Image processing device and image processing method

Abstract

An image processing apparatus and an image processing method are disclosed. The image processing apparatus comprises: a three-dimensional volume data acquisition unit for acquiring three-dimensional volume data of an examination area of ​​a subject; a two-dimensional image data acquisition unit for acquiring multiple two-dimensional image data of a portion of the examination area; an image quality index calculation unit for calculating an image quality index of the multiple two-dimensional image data; a candidate two-dimensional image data determination unit for calculating an overall image quality index of a predetermined number of two-dimensional image data based on the image quality index of the multiple two-dimensional image data, and determining the predetermined number of two-dimensional image data as candidate two-dimensional image data when the overall image quality index is above a pre-set image quality index standard value for the examination area; and an image data registration unit for registering the candidate two-dimensional image data or thin three-dimensional volume data reconstructed based on the candidate two-dimensional image data with the three-dimensional volume data.

Description

Technical Field

[0001] The present invention relates to an image processing apparatus and an image processing method capable of registering two-dimensional images with three-dimensional images or three-dimensional images with each other, i.e., registering multimodal images. Background Technology

[0002] In the field of medical imaging, during examinations or treatments, for the examination site of the subject, it is necessary to register the two-dimensional or three-dimensional image data of the examination site with the three-dimensional image data obtained before the examination or treatment.

[0003] In existing technologies, when diagnosing or performing surgery on the examination site of a subject, there is DRR-CT registration, which is used to register two-dimensional image data with three-dimensional image data. In DRR-CT registration, CT (computed tomography) image data of the examination site is obtained in advance. During diagnosis or surgery, DRR image data (X-ray data) as two-dimensional data is obtained. The DRR image data is registered with the CT image data, thereby analyzing and diagnosing the examination site.

[0004] However, when ultrasound (US) guidance is used during diagnosis or surgery, it is necessary to register the real-time US image data (two-dimensional image data) with the CT image data or MR image data of the examination site obtained in advance. That is, to perform US-CT / MR registration. However, not all US image data obtained is clear enough to meet the registration requirements. Therefore, there are often cases where the US image data is not clear and the registration with CT image data or MR image data fails.

[0005] Furthermore, since the image quality deviation between US image data and CT or MR image data is greater than that between DRR image data and CT or MR image data, the DRR-CT registration technique cannot be applied to the aforementioned US-CT / MR registration.

[0006] Furthermore, in existing technologies, when diagnosing or performing surgery on the examination site of a subject, there are techniques for registering the acquired three-dimensional US image data with pre-acquired three-dimensional US image data, CT image data, or MR image data. When a subject breathes, the size and position of organs such as the heart or liver, which are the examination sites, can change. Therefore, it is necessary to register the overall three-dimensional US image data of the heart or liver with pre-acquired three-dimensional image data. However, the process of acquiring the data for reconstructing the overall three-dimensional US image data of the heart or liver is complex and time-consuming, making this registration unsuitable, especially during surgery. Summary of the Invention

[0007] Therefore, in view of the above, the present invention provides an image processing apparatus and an image processing method that can automatically obtain multiple two-dimensional US image data that meet the registration requirements when registering between multimodal images including two-dimensional US images.

[0008] The image processing apparatus of the present invention includes: a three-dimensional volume data acquisition unit for acquiring three-dimensional volume data of an examination area of ​​a subject; a two-dimensional image data acquisition unit for acquiring multiple two-dimensional image data of a portion of the examination area; an image quality index calculation unit for calculating an image quality index of the multiple two-dimensional image data; a candidate two-dimensional image data determination unit for calculating an overall image quality index of a predetermined number of two-dimensional image data based on the image quality index of the multiple two-dimensional image data, and determining the predetermined number of two-dimensional image data as candidate two-dimensional image data when the overall image quality index is above a pre-set image quality index standard value for the examination area; and an image data registration unit for registering the candidate two-dimensional image data or thin three-dimensional volume data reconstructed based on the candidate two-dimensional image data with the three-dimensional volume data.

[0009] The image processing method of the present invention includes: a step of acquiring three-dimensional volume data of an examination area of ​​a subject; a step of acquiring multiple two-dimensional image data of a portion of the examination area; a step of calculating an image quality index of the multiple two-dimensional image data; a step of calculating an overall image quality index of a predetermined number of two-dimensional image data based on the image quality index of the multiple two-dimensional image data, and determining the predetermined number of two-dimensional image data as candidate two-dimensional image data when the overall image quality index is above a preset image quality index standard value for the examination area; and a step of registering the candidate two-dimensional image data or thin three-dimensional volume data reconstructed based on the candidate two-dimensional image data with the three-dimensional volume data.

[0010] Invention Effects

[0011] According to the image processing apparatus and image processing method of the present invention, when performing registration between multimodal images including two-dimensional US images, it can automatically obtain two-dimensional image data or small three-dimensional volume data suitable for registration with pre-obtained three-dimensional volume data, and can perform good and stable registration in real time, thus avoiding registration failure. Attached Figure Description

[0012] Figure 1 This is a structural block diagram of the image processing apparatus of the present invention.

[0013] Figure 2 This is a graph showing an example of the calculated overall image quality index.

[0014] Figure 3 This is a diagram illustrating the method for calculating the overall image quality index of a specified amount of two-dimensional US image data.

[0015] Figure 4 This is a diagram illustrating an implementation method for registering images between multiple modalities.

[0016] Figure 5 This is a diagram illustrating another implementation of a registration method between multimodal images. Detailed Implementation

[0017] The image processing apparatus and image processing method of the present invention will now be described with reference to the accompanying drawings.

[0018] Implementation Method 1

[0019] Figure 1 This is a block diagram of the image processing device 10.

[0020] like Figure 1 As shown, the image processing device 10 includes a three-dimensional volume data acquisition unit 11, a two-dimensional image data acquisition unit 12, an image quality index calculation unit 13, a candidate two-dimensional image data determination unit 14, and an image data registration unit 15.

[0021] The three-dimensional volume data acquisition unit 11 acquires the three-dimensional volume data of the examination area of ​​the subject.

[0022] Before examining or treating a patient, a three-dimensional scan is usually performed to obtain comprehensive information about the area to be examined, such as three-dimensional ultrasound (US), three-dimensional CT (computed tomography), or three-dimensional MR (magnetic resonance) data. The area to be examined can be a single organ, such as the liver or heart, or multiple organs including the liver and heart.

[0023] The two-dimensional image data acquisition unit 12 acquires multiple two-dimensional image data of a portion of the inspected area.

[0024] When examining a site such as the liver, a two-dimensional ultrasound (US) scan of the liver is usually required. During the examination of the liver, the probe is pressed against the liver area in the patient's abdomen, and the probe is slid or rotated to the appropriate position to obtain multiple real-time two-dimensional US images of a portion of the liver.

[0025] The multiple 2D US image data contain relatively little image information, so it is necessary to register these multiple 2D US image data to pre-obtained 3D volume data, such as 3D MR volume data, for further judgment.

[0026] When the image contours and grayscale variations in the multiple 2D US image data are clear, the registration of the 2D US image data and the 3D MR volume data can be successful. However, when the image contours and grayscale variations in the multiple 2D US image data are unclear, the registration usually cannot be performed, resulting in registration failure.

[0027] In this invention, the image quality index of multiple two-dimensional image data is calculated by the image quality index calculation unit 13.

[0028] This explanation will use multiple two-dimensional image data sets as an example.

[0029] The image quality index is set based on the gradient information (i.e., changes in edge contours) and grayscale information of the 2D US image data. A larger image quality index is set for 2D US image data with large gradients (i.e., large changes in edge contours) and / or large grayscale differences in regions of the image, and vice versa.

