Medical image analysis system and method

Abstract

The present invention provides an innovative medical imaging analysis system and method, addressing to improves the accuracy of boron quantitative analysis between the boundary of the human body and the air through the fusion of anatomical image and functional image. The system comprises the following components: a contour module, receiving anatomical image and functional image, said contour module further includes a contour identification unit and a contour definition unit, wherein the contour identification unit identifies geometric contour in the anatomical image, the contour definition unit defines distribution contour based on the signal intensity of boron in the functional image. A image fusion module, overlaying the anatomical mage and functional image, outputting the result of the intersection area of the geometric and distribution contours. A distribution analysis module, making the quantitative analysis of the boron distribution on the tissue surface according the intersection area of the geometric contour and distribution contour mentioned above.

Description

[Technical Field]

[0001] This case proposes a medical image analysis system and method, more specifically, a medical image analysis system and method that improves the accuracy of medical images in qualitative and quantitative analysis by fusing anatomical images and functional images. [Previous Technology]

[0002] Boron Neutron Capture Therapy (BNCT) is an innovative radiotherapy for cancer. It uses boron-containing drugs (such as those using amino acids as boron carriers) designed to specifically accumulate in the patient's tumor cells, while normal cells absorb less. When a patient is irradiated with an epithermal neutron beam, the tumor cells with accumulated boron-containing drugs absorb neutrons more easily than normal healthy cells. When the neutrons interact with the boron-containing drugs, a nuclear reaction occurs, producing lithium nuclei (7Li) and alpha particles. The generated alpha particles and lithium nuclei (7Li) atoms have an extremely short range (approximately 5-9 micrometers) to destroy tumor cells. This range is roughly equal to the diameter of a cell, allowing energy to be concentrated inside the tumor cells with minimal damage to adjacent healthy tissues. Therefore, BNCT treatment has advantages such as high selectivity for tumor sites, low toxicity, fewer treatment sessions than traditional radiotherapy, less burden on the body, and faster recovery.

[0003] For traditional radiation medicine, or the aforementioned boron neutron capture therapy, the use of medical imaging is crucial because it helps to accurately locate the patient's tumor, protect normal tissues during treatment, assess treatment effectiveness, and develop new medical tools in the medical engineering field. Medical imaging is currently mainly divided into anatomical imaging (such as computed tomography and magnetic resonance imaging, providing information about physical geometry) and functional imaging (such as positron emission tomography, gamma scintillation detectors, and single-photon emission computed tomography, providing information about the distribution of targeted drugs or other calibrated substances in the human body).

[0004] For example, in US Patent US20190030369A1, a treatment planning analysis system based on anatomical images is proposed for preliminary analysis before boron neutron capture therapy. It is used to distinguish mucosal regions, bone regions, air regions and corresponding body contours in medical images to assist medical personnel in identifying affected areas in body tissues. In this '369 case, the system components include an image acquisition unit, a body contour setting unit, an air hole region setting unit, and a mucosal region setting unit. Through the body contour setting unit, the patient's body contour region is set using the image acquired by the image acquisition unit. Next, the air hole region setting unit sets the air hole region inside the body contour region based on pixel quality. Finally, based on the aforementioned body contour region and air hole region, the mucosal region setting unit extends a portion of a first thickness from the inside of the body contour region toward the outside of the air hole region surface, analyzes the mucosal region with mucous membrane, and outputs the patient's body contour region, air hole region, and mucosal region as an anatomical image to achieve the purpose of locating the body contour, skeletal region, air region, air hole region, and mucosal region in the patient's body.

[0005] However, although anatomical images can indeed be used to distinguish physical geometric structures such as body contours and skeletal regions in conventional techniques, their resolution for soft tissues is still relatively poor. Furthermore, if the distribution information of targeted drugs or other calibrated substances (such as isotopes, or boron drugs in the case of BNCT) is required, functional imaging scans are still relied upon. However, functional imaging suffers from poor resolution for the aforementioned physical geometric structures. Therefore, medical personnel often have to refer to multