Medical image processing method and medical image processing device

Abstract

The present application discloses a medical image processing method and a medical image processing device. The medical image processing method includes the following steps: first, receiving a test image of a subject; then, receiving a plurality of recorded images of the subject; then, comparing the test image with the plurality of recorded images to select at least one recorded image from the plurality of recorded images that is similar to the test image; then, adjusting at least one recorded image to generate a similar image; and finally, comparing the test image with the similar image to determine at least one different part.

Description

Technical Field

[0001] The present application relates to an image processing technology, and in particular to a medical image processing method and a medical image processing device. Background Art

[0002] With the advancement of technology, various medical imaging devices, such as ultrasound, magnetic resonance imaging (MRI), positron emission tomography (PET), computed tomography (CT), mammography, and X-rays, produce images. These images can be used to diagnose certain diseases and provide insights into medical research.

[0003] The storage and transmission of diagnostic images generally adhere to the Digital Imaging and Communications in Medicine (DICOM) international medical imaging standard, a standard protocol for medical image processing, storage, printing, and transmission. It includes file format definitions and network communication protocols. DICOM is a TCP / IP-based application protocol that connects various systems. Two medical devices that accept the DICOM format can receive and exchange images and patient data using DICOM-formatted files.

[0004] Through the DICOM image transfer protocol, users can readily access both current and historical image data from past examinations. Through proper observation and comparison, discrepancies can be identified and the patient's physiological condition determined. However, conventional techniques typically rely on experience to visually compare past and present images after obtaining relevant patient images. This diagnostic approach is often influenced by the practitioner's experience and can easily lead to misjudgment.

[0005] On the other hand, for example, the same patient may have changes in body shape, such as weight gain or weight loss, at different time points, or medical images produced by different equipment may have angle deviations, making it difficult for medical personnel to find the image they want to compare when reviewing the before and after images. It is even difficult to accurately and instantly compare the differences due to different perspectives.

[0006] In summary, a medical image processing method and a medical image processing device thereof are needed to address the deficiencies of conventional technologies, improve the efficiency of medical image comparison, and reduce misjudgment. Summary of the Invention

[0007] The present application provides a medical image processing method and a medical image processing device thereof, wherein an image input terminal receives a detection image from a detection device, selects a recorded image that is most similar to the detection image from past recorded images through an image processing mechanism, and processes it to generate a similar image that can be compared and superimposed with the detection image, and automatically generates comparison information, thereby resolving the inconvenience and proneness to misjudgment of users when performing visual comparisons in conventional technologies. Specifically, when generating similar images, the present application further provides a method for interpolating multiple recorded images to form similar images, eliminating the conventional time-consuming imaging process of full 3D modeling and subsequent slicing, thereby achieving the effect of improving computational speed.

[0008] In one embodiment, the present application provides a medical image processing method comprising the following steps: first, receiving a test image of a subject; then, receiving a plurality of recorded images of the subject; then, comparing the test image with the plurality of recorded images to select at least one recorded image from the plurality of recorded images that is similar to the test image; then, adjusting the at least one recorded image to generate a similar image; and finally, comparing the test image with the similar image to determine at least one different area.

[0009] In another embodiment, the present application provides a medical image processing device, comprising an image processing unit and a first image output unit. The image processing unit is electrically connected to a database storing a plurality of recorded images, and is used to receive a plurality of recorded images and a detection image from an image generating device, wherein the image processing unit detects a plurality of first distinguishing feature points on the detection image, and detects a plurality of second distinguishing feature points for each of the plurality of recorded images, and compares the plurality of first distinguishing feature points of the detection image with the second distinguishing feature points in the plurality of recorded images, respectively, to determine and select at least one recorded image that is similar to the detection image from the plurality of recorded images, wherein the image processing unit adjusts at least one recorded image to generate a similar image. The first image output unit is electrically connected to the image processing unit, and is used to output the similar image. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0011] Figure 1A This is a flow chart of an embodiment of the medical image processing method of the present application.

[0012] Figure 1B This is a flow chart of an embodiment of the present application for comparing a detected image with a recorded image.

[0013] Figure 2 FIG. 1 is a schematic diagram of an embodiment of a medical image processing device of the present application.

[0014] Figure 3A This is a schematic diagram of an embodiment of the detection image of the present application.

[0015] Figure 3B Schematic diagram of multiple recorded images of this application.

[0016] Figure 3C This is a schematic diagram of an embodiment of a recorded image before adjustment calculation is performed in the present application.

[0017] Figure 3D A schematic diagram of an embodiment of a similar image generated after adjustment calculation in this application.

[0018] Figure 3E A schematic diagram of an embodiment of comparing a similar image with a detection image.

[0019] Figure 4A FIG. 1 is a flow chart of another embodiment of the medical image processing method of the present application.

[0020] Figure 4B This is a flow chart of another embodiment of the present application for comparing a detected image with a recorded image.

[0021] Figure 4C FIG. 1 is a flow chart of an embodiment of generating similar images.

[0022] Figure 5A A schematic diagram of a detection image with multiple comparison areas in this application.

[0023] Figure 5B This is a schematic diagram of a recorded image with multiple comparison areas in the present application.

[0024] Figure 5C and Figure 5D Schematic diagrams of similar concentration distributions respectively.

[0025] Figure 6A and 6B Schematic diagrams of recorded images with high similar concentrations in different comparison areas.

[0026] Figure 6C and 6D A schematic diagram of an embodiment of the adjustment calculation to obtain similar images according to the present application.

[0027] Figures 6E to 6G This is a schematic diagram of another embodiment of the adjustment calculation to obtain similar images according to the present application.

[0028] Figure 6HThis is a schematic diagram of another embodiment of the adjustment calculation to obtain similar images in the present application.

[0029] Figure 7 FIG. 1 is a schematic diagram of a system architecture of the medical image processing device of the present application applied to medical image processing. DETAILED DESCRIPTION

[0030] Various exemplary embodiments will be described more fully below with reference to the accompanying drawings, some of which are shown in the accompanying drawings. However, the concepts of the present application may be embodied in many different forms and should not be construed as limited to the exemplary embodiments described herein. Rather, these exemplary embodiments are provided so that the present application will be detailed and complete and will fully convey the scope of the concepts of the present application to those skilled in the art. Like numbers always indicate like elements. The following will illustrate a medical image processing method and a medical image processing device thereof using a variety of embodiments in conjunction with drawings, however, the following embodiments are not intended to limit the present application.

[0031] First, refer to Figure 1 and Figure 2 1 is a flow chart of an embodiment of a medical image processing method of the present application; Figure 2 This is a schematic diagram of an embodiment of a medical image processing device for realizing the image capture means of the present application. In step 20 of the medical image processing method 2, first, the image generating device 4 receives the detection image of the subject to be tested. The image generating device 4 that outputs the detection image can be various medical detection equipment, such as: an ultrasound scanning device, a nuclear magnetic resonance scan (MRI), a positron emission tomography (PET), a computerized tomography (CT), a mammography and an X-ray device, etc. The detection image can be one or more static images, or a continuous dynamic detection image or one or more static images in the detection image, and there is no specific limitation. In one embodiment, after the image generating device 4 generates a detection image of the subject to be tested (for example, one of the body parts of the subject), one of the detection images is selected from the detection image. One of the detection images in the detection image is obtained by the image capture means for subsequent comparison. In another embodiment, the image capture means is Figure 2 The medical image processing device 3 shown in FIG. 1 is used to perform the detection. The medical image processing device 3 includes an image capture unit 30, an image processing unit 31, and a first image output unit 32. The image capture unit 30 is coupled to the image generating device 4 to receive the detection image output by the image generating device 4 and to capture the detection image from the detection image.

[0032] Then, in step 21, a plurality of recorded images of the subject to be tested are obtained from the database. These are past test record images stored in the database, which are mainly used for comparison with the test image. In this embodiment, the medical image processing device 3 is electrically connected to the cloud server 5 and the image processing unit 31 in the medical image storage and transmission system (Picture Archiving and Communication System, PACS) by a network connection unit 36. The cloud server 5 has a database, which stores all the test image records generated by the image generating device 4 for each subject during the test process; on the other hand, it also stores medical image records of the same subject tested at different time points. The medical image processing device 3 uses a specific communication protocol to obtain the plurality of recorded images from the database 50. In one embodiment, the communication protocol can be the Digital Imaging Communications Protocol for Medicine (DICOM).

[0033] Then, step 22 is performed to compare the detection image with the plurality of recorded images to select a recorded image that is close to the detection image from the plurality of recorded images. The comparison method of this step is described as follows: Figure 1B As shown, first, step 220 is performed to detect a plurality of first distinguishing feature points on the detection image 9. Please also refer to Figure 3A , which is a schematic diagram of an embodiment of a detection image of the present application. In the detection image 9, C0 represents a first distinguishing feature point, which can be automatically defined by the image processing unit 31 executing an algorithm on the detection image 9. In one embodiment, the aforementioned algorithm can be a scale-invariant feature transform (SIFT) algorithm, but is not limited to this. For example, a speed-up robust feature (SURF) algorithm can also be implemented. These algorithms are well known to those skilled in the art and will not be described in detail here.

[0034] After detecting a plurality of first distinguishing feature points C0 in the detection image 9, step 221 is then performed to detect a plurality of second distinguishing feature points for the plurality of recorded images stored in the database. In this embodiment, the medical image processing device 3 is electrically connected to the cloud server 5 and the image processing unit 31 in the medical image storage and transmission system (PACS) by the network connection unit 36. The cloud server 5 has a database that stores the detection image records generated by the image generating device 4 during the detection process. The medical image processing device 3 obtains the plurality of recorded images from the database 50 using a specific communication protocol. In one embodiment, the communication protocol can be the Digital Imaging Communications Protocol for Medicine (DICOM). Please refer to Figure 3B, which is a schematic diagram of multiple recorded images of the present application. In one embodiment, Figure 3B The plurality of recorded images 9a-9f shown represent images from a previous examination of the same subject. Each recorded image 9a-9f represents a different scan layer of a specific organ. Step 221 also utilizes an algorithm, in this embodiment a SIFT algorithm, to detect and generate a plurality of second distinguishing feature points C1-C6 on each recorded image 9a-9f, although this is not limiting.

[0035] Next, step 222 is performed to compare the plurality of first distinguishing feature points C0 of the detection image 9 with the second distinguishing feature points C1 to C6 in the plurality of recorded images 9a to 9f, so as to determine the similar image that is most similar to the detection image 9 from the plurality of recorded images 9a to 9f. In one embodiment of step 222, an algorithm, such as a SIFT or SURF algorithm, is used in this embodiment to calculate the detection image 9 and each of the recorded images 9a to 9f, that is, the second distinguishing feature points C1 to C6 of each of the recorded images 9a to 9f are matched with the first distinguishing feature point C0 of the detection image 9, and the similarity concentration of each recorded image 9a to 9f relative to the detection image is obtained by SIFT calculation. The higher the similarity concentration, the more matching first and second feature points there are. For example, in Figure 3B Among them, the record image 9e has the highest similarity density. Therefore, after the calculation in step 222, the record image 9e is regarded as the most similar record image.

[0036] Back to Figure 1A After step 22, the recorded image 9e that is closest to the detected image is obtained. Then, step 23 is performed to adjust the contour, orientation, or scale of the recorded image to produce a similar image. In one embodiment of this step, the adjustment process mainly adjusts the recorded image found by the aforementioned comparison to obtain a similar image with the same perspective, orientation, contour, or scale as the detected image. In one embodiment, the recorded image obtained in step 22 may have differences in perspective, orientation, contour, or scale with the detected image. For example, Figure 3C Therefore, through the adjustment process of step 23, the similar image can be corrected through the adjustment calculation. In one embodiment, Figure 3C and 3D As shown, Figure 3C After the recorded image is calculated in step 23, the following is obtained: Figure 3DThe image state shown. The calculation method of step 23 is mainly to generate a transposition matrix based on at least four similar second feature points CS1 to CS4 on the recorded image 9e obtained in step 22. This transposition matrix can adjust the image space (x, y, z) so that the recorded image 9e is transformed into a similar image 9se with the same viewing angle, outline or proportion as the detection image 9 through the transformation of the transposition matrix. Figure 3D Similar images 9se can be seen with Figure 3A The detection image 9 shown has a similar viewing angle. It should be noted that the adjustment algorithm is a conventional technique and will not be described in detail here.