[0030] The image quality index calculation unit 13 calculates the image quality index for each of multiple real-time two-dimensional US image data of a portion of the liver obtained by the probe.

[0031] The candidate two-dimensional image data determination unit 14 calculates the overall image quality index of a specified number of two-dimensional image data based on the image quality index of multiple two-dimensional image data. When the overall image quality index is above the image quality index standard value preset for the inspection area, the specified number of two-dimensional image data is determined as candidate two-dimensional image data.

[0032] Specifically, when the two-dimensional image data is two-dimensional US image data, the candidate two-dimensional image data determination unit 14 calculates the overall image quality index of a specified number of two-dimensional US image data based on the image quality index of multiple two-dimensional US image data. The calculation method of the overall image quality index will be explained later.

[0033] Furthermore, when the calculated overall image quality index is above a pre-set standard value for the examined area, such as the liver, a predetermined number of two-dimensional US image data are selected as candidate two-dimensional US image data. For example, let the pre-set standard value for the liver be 0.8. Figure 2 As shown, the overall image quality index calculated in (a) is 1, and the overall image quality index calculated in (b) is 0.1. The overall image quality index of the specified number of 2D US image data in (a) is 1, which is above the preset image quality index standard value of 0.8. Therefore, the specified number of 2D US image data is determined as candidate 2D US image data. The overall image quality index of the specified number of 2D US image data in (b) is 0.1, which is below the preset image quality index standard value of 0.8. Therefore, the specified number of 2D US image data is not determined as candidate 2D US image data. In other words, the corresponding specified number of 2D US image data in (a) meets the registration standard with the previously obtained 3D volume data, such as 3D MR volume data, while the corresponding specified number of 2D US image data in (b) does not meet the above registration standard.

[0034] At this time, Figure 2 In image (a), "Image Quality Index: 1" is displayed in green, reminding the operator that the specified number of 2D US image data meets the registration standard, thus enabling the specified number of 2D US image data to be registered with the 3D MR volume data for further inspection. Figure 2 In image (b), "Image Quality Index: 0.1" is displayed in red or yellow, reminding the operator that the specified number of two-dimensional US images does not meet the registration criteria. The operator will then change the pressure position on the liver of the subject, i.e., the scanning position, or change the rotation angle at the current pressure position until two-dimensional US images that meet the registration criteria are found.

[0035] Image data registration unit 15 registers candidate two-dimensional image data or thin three-dimensional volume data reconstructed based on candidate two-dimensional image data with three-dimensional volume data.

[0036] For example, when two-dimensional US image data meets the registration criteria and is determined to be candidate two-dimensional US image data, the candidate two-dimensional US image data can be registered with three-dimensional volume data, such as three-dimensional MR volume data, using a known registration method between two-dimensional image data and three-dimensional volume data. Furthermore, thin three-dimensional US volume data, i.e., a thin sheet-like three-dimensional US volume data that is much smaller than the three-dimensional volume data of the whole liver, can be reconstructed based on the candidate two-dimensional US image data, and the thin three-dimensional US volume data can be registered with the three-dimensional volume data of the whole liver using a known registration method between three-dimensional volume data.

[0037] Below, based on Figure 3 A method for calculating the overall image quality index of a specified number of two-dimensional US image data from multiple two-dimensional US image datasets is described.

[0038] Figure 3 (a) shows multiple real-time 2D US images of a portion of the liver obtained by the probe. Six 2D US images are shown in this figure, but the number of 2D US images obtained by the probe is typically not six, but several hundred. For these several hundred 2D US images, the image quality index Q is calculated for each data point, and the result is... Figure 3 As shown in (b), the image quality index Q of five two-dimensional US image data sets is presented as a representative calculation result. The candidate two-dimensional image data decision unit 14 selects the benchmark two-dimensional US image data set with the largest image quality index Q from hundreds of two-dimensional US image data sets. Figure 3 In (b), the baseline two-dimensional US image data is the two-dimensional US image data with an image quality index Q of 0.9 located in the center.

