JP2743796B2 - Correction method for nuclear medicine imaging device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a nuclear medicine diagnostic apparatus, and more particularly to a SPECT (Single Photo) constructed using a scintillation camera and this camera.
n Emission Computed Tomog
The present invention relates to a correction method for a nuclear medicine imaging device such as a raph device.

[0002]

2. Description of the Related Art A nuclear medicine imaging apparatus administers a radioisotope (RI) to a subject and, when accumulated in a specific organ or the like, detects radiation emitted from the body to the outside of the body to obtain emission data. It collects and obtains an image. Here, the radiation emitted from the RI causes Compton scattering in the tissue inside the body before coming out of the body from the body.
Therefore, it is necessary to perform correction on the obtained emission data so as to eliminate the influence of the scattering.

For this reason, an energy-weighted scattered radiation removal correction method has been proposed. The energy weighting function used therein is disclosed in Japanese Patent Application No. 60-298164 (Japanese Patent Application Laid-Open No.
The scattering coefficient is calculated from the scattered radiation removal correction coefficient f {E, (Ai / Si)} = (Ao / So) / (Ai / Si). Here, Ai is a count at an arbitrary energy E, and Si is a count in a higher energy region than a peak in the photoelectric effect region in the energy spectrum. Ao and So are counts when there is no scatterer. That is, a scatterer causing the same degree of scattering as the object is measured in advance, an energy spectrum is obtained for each thickness of the scatterer, and Ai and Si for each E are obtained for each spectrum. The values Ao and So when there is no body are obtained.

[0004]

However, in the conventional correction method, it is preferable to collect single nuclides with a single peak. However, when there is a background such as simultaneous collection of two nuclides or multi-peak with only one nuclide, the conventional correction method has a problem. In addition, there is a problem that an error occurs in the correction. In addition, the error increases as the amount of the background increases, and the amount of the background is troublesome because the amount of the background differs for each inspection and each position.

In view of the above, the present invention makes it possible to satisfactorily perform scattered radiation removal correction in the case of simultaneous collection of two nuclides or in the case of multiple peaks even in the case of single nuclide collection, as in the case of single nuclide collection of a single peak. It is an object of the present invention to provide a method of correcting a nuclear medicine imaging device which is improved as much as possible.

[0006]

In order to achieve the above object, a correction method for a nuclear medicine imaging apparatus according to the present invention uses a weighting function to add weights to each count of a measured energy spectrum. As the energy weighting function, a positive value in the range of the photoelectric effect region, a negative value in each range of the region on the low energy side and the region on the high energy side adjacent to the photoelectric effect region,
It is characterized in that a material having a value of zero in other energy ranges and having such a characteristic that the first-order integral value in the entire energy range becomes zero is used.

[0007]

[Action] 1 in the entire energy range of the energy weight function
By calculating the function so that the next integral value becomes zero, and using the function to weight and add each count of the measured energy spectrum, if the energy spectrum includes a background component, the background The component of the constant term of the ground component is removed.
Therefore, when the amount of the background can be approximated to be constant with respect to the energy, effective scattered radiation removal correction can be performed regardless of the increase or decrease of the amount of the background.

[0008]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, a preferred embodiment of the present invention will be described in detail with reference to the drawings. First, data for obtaining a correction coefficient is collected using the system shown in FIG. This system is mainly composed of a scintillation camera 11, and a position signal X,
Y, an energy signal Z, and an unblank signal are sent to the interface circuit 12. The energy signal Z is input to the multi-channel analyzer 13 via the interface circuit 12, and the wave height of the energy signal Z is analyzed. In this way, energy analysis is performed for each radiation incident event on the scintillation camera 11, and the result is sent to the energy spectrum collection device 14 to perform counting for each energy. In the energy spectrum collecting apparatus 14, since the counting for each energy is performed for each radiation incident position represented by the position signals X and Y, an energy spectrum is obtained for each incident position.

First, the scatterer 20 is measured prior to the actual measurement on the subject, and energy spectrum data for obtaining a correction coefficient is collected. Since a subject is usually a human body, a scatterer 20 that produces scattering similar to a human body is used as the scatterer 20, and energy spectrum data is collected for portions having different thicknesses.
For example, the scatterer 20 is formed in a wedge shape, and radiation sources 30, 31, 32,... Are arranged at various thickness portions, and the energy spectrum as shown in FIG. Get. 2A is an energy spectrum at the position of the source 30, FIG. 2B is an energy spectrum obtained at the position of the source 31, and FIG. 2C is an energy spectrum obtained at the position of the source 32. It is a spectrum. At the position of the source 30, the thickness of the scatterer 20 is zero and there is no Compton scattering.
Since the scatterer 20 has a small thickness at the position of the radiation source 31 and has a larger thickness at the position of the radiation source 32, Compton scattering occurs corresponding to the thickness.