[0037] Then, return to Figure 1A After obtaining the similar image that is closest to the detection image in step 23, the process proceeds to step 24 to compare the detection image with the similar image to determine at least one difference. In one embodiment of step 24, scales can be superimposed on the similar image and the detection image in the same proportion to facilitate user identification, such as Figure 3E As shown in Figures (a) and (b), the unit length has a specific number of pixels, for example: Figure 3E In another embodiment of step 24, the pixel size is 100 pixels, but is not limited thereto. Figure 3E As shown in FIG. (c), the similar image 9se is calculated with the detection image 9, and only the difference image 9x where the difference area 900 between the two images is highlighted is displayed for the user's reference. Alternatively, further steps may be performed, such as Figure 3E As shown in FIG. 5 (d), the difference image 9x and the detection image 9 are superimposed to generate a superimposed image 9', and the position of the difference area 900 from the similar image is framed on the detection image to highlight the different areas for user reference.

[0038] It should be noted that when comparing a test image with recorded images stored in the database, two scenarios are possible. The first involves comparing the similarity density of the entire test image and the recorded image, selecting the single recorded image with the highest similarity density to the test pattern. The second scenario involves comparing one or more recorded images to be most relevant to the test image. The following describes an embodiment of the second scenario.

[0039] like Figures 4A to 4C As shown, Figure 4A This is a flow chart of another embodiment of the medical image processing method of the present application. Figure 4B A flowchart of an embodiment of calculating and generating a plurality of similar recorded images is provided. Figure 4C FIG. 1 is a flow chart of an embodiment of generating similar images. Figure 4AThe medical image processing method 2a shown in FIG. 2a is similar to the aforementioned steps 20 and 21 in steps 20a to 21a and will not be described in detail. Figure 4B As shown, it further includes defining a comparison area in the detection image and the recorded image, and comparing the detection image with the plurality of recorded images to select at least one recorded image that is similar to the detection image from the plurality of recorded images. In one embodiment, step 22a can further include step 220a, Figure 5A ) and each recorded image 9a to 9f (as shown Figure 5B As shown in FIG. 1 , the image is divided into a plurality of regions. In this embodiment, the image is divided into four regions by the center of its length and width to form four quadrants. In the detection image 9, four first comparison regions 90 to 93 are defined. The first quadrant is the first comparison region 90, the second quadrant is the first comparison region 91, the third quadrant is the first comparison region 92, and the fourth quadrant is the first comparison region 93. The recorded images 9a to 9f are respectively defined as a plurality of second comparison regions 90a to 93a to 90f to 93f corresponding to the first comparison regions 90 to 93. The first quadrant is the second comparison region 90a to 90f, the second quadrant is the second comparison region 91a to 91f, the third quadrant is the second comparison region 92a to 92f, and the fourth quadrant is the second comparison region 93a to 93f.

[0040] Then, step 221a is performed to calculate the first distinguishing feature point C0 of each first comparison area 90-93 and the second distinguishing feature points C1-C6 of the corresponding second comparison areas 90a-93a to 90f-93f in each recorded image 9a-9f. In one embodiment of step 221a, a SIFT algorithm is used to calculate each first comparison area 90-93 of the detection image 9 and the corresponding second comparison area 90a-93a to 90f-93f of each recorded image 9a-9f. During the calculation process, the second distinguishing feature points C1-C6 of each second comparison area 90a-93a to 90f-93f of each recorded image 9a-9f and the first distinguishing feature point C0 of the corresponding first comparison area 90-93 are calculated to obtain the similarity concentration of each second comparison area 90a-93a to 90f-93f, which represents the degree of similarity between the second comparison area 90a-93a to 90f-93f of each recorded image 9a-9f and the corresponding first comparison area 90-93 in the detection image 9.