[0039] Furthermore, the candidate two-dimensional image data determination unit 14 selects a predetermined number of two-dimensional US image data in such a way that the same number of two-dimensional US image data exist before and after the reference two-dimensional US image data. When this same number is set to n (n is a natural number), the predetermined number is (2n+1) images, so here the predetermined number is set to an odd number of 3 or more. The multiple two-dimensional US image data obtained by the probe are multiple data arranged along a certain axis, so in the selected predetermined number of two-dimensional US image data, the reference two-dimensional US image data is located in the central position. That is, when 3 two-dimensional US image data are selected, the reference two-dimensional US image data is located in the second position, and when 7 two-dimensional US image data are selected, the reference two-dimensional US image data is located in the fourth position. Moreover, this predetermined number of two-dimensional US image data are arranged (exist) adjacent to each other in hundreds of two-dimensional US image data.

[0040] The candidate two-dimensional image data determination unit 14 is configured to assign a first weight to the reference two-dimensional US image data in a specified number of two-dimensional US image data, and assign a weight less than the first weight to other two-dimensional US image data in such a way that the farther away from the reference two-dimensional US image data, the smaller the weight is assigned. The image quality index of each two-dimensional US image data is calculated as the product of its weight, and the value obtained by adding the products of the specified number of two-dimensional US image data is used as the overall image quality index.

[0041] Figure 3 (c) shows the case where weights are assigned to a specified number of two-dimensional US image data selected. Among the selected number of 2D US image data, the baseline 2D US image data with an image quality index Q of 0.9 is assigned a first weight, i.e., the largest weight of 0.5 (i.e., W: 0.5). Other 2D US image data besides the baseline 2D US image data are assigned a weight less than the first weight, i.e., 0.5. Moreover, the weights are assigned in a manner that the farther away from the baseline 2D US image data, the smaller the weight. Specifically, for the baseline 2D US image data with an image quality index Q of 0.9, only the 2D US image data with an image quality index Q of 0.7 in front of it is assigned a weight of 0.25. For the baseline 2D US image data with an image quality index Q of 0.9, only the 2D US image data with an image quality index Q of 0.8 behind it is assigned a weight of 0.25, and so on. For the 2D US image data with an image quality index Q of 0.3 located in front of the baseline 2D US image data, a weight of 0.02 is assigned. For the 2D US image data with an image quality index Q of 0.1 located in back of the baseline 2D US image data, a weight of 0.02 is assigned. Figure 3 The weights assigned to the specified number of two-dimensional US image data in (c) are shown as examples only and could be other values.

[0042] Figure 3 (d) shows the formula for calculating the overall image quality index. The candidate two-dimensional image data determination unit 14 calculates the product of the image quality index Q of each two-dimensional US image data and its weight W, i.e., calculates 0.9*0.5, 0.8*0.25, etc., and, as shown... Figure 3 As shown in the formula for (d), the value obtained by adding the products of a specified number of two-dimensional US image data, such as 0.9*0.5, 0.8*0.25, etc., is used as the overall image quality index Qvol.

[0043] As mentioned above, according to Figure 3A method for calculating the overall image quality index of a predetermined number of two-dimensional US image data sets from multiple two-dimensional US image datasets has been described. However, this overall image quality index can also be calculated using other methods. For example, the candidate two-dimensional image data determination unit 14 is configured to use the average of the image quality indices of the predetermined number of two-dimensional US image data sets selected in the manner described above as the overall image quality index. This allows for the calculation of the overall image quality index using a simpler method than described above, reducing the amount of data processed by the system and accelerating the processing speed.

[0044] Next, use Figure 4 The registration method between multimodal images is explained.