[0010] In each spectrum of FIG. 2, a dotted line indicates a background distribution. In the case of simultaneous collection of two nuclides, such as the combination of Tl-201 and Tc-99m or the combination of Tl-201 and I-123, there is a crosstalk between nuclides, and the spectra of nuclides on the low energy side show high spectra. Such a background is observed because a count due to scattered radiation on the energy side is added. The same applies to a single nuclide in the case of a multipeak. Here, the radiation sources 30, 31, 32,... Are the same combination of two nuclides or the same multi-peak one nuclide as administered to the actual subject.

When the energy spectrum of the scatterer 20 at various thicknesses is obtained, a region on the higher energy side than the peak in the photoelectric effect region (for example, an energy region of 105% to 110% of the energy of the photoelectric peak).
The count So, Si (i = 1, 2,...) At S is obtained. Since this region S is a region which is not affected by Compton scattering due to radiation from the nuclide having the photoelectric peak, assuming that there is no background indicated by a dotted line, the count in this region S is referred to. By normalization, a spectrum as shown in FIG. 3 can be obtained. From FIG. 3, if Ao / A1n and Ao / A2n are obtained, when the actual measurement counts are A1n and A2n, the values are multiplied by Ao / A1n and Ao / A2n to obtain a case where there is no scatterer. It can be seen that the value can be converted to Ao. Ao, Ai (i = 1, 2,...) Are counts at an arbitrary energy E, and A1n and A2n are normalized on the basis of the count in the region S, that is, A1n = A1 · (So / S1), A2n = A2 · (S
o / S2). That is, the scatterer 20 differs from the prior art.
The scattered radiation correction coefficient f {E, (Ai / Si)} = (Ao / So) / (Ai /
If an energy spectrum is obtained at a certain position with respect to the actual subject, the ratio Ai / Si with respect to the count Ai at an arbitrary energy E in the energy spectrum can be obtained from the Si obtained from the energy spectrum. The above function f is determined by Si and its energy E.
When multiplied by the correction value obtained from {E, (Ai / Si)},
This can be converted to a count when there is no scatterer, and scattered radiation removal correction can be performed. However, in reality, there is a background, and scattered radiation removal correction cannot be accurately performed by such correction.

Therefore, the photoelectric effect region P is set to the energy E2
To E3 (for example, 90% of the energy of the photopeak).
% To 110%), and consider the low energy side region SL and the high energy side region SH adjacent thereto. The region SL is a region of energy E1 to E2 (for example, 85% to 90% of the energy of the photoelectric peak), and the region SH is a region of energy E3 to E4 (for example, 110% to 115% of the energy of the photoelectric peak). And the energy weighting function f (E) in the region P is
Counts Ai and Si at an arbitrary energy E obtained from the energy spectrum actually measured for the subject
Is determined according to the above function f {E, (Ai / Si)}. Here, f (E) ≧ 0. It is assumed that f (E) ≦ 0 in the regions SL and SH, and f (E) = 0 in regions outside these, that is, in the regions of E <E1 and E> E4. The energy weighting function f (E) thus obtained is weighted and added to each count of the measured energy spectrum to perform scattered radiation removal correction.

The energy spectrum a (E) is represented by the sum of a net spectral component b (E) such as the photoelectric peak of the radiation of interest and its related Compton scattering and a background component c (E), a (E) = b (E) + c (E). Here, assuming that c (E) can be approximated by c (E) ≒ αE + β (α, β: constant), the energy spectrum a
The weighted addition operation of the energy weighting function f (E) for (E) is performed by the following equation 1.

(Equation 1) Will be represented by Therefore, the first-order integral value (one-time integral) of the energy weighting function f (E) becomes zero, that is,

(Equation 2) F in the range of E1 to E2 and E3 to E4 to satisfy
Determine the value of (E). Then, this f (E) becomes like A, B, C in FIG. 4, and these are A, B in FIG.
4A corresponds to B and C, FIG. 4A shows the case where the scatterer 20 is not provided, FIG. 4B shows the case where the scatterer 20 is thin, and FIG.
Is thick.

The measurement performed by administering the same nuclide RI to the actual subject is performed by using the scattered radiation correction coefficient f {E, (Ai / Si)} obtained by the measurement of the scatterer 20 for correction data. This is performed after the processing up to the storage in the memory 16 is completed. In the measurement, when an energy spectrum is obtained by the energy spectrum collecting device 14 for each position represented by the position signals X and Y from the scintillation camera 11, the correction arithmetic device 15 The correction coefficient f ｛E, (A
i / Si)｝ and count A at an arbitrary energy E obtained from the obtained energy spectrum.
The energy weighting function f (E) in the region P is calculated from the ratio Ai / Si between i and Si and its energy E. Further, f in the range of E1 to E2 and E3 to E4 is set so that the first-order integral value of the energy weighting function f (E) becomes zero.
Determine the value of (E). The energy weighting function f (E) thus obtained is weighted and added to each count of the measured energy spectrum. Then, since the above equation 2 is satisfied, the third term on the right side of the above equation 1 becomes zero, and the energy weighting correction can be performed without being affected by the amount of the component β of the background constant term. I understand. Therefore, when the amount of the background can be approximated to be constant with respect to the energy (α ≒ 0), effective scattered radiation removal correction can be performed regardless of the increase or decrease of the amount of the background.