[0041] Next, step 222a is performed to determine at least one recorded image that is most similar to the detection image 9 based on the calculation results of the second comparison areas 90a-93a to 90f-93f in each recorded image 9a-9f. The results of step 222a can be in two forms. The first form is that after the calculation, the second comparison areas of the four quadrants with the highest density all appear in the same recorded image. The second form, which is the main feature introduced in this embodiment, is that the quadrants with the highest density are distributed in different recorded images. This will be described in detail below. Figure 4A As shown, after step 22a, step 23a is performed to adjust the contour, orientation or ratio of at least one recorded image to generate a similar image. Figure 4C As shown, in the adjustment process of step 23a, firstly, step 230a is used to determine whether all the second comparison areas with the highest similar concentrations appear in the same recorded image. Figure 5C As shown in the figure, each of the recorded images 9a-9f has a similar density value M (M 90a～90f ,M 91a～91f ,M 92a～92f ,M 93a～93f ), which are respectively composed of similar concentration values ​​corresponding to the four second comparison areas 90a-93a to 90f-93f, from Figure 5C The schematic diagram shows that the similarity concentration M(9, 8, 9, 8) of each second comparison area 90d~93d of the recorded image 9d is the highest. Therefore, step 231a will be performed to use the recorded image 9d with the highest similarity concentration in all comparison areas as the most similar recorded image. Then, the adjustment procedure of step 232a is performed, which can be, for example, the adjustment of the image perspective. It mainly adjusts the recorded image in step 231a that is closest to the detection image to obtain a similar image with the same perspective, orientation, outline or proportion as the detection image. In one embodiment, the recorded image after step 231a may have differences in perspective, orientation, outline or proportion with the detection image, for example Figure 3C Therefore, through the adjustment process of step 232a, the similar image can be corrected through the adjustment calculation to obtain the following Figure 3D The image status shown. The adjustment algorithm is a conventional technique and will not be described in detail here.

[0042] In another embodiment, if Figure 5DAs shown, it can be clearly seen that the similarity concentration M occurs on different recorded images. In this embodiment, the similarity concentrations of the second comparison areas 90d to 93d of the recorded image 9d are M(8, 7, 6, 9), respectively, wherein the second comparison areas 90d and 93d have higher similarity concentrations, and the second comparison areas 90e to 93e of the recorded image 9e have similarity concentrations M(7, 8, 9, 6), wherein the second comparison areas 91e and 92e have higher similarity concentrations. In this case, since there is no single recorded image that is most similar to the detection image 9, step 233a will be performed after step 231a, and the image with higher similarity concentration will be interpolated to generate a similar image. In this step, there are many ways to interpolate. For example, in one embodiment, Figure 5D The characteristic value (e.g., grayscale value, brightness value, or contrast value) of each pixel (x, y) in recorded image 9d is added to the characteristic value of the corresponding pixel (x, y) in recorded image 9e and divided by two to obtain the new grayscale value of the pixel (x, y). The new image obtained by calculating the corresponding pixels of the two recorded images is called the similar image.

[0043] In addition to the aforementioned method of determining similar images by adding the feature values ​​of the two recorded images and dividing by two, another calculation in step 233a can be performed by assigning a weight to each of the second comparison regions 90d-93d and 90e-93e in the recorded images 9d and 9e, with the second comparison regions having a greater number of matching second distinguishing feature points being assigned a higher weight. The pixel feature values ​​of the pixels within each of the second comparison regions 90d-93d and 90e-93e in the recorded images 9d and 9e are then calculated based on the weights to determine similar images. The above calculation method is described below.

[0044] like Figure 6A and 6B As shown, it is assumed that the two recorded images of the second comparison area with relatively high similar concentrations are Figure 5D Recorded images 9d and 9e are shown. In recorded image 9d, the similarity density in second comparison area 90d and second comparison area 93d, that is, the two second comparison areas 90d and 93d extending along the positive X-axis and the Y-axis, is higher than that in the corresponding second comparison areas 90e and 93e of recorded image 9e. Meanwhile, in recorded image 9e, the similarity density in second comparison area 91e and second comparison area 92e, that is, the two second comparison areas 91e and 92e extending along the negative X-axis and the Y-axis, is higher than that in the corresponding second comparison areas 91d and 92d of recorded image 9d.

[0045] In the above situation, if Figure 6B As shown, since the second comparison areas with high similarity concentrations are located in the right half of recorded image 9d and the left half of recorded image 9e, respectively, the calculation is adjusted by weighting the left and right halves. Therefore, the weight is determined by the distance x from the Y axis. Based on the above rule, the weight distribution of the second comparison areas in each quadrant is as follows:

[0046] 90d 91d 92d 93d w11=(W+x) / 2W w12=(Wx) / 2W w13=(Wx) / 2W w14=(W+x) / 2W 90e 91e 92e 93e w21=(Wx) / 2W w22=(W+x) / 2W w23=(W+x) / 2W w24=(Wx) / 2W