[0045] In step S101, three-dimensional volume data of the examination site of the subject is obtained. The examination site is, for example, the liver. The three-dimensional volume data is, for example, three-dimensional US volume data, three-dimensional CT volume data, or three-dimensional MR volume data.

[0046] In step S102, the probe scans the area to obtain multiple two-dimensional US image data of a portion of the area to be inspected.

[0047] In step S103, for the multiple two-dimensional US image data obtained in step S102, the image quality index of each image data is calculated.

[0048] In step S104, the two-dimensional US image data with the highest image quality index is selected from multiple two-dimensional US image data as the reference two-dimensional US image data.

[0049] In step S105, the same number of two-dimensional US image data are selected from the front and back, centered on the reference two-dimensional US image data, and the selected two-dimensional US image data and the reference two-dimensional US image data are used as the specified number of two-dimensional US image data.

[0050] In step S106, the overall image quality index of a specified number of two-dimensional US image data is calculated, which can be done using... Figure 3 The calculation method shown involves assigning a first weight to the baseline 2D US image data, and assigning weights smaller than the first weight to other 2D US image data, with the weight decreasing as the data is farther from the baseline. The image quality index of each 2D US image data is calculated as a product of its weight, and the products of a specified number of 2D US image data are summed to obtain the overall image quality index. Alternatively, the average of the image quality indices of a specified number of 2D US image data can be calculated and used as the overall image quality index.

[0051] In step S107, the overall image quality index calculated in step S106 is compared with the standard image quality index value set for the inspected area. If the overall image quality index is above the standard value, the process proceeds to step S108, where a specified number of two-dimensional US image data are selected as candidate two-dimensional US image data. If the overall image quality index is not above the standard value, the process returns to step S102, where the operator changes the probe's scanning position relative to the inspected area or changes the probe's scanning direction relative to the inspected area, and then acquires multiple two-dimensional US image data of a portion of the inspected area again.

[0052] In step S109, the candidate two-dimensional US image data is registered with the three-dimensional volume data of the inspection area obtained in step S101. This registration can be performed using known registration methods for two-dimensional image data and three-dimensional volume data.

[0053] The above registration method can be, for example, manually setting the conversion parameters between two-dimensional US image data and three-dimensional volume data, and finding the two-dimensional cross-sectional image corresponding to the candidate two-dimensional US image data in the three-dimensional volume data through the affine transformation matrix, thereby achieving the registration of the candidate two-dimensional US image data and the three-dimensional volume data of the examination site.

[0054] In the above embodiment, the liver was used as an example for the examination site, but the examination site can also be other organs such as the heart and lungs. It is necessary to pre-set the image quality index standard value of the two-dimensional image data for each organ.

[0055] In the above embodiments, when the acquired two-dimensional image data is not suitable for registration, the operator is alerted by displaying the image quality index value in red or yellow. However, the operator can also be alerted by voice or other means.

[0056] According to this embodiment, when scanning the examination area of ​​the subject to obtain two-dimensional image data, two-dimensional image data suitable for registration with the pre-obtained three-dimensional volume data can be automatically obtained, and good and stable registration can be performed in real time to avoid registration failure.

[0057] Furthermore, when the acquired two-dimensional image data is not suitable for registration, the operator can be informed by displaying images, thereby allowing the operator to change the scanning conditions to reacquire the two-dimensional image data, which can shorten the inspection time and improve the operator's work efficiency.

[0058] Implementation Method 2

[0059] Next, use Figure 5 The image processing method of the second embodiment of the present invention will be described.

[0060] The structure of the image processing apparatus in the second embodiment is the same as that in the first embodiment, and will not be described again here.

[0061] Moreover, such as Figure 5 As shown, steps S201 to S208 of the image processing method of the second embodiment are the same as steps S101 to S108 of the image processing method of the first embodiment, and therefore will not be described again.