Further, the following equation (3)

(Equation 3) Holds. This is because F (E1) = F (E4) = 0. Therefore, the energy weighting function f
(E) so that the quadratic integrated value (integrated twice) becomes zero, that is,

(Equation 4) F in the range of E1 to E2 and E3 to E4 to satisfy
Determine the value of (E). Then, this f (E) becomes like A, B, and C in FIG. 5, which are respectively A, B, and C in FIG.
5A corresponds to B and C, FIG. 5A illustrates the case where the scatterer 20 is not provided, FIG. 5B illustrates the case where the scatterer 20 is thin, and FIG.
Is thick. F (E) corresponding to each of these is as shown by the solid lines of A, B, and C in FIG. For comparison with this, FIG.
When F (E) is obtained from f (E) of FIG. 6, it becomes as shown by dotted lines of A, B, and C in FIG.

(Equation 5) And the coefficient (slope) α of the first-order term of the background
It can be seen that it is affected by

When the energy spectrum is obtained by the measurement on the actual subject, the correction arithmetic unit 15
However, the region P portion of the energy weighting function f (E) is obtained in the same manner as described above, and for the values in the range of E1 to E2 and E3 to E4, the first and second integral values of f (E) are The energy weighting function f (E) is obtained assuming that it becomes zero, and the energy weighting function f (E) is weighted and added to each count of the measured energy spectrum. In this case, since Equations 2 and 4 are satisfied, the third term on the right side of Equation 1 becomes zero, and the second term on the right side of Equation 1 becomes zero. Coefficient (slope) α of first-order term
And the energy weighting correction can be performed without being affected by both the amount of the component β of the constant term.

By performing the scattered radiation removal correction on the energy spectrum at each position by such correction calculation processing, a scintigram excellent in quantitative performance can be obtained. In addition, the scintillation camera 11 is rotated around the subject to perform measurements in all directions around the subject, and only the data on the desired slice plane is taken out of the data collected in each direction, and the data is back-projected as projection data. By performing the calculation, the emission CT image can be reconstructed. In this case, however, the value of each pixel of the reconstructed image can be corrected for Compton scattered radiation without influence of the background. Increase.

In the above embodiment, f (E) in the energy range of energy E2 to E3 was obtained by measuring the scatterer 20 (Ao / So) / (Ai / S).
Although it is determined based on i), it can be determined by other equations. Energy weighting function f
(E) is determined according to the energy spectrum (specifically, determined from the scattered ray correction coefficient f {E, (Ai / Si)} using Ai / Si and E as described above).
Uses a single energy weighting function, f (E), derived from a representative one of the energy spectra, whereby the energy weighting is independent of the energy spectrum measured on the actual subject and is merely dependent on its energy. The scattered radiation removal correction may be performed. In addition, various changes can be made without departing from the spirit of the present invention.

[0019]

As described above, according to the correction method of the nuclear medicine imaging apparatus of the present invention,
With little effect on increasing or decreasing the amount of background,
Energy weighted scattered radiation removal correction can be performed with high accuracy. Therefore, especially, for example, Tl-201
In the case of simultaneous collection of two nuclides, such as the combination of Tc-99m or Tl-201 and I-123, variations in the dose of each nuclide, differences in shape between subjects, and drugs labeled with the nuclide Excellent scattered radiation removal correction can be performed without depending on the type of the sample, the variation in the intake rate among the subjects, and between the sites, and the like, and the quantitativeness of the image can be improved. In the collection of one nuclide, for example, Tl
Even in the case of multi-peaks such as -201, In-111, and Ga-67, good scattered radiation removal correction can be performed without depending on the difference in shape between subjects, etc., and the quantitativeness of the image is improved. can do.

[Brief description of the drawings]

FIG. 1 is a block diagram of one embodiment of the present invention.

FIG. 2 is a graph showing an energy spectrum.

FIG. 3 is a graph showing a normalized spectrum.

FIG. 4 is a graph showing an example of an energy weighting function f (E).

FIG. 5 is a graph showing another example of the energy weighting function f (E).

FIG. 6 shows a linear integral function F of an energy weighting function f (E).
The graph showing an example of (E).

[Explanation of symbols]

 REFERENCE SIGNS LIST 11 scintillation camera 12 interface circuit 13 multi-channel analyzer 14 energy spectrum collection device 15 correction operation device 16 memory for correction data 20 scatterer 30, 31, 32 radiation source

Claims (1)

(57) [Claims] 1. A positive value in a range of a photoelectric effect region, a negative value in each of a low energy side region and a high energy side region adjacent to the photoelectric effect region, and a zero value in other energy ranges. Nuclear medicine imaging apparatus having a weighted addition to each count of a measured energy spectrum using an energy weighting function having a characteristic such that a first-order integral value in the entire energy range is zero. Correction method.