[0047] Where w11, w12, w13, and w14 represent the weights of the second comparison area 90d-93d in recorded image 9d, and w21, w22, w23, and w24 represent the weights of the second comparison area 90e-93e in recorded image 9e. W is half the image height of recorded images 9d and 9e along the X-axis, expressed in pixels. x represents the absolute distance from each pixel Pd and Pe on recorded images 9d and 9e to the Y-axis. Therefore, based on the above weights, Figure 6C and 6D As shown, Figure 6D Similar images of 9s are Figure 6C The recorded images 9d and 9e in the image are synthesized by weight calculation. Figure 6D For example, pixel 90s in region 90s of similar image 9s has a feature value, such as a grayscale value, calculated by adding a weight to the feature value G90d(x, y) of pixel 90d in second comparison region 90d corresponding to region 90s in recorded image 9d and the feature value G90e(x, y) of pixel 90e in second comparison region 90e corresponding to region 90s in recorded image 9e. The calculation equation is G90s(x, y) = w11*G90d(x, y) + w21*G90e(x, y). Similarly, the characteristic value G91s(x,y) of each pixel in the region 91s corresponding to the second comparison regions 91d and 91e on the new similar image 9s is expressed as G91s(x,y)=w12*G91d(x,y)+w22*G91e(x,y), and so on, G92s(x,y)=w13*G92d(x,y)+w23*G92e(x,y) and G93s(x,y)=w14*G93d(x,y)+w24*G93e(x,y). Through the above interpolation calculation, the grayscale of each pixel on the new similar image 9s can be obtained, thereby forming the following: Figure 6D The similar image shown is obtained by interpolation in the above-mentioned manner. The similar image 9s has the advantages of fast calculation and no need for full 3D modeling and subsequent slicing.

[0048] It should be noted that the above examples are calculation formulas for two second comparison areas on the Y axis of the positive X axis or the negative X axis respectively having high similar concentrations. In another embodiment, Figure 6E and 6F As shown in this embodiment, the second comparison areas with high similarity density are located in two second comparison areas 90d and 91d on the positive Y-axis and the positive and negative X-axis directions in recorded image 9d, as well as two comparison areas 92e and 93e on the negative Y-axis and the positive and negative X-axis directions in recorded image 9e. Since the second comparison areas with high similarity density are located in the upper and lower halves of recorded image 9d and 9e, respectively, the calculation is adjusted using the weight distribution of the upper and lower halves. Therefore, the weight is determined by the distance y from the X-axis. Based on the aforementioned rule, the weight distribution of the second comparison areas in each quadrant is as follows, and the weight relationship is as follows:

[0049] 90d 91d 92d 93d w11=(H+y) / 2H w12=(H+y) / 2H w13=(Hy) / 2H w14=(Hy) / 2H 90e 91e 92e 93e w21=(Hy) / 2H w22=(Hy) / 2H w23=(H+y) / 2H w24=(H+y) / 2H

[0050] Where w11, w12, w13, and w14 represent the weights of the second comparison area 90d-93d in recorded image 9d, and w21, w22, w23, and w24 represent the weights of the second comparison area 90e-93e in recorded image 9e. H is half the image width of recorded images 9d and 9e along the Y axis, expressed in pixels. y represents the absolute distance between each pixel Pd and Pe on recorded images 9d and 9e and the X axis. Therefore, based on the above weight distribution, Figure 6G As shown, the characteristic value, e.g., grayscale value, of each pixel in region 90s1 of the new similar image 9s1 corresponding to the second comparison regions 90d and 90e can be expressed as G90s1(x, y) = w11*G90d(x, y) + w21*G90e(x, y). Similarly, the characteristic value of each pixel in region 91s1 of the new similar image 9s1 corresponding to the second comparison regions 91d and 91e is G91s1(x, y) = w12*G91d(x, y) + w22*G91e(x, y), and so on. Through the above interpolation calculation, the similar image 9s1 can be obtained.

