.

FIG. 4 is an explanatory diagram for explaining the operation of the apparatus according to the embodiment of the present invention.

FIG. 5 is a histogram for explaining the operation of the apparatus according to the embodiment of the present invention.

FIG. 6 is an explanatory diagram showing an example of a cine image obtained by the apparatus according to the embodiment of the present invention.

FIG. 7 is an explanatory diagram showing an example of a cine image obtained by a conventional cine image device.

FIG. 8 is an explanatory diagram showing an example of a cine image obtained by a conventional cine image device.

FIG. 9 is an explanatory diagram showing an example of a cine image obtained by a conventional cine image device.

FIG. 10 is an explanatory diagram showing an example of a cine image obtained by a conventional cine image device.

[Explanation of symbols]

 10 MRI Cine Image Processing Device 11 ROI Selection Device 12 Standard Preprocessing Circuit 13 Evaluation Circuit 14 Control Circuit 15 Display Control Circuit

Claims (3)

[Claims] 1. In an MRI cine image processing apparatus for receiving raw image data of a plurality of images obtained by nuclear magnetic resonance to generate a cine image, a region of interest selecting means (1) for selecting a region of interest to be noticed.
1) and the pixel signal strength of the raw image data corresponding to the highest image display level and the lowest image display level for cine image processing from the pixel signal strengths in the region of interest for the raw image data of a plurality of images as standard values Required standard preprocessing means (12)
And an evaluation unit (13) for generating cine image data by evaluating raw image data of a plurality of images based on the standard value obtained by the standard preprocessing unit. Cine image processing device. 2. The standard preprocessing means obtains a maximum value of signal intensity from pixels in a region of interest for raw image data of a plurality of images, and obtains a histogram of all pixels,
The MRI cine image processing apparatus according to claim 1, comprising means for obtaining a noise level from these and means for setting a value obtained from these as a standard value. 3. The standard preprocessing means performs standardization by the standard value thus found to find an average value of all pixels of all raw image data, and when the average value is smaller than a preset set value, The M according to claim 1, further comprising means for re-determining a standard value such that the average value and the set value match.
RI cine image processing device.