[0062] like Figure 5 As shown, in step S209, thin three-dimensional volume data, i.e. thin three-dimensional US volume data, is reconstructed based on the candidate two-dimensional US image data determined in step S208. This reconstruction is performed using a known method for reconstructing three-dimensional US volume data from two-dimensional US image data.

[0063] In step S210, the thin three-dimensional US volume data is registered with the three-dimensional volume data of the inspection area obtained in step S201. This registration can be performed using known registration methods between three-dimensional volume data.

[0064] According to this embodiment, when scanning the examination area of ​​the subject to obtain two-dimensional image data, it is possible to automatically obtain two-dimensional image data with excellent image quality, and to reconstruct small three-dimensional volume data based on the two-dimensional image data with excellent image quality. It is possible to achieve real-time and stable registration between the small three-dimensional volume data and the original three-dimensional volume data, that is, between the three-dimensional volume data, and to avoid registration failure.

[0065] As described above, although several embodiments of the present invention have been explained, these embodiments are shown as examples and are not intended to limit the scope of the invention. These embodiments can be implemented in various other ways, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and their variations are included in the scope and spirit of the invention, and are included in the invention described in the technical solution and its equivalents.

Claims

1. An image processing apparatus, characterized in that, have: The three-dimensional volume data acquisition unit acquires the three-dimensional volume data of the examination area of ​​the subject. The two-dimensional image data acquisition unit acquires multiple two-dimensional image data of a portion of the above-mentioned inspection area; The image quality index calculation unit calculates the image quality index of the above multiple two-dimensional image data; The candidate two-dimensional image data determination unit calculates the overall image quality index of a predetermined number of two-dimensional image data based on the image quality index of the plurality of two-dimensional image data. When the overall image quality index is above a pre-set image quality index standard value for the inspection area, the predetermined number of two-dimensional image data is determined as candidate two-dimensional image data. The image data registration unit registers the aforementioned candidate two-dimensional image data or the thin three-dimensional volume data reconstructed based on the aforementioned candidate two-dimensional image data with the aforementioned three-dimensional volume data. The aforementioned candidate two-dimensional image data determination unit selects the benchmark two-dimensional image data with the highest image quality index from the aforementioned plurality of two-dimensional image data, and selects the aforementioned predetermined number of two-dimensional image data in such a way that the same number of two-dimensional image data exist before and after the benchmark two-dimensional image data. The aforementioned candidate two-dimensional image data determination unit is configured to assign a first weight to the aforementioned reference two-dimensional image data among the aforementioned specified number of two-dimensional image data, and assign a weight less than the first weight to other two-dimensional image data in such a way that the farther away from the aforementioned reference two-dimensional image data, the smaller the weight is assigned, calculate the product of the image quality index of each two-dimensional image data and its weight, and add up the values ​​of the aforementioned specified number of two-dimensional image data to obtain the overall image quality index.

2. An image processing apparatus, characterized in that, have: The three-dimensional volume data acquisition unit acquires the three-dimensional volume data of the examination area of ​​the subject. The two-dimensional image data acquisition unit acquires multiple two-dimensional image data of a portion of the above-mentioned inspection area; The image quality index calculation unit calculates the image quality index of the above multiple two-dimensional image data; The candidate two-dimensional image data determination unit calculates the overall image quality index of a predetermined number of two-dimensional image data based on the image quality index of the plurality of two-dimensional image data. When the overall image quality index is above a pre-set image quality index standard value for the inspection area, the predetermined number of two-dimensional image data is determined as candidate two-dimensional image data. The image data registration unit registers the aforementioned candidate two-dimensional image data or the thin three-dimensional volume data reconstructed based on the aforementioned candidate two-dimensional image data with the aforementioned three-dimensional volume data. The aforementioned candidate two-dimensional image data determination unit selects the benchmark two-dimensional image data with the highest image quality index from the aforementioned plurality of two-dimensional image data, and selects the aforementioned predetermined number of two-dimensional image data in such a way that the same number of two-dimensional image data exist before and after the benchmark two-dimensional image data. The aforementioned candidate two-dimensional image data determination unit is configured to use the average of the image quality indices of the specified number of two-dimensional image data as the overall image quality index.