[0051] If a single recorded image has three similar concentrations, such as Figure 6H As shown, in this embodiment, the second comparison areas 90d to 92d of the recorded image 9d have a higher similarity density, while the second comparison area 93e of the recorded image 9e has a high similarity density. Therefore, their weight relationship is as follows:

[0052] 90d 91d 92d 93d w11=(L+r) / 2L w12=(L+r) / 2L w13=(L+r) / 2L w14=(Lr) / 2L 90e 91e 92e 93e w21=(Lr) / 2L w22=(Lr) / 2L w23=(Lr) / 2L w24=(L+r) / 2L

[0053] Where w11, w12, w13, and w14 represent the weights of the second comparison area 90d-93d in recorded image 9d, respectively, while w21, w22, w23, and w24 represent the weights of the second comparison area 90e-93e in recorded image 9e, respectively. L is half the sum of the half-width W and half-height H of recorded images 9d and 9e, or 0.5(H+W), expressed in pixels. r represents the absolute distance between each pixel Pd and Pe in recorded images 9d and 9e and the origin. Therefore, based on the above weight distribution, a new similar image can be obtained by interpolation. It should be noted that the method for determining weights is not limited to the aforementioned x, y, and r distances and W, H, and L. Those skilled in the art can define their own weighting rules as needed.

[0054] Back to Figure 4C After obtaining the similar image through step 233a, the viewing angle, contour, or scale adjustment process of step 234a is performed. In this step, the similar image obtained in step 233a is adjusted with the detection image to obtain a similar image with the same viewing angle, contour, or scale as the detection image. In one embodiment, the similar image obtained through step 233a may have differences in viewing angle, contour, or scale with the detection image, such as Figure 3C Therefore, through the viewing angle, contour or ratio adjustment process of step 234a, the similar image can be corrected through adjustment calculation to obtain the following Figure 3D The image state shown. The adjustment algorithm is a conventional technique and will not be described in detail here. Finally, step 24a is performed to compare the detection image with the similar image to determine whether at least one difference portion (such as Figure 3E The procedure of this step is the same as that of step 24 above and will not be described in detail here.

[0055] See also Figure 7As shown in FIG, this figure is a schematic diagram of the architecture of an embodiment of the medical image processing device of the present application. The medical image processing device 3 in this embodiment includes an image capture unit 30, an image processing unit 31, and a first image output unit 32. The image capture unit 30 in this embodiment is coupled to the image generating device 4 by an analog-to-digital conversion unit (ADC) 37. The image capture unit 30 receives the detection image output by the image generating device 4 and captures the detection image from the detection image. In this embodiment, the image capture unit 30 is an FPGA component for performing image-related calculations. The image generating device 4 can be an ultrasound scanning device, a magnetic resonance imaging (MRI), a positron emission tomography (PET), a computerized tomography (CT), a mammography device, or an X-ray device. The image generating device 4 in this embodiment includes a detection device 40, a computing host 41 electrically connected to the detection device 40, and a second display device 42 electrically connected to the computing host 41.

[0056] The image processing unit 31 is electrically connected to the image capture unit 30 to receive the detection image output by the image capture unit 30. The image processing unit 31 is further connected to the network 100 via the network interface 36, and is connected to the cloud server 5 through the network 100. The cloud server 5 has a database 50, which stores a plurality of recorded images. The aforementioned recorded images are image records left when the patient is tested through the image generating device 4. The medical image processing device 3 obtains a plurality of recorded images from the database 50 and transmits them to the image processing unit 31 for subsequent calculation processing. The image processing unit 31 can be a system chip (SOC) with computing processing capabilities. In this embodiment, the image processing unit 31 is used to perform the following operations: Figure 1A-1B 、 Figures 4A to 4C The calculation process is as described above and will not be repeated here. The image capture unit 30 and the image processing unit 31 are also coupled to a memory unit 38, such as a flash memory, dynamic random access memory, or a combination thereof. The first image output unit 32 is electrically connected to the image processing unit 31 and the first display device 39. The first image output unit 32 is configured to output the calculated image to the first display device 39 for display.

[0057] The medical image processing device 3 further includes an input interface unit 33 electrically connected to the image processing unit 31 and an input device 34, such as a keyboard, mouse, trackball, or touch panel. The input interface unit 33 receives input signals from the input device 34 to control the image output mode of the image processing unit 31. Furthermore, in another embodiment, the medical image processing device 3 further includes a second image output unit 35 electrically connected to the image capture unit 30 to transmit images back to the computing host 41 of the image generating device 4. The computing host 41 can output the transmitted images via a second display device 42. It should be noted that the images transmitted from the medical image processing device 3 back to the computing host 41 can be detection images, recorded images, detection images, similar images, or subtraction images or superimposed images obtained after a comparison process between similar images and detection images. This allows the image generating device 4 to display images processed by the medical image processing device 3, assisting the examiner in making an immediate judgment.

[0058] In summary, the present application is described in the embodiments of the present application. The detection image generated by the detection device is compared with the image in the detection image through the image comparison mechanism. The image in the detection image is then selected or interpolated from the past recorded image to obtain a similar image. The similar image can then be compared, superimposed or subtracted with the detection image to generate new image information for the user to use as a reference for medical judgment. The aforementioned automatic comparison method can solve the problem of inconvenience and easy misjudgment when the user uses visual comparison in the conventional technology. In addition, the embodiments of the present application further illustrate the calculation mechanism of the similar image obtained by interpolation calculation, which solves the calculation efficiency problem of obtaining similar images through full 3D modeling and then slicing, thereby achieving a fast calculation effect.

[0059] It should be noted that, in this document, the terms "comprise," "include," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or apparatus comprising a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus. In the absence of further limitations, the phrase "comprising a..." does not preclude the presence of other identical elements in the process, method, article, or apparatus comprising the elements.

[0060] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of this application, ordinary technicians in this field can also make many forms without departing from the purpose of this application and the scope of protection of the claims, all of which are within the protection of this application.

Claims

1. A medical image processing method, characterized in that: include: Receiving a detection image of a person to be tested; receiving a plurality of recorded images of the subject; Comparing the detection image with the plurality of recorded images further comprises: defining a plurality of different first comparison regions in the detection image, such that each of the first comparison regions includes a portion of the plurality of first distinguishing feature points; defining a plurality of second comparison areas, which are different from and correspond to the plurality of first comparison areas, in each recorded image, so that each of the second comparison areas includes a portion of the plurality of second distinguishing feature points; Calculating the first distinguishing feature point of each first comparison area with the second distinguishing feature point of the corresponding second comparison area in each recorded image to obtain a similarity concentration for each second comparison area, and determining whether all second comparison areas with the highest similarity concentration appear in the same recorded image; If not, interpolating the first recorded image and the second recorded image with higher similarity density to generate a first similar image; Adjusting and calculating the first similar image to obtain the same viewing angle, contour or proportion as the detection image to form a second similar image; and The detection image is compared with the second similar image to determine at least one different portion.

2. The medical image processing method according to claim 1, wherein: The interpolation calculation step of the first recorded image and the second recorded image further includes: defining a weight value for each second comparison region in the first recorded image and the second recorded image, wherein a second comparison region having a greater number of matching second distinguishing feature points has a greater weight value; The pixel features in each second comparison area of ​​the first recorded image and the second recorded image are calculated according to the weight value of the second comparison area to generate the first similarity image.

3. The medical image processing method according to claim 1, wherein: The step of comparing the detection image with the second similar image further includes subtracting or superimposing the second similar image and the detection image to generate a difference image.

4. A medical image processing device, characterized in that: include: An image processing unit is electrically connected to a database storing a plurality of recorded images, wherein the image processing unit receives the plurality of recorded images and a detection image from an image generating device, wherein the image processing unit compares the detection image with the plurality of recorded images, and is characterized in that it further includes: defining a plurality of different first comparison areas for the detection image, so that each of the first comparison areas includes a portion of the plurality of first distinguishing feature points, and defining a plurality of different second comparison areas for each recorded image that correspond to the plurality of first comparison areas, so that each of the second comparison areas includes a portion of the plurality of second distinguishing feature points, and Calculating the first distinguishing feature points of the first comparison area with the second distinguishing feature points of the corresponding second comparison area in each recorded image to obtain a similarity concentration for each second comparison area, determining whether all second comparison areas with the highest similarity concentration appear in the same recorded image, and if not, interpolating the first recorded image with the higher similarity concentration with the second recorded image to generate a first similar image, adjusting the first similar image to obtain the same viewing angle, contour, or proportion as the detection image to form a second similar image, and comparing the detection image with the second similar image to determine whether at least one different location is found; The first image output unit is electrically connected to the image processing unit and is used to output the second similar image.

5. The medical image processing device according to claim 4, wherein: The image processing unit subtracts or superimposes the detection image and the second similar image to generate a difference image.

Images