3. The image processing apparatus as described in claim 1 or 2, characterized in that, The aforementioned three-dimensional volume data can be 3D US volume data, 3D CT volume data, or 3D MR volume data. The above two-dimensional image data is two-dimensional US image data.

4. The image processing apparatus as described in claim 1 or 2, characterized in that, The above-mentioned examination sites on the subject are the organs of the subject.

5. The image processing apparatus as described in claim 1 or 2, characterized in that, The aforementioned image quality index is set based on the gradient information and grayscale information of the two-dimensional image data.

6. An image processing method, characterized in that, have: The steps to obtain three-dimensional volumetric data of the examination site of the subject; The steps to obtain multiple two-dimensional image data of a portion of the above-mentioned examination area; The steps for calculating the image quality index of the above multiple two-dimensional image data; The steps include: calculating the overall image quality index of a predetermined number of two-dimensional image data based on the image quality index of the aforementioned multiple two-dimensional image data; and determining the predetermined number of two-dimensional image data as candidate two-dimensional image data when the overall image quality index is above a pre-set standard value for the inspected area. The step of registering the aforementioned candidate two-dimensional image data or the thin three-dimensional volume data reconstructed based on the aforementioned candidate two-dimensional image data with the aforementioned three-dimensional volume data. In the step of determining candidate two-dimensional image data, a reference two-dimensional image data with the highest image quality index is selected from the multiple two-dimensional image data. The specified number of two-dimensional image data is selected by ensuring that the same number of two-dimensional image data exist before and after the reference two-dimensional image data. In the step of determining candidate two-dimensional image data, among the specified number of two-dimensional image data, the reference two-dimensional image data is assigned a first weight, and other two-dimensional image data are assigned a weight less than the first weight in such a way that the farther away from the reference two-dimensional image data is, the smaller the weight is assigned. The product of the image quality index of each two-dimensional image data and its weight is calculated, and the value obtained by adding the products of the specified number of two-dimensional image data is used as the overall image quality index.

7. An image processing method, characterized in that, have: The steps to obtain three-dimensional volumetric data of the examination site of the subject; The steps to obtain multiple two-dimensional image data of a portion of the above-mentioned examination area; The steps for calculating the image quality index of the above multiple two-dimensional image data; The steps include: calculating the overall image quality index of a predetermined number of two-dimensional image data based on the image quality index of the aforementioned multiple two-dimensional image data; and determining the predetermined number of two-dimensional image data as candidate two-dimensional image data when the overall image quality index is above a pre-set standard value for the inspected area. The step of registering the aforementioned candidate two-dimensional image data or the thin three-dimensional volume data reconstructed based on the aforementioned candidate two-dimensional image data with the aforementioned three-dimensional volume data. In the step of determining candidate two-dimensional image data, a reference two-dimensional image data with the highest image quality index is selected from the multiple two-dimensional image data. The specified number of two-dimensional image data is selected by ensuring that the same number of two-dimensional image data exist before and after the reference two-dimensional image data. In the step of determining candidate two-dimensional image data, the average of the image quality indices of the specified number of two-dimensional image data is used as the overall image quality index.

8. The image processing method as described in claim 6 or 7, characterized in that, The aforementioned three-dimensional volume data can be 3D US volume data, 3D CT volume data, or 3D MR volume data. The above two-dimensional image data is two-dimensional US image data.

9. The image processing method as described in claim 6 or 7, characterized in that, The above-mentioned examination sites on the subject are the organs of the subject.

10. The image processing method as described in claim 6 or 7, characterized in that, The aforementioned image quality index is set based on the gradient information and grayscale information of the two-dimensional image data.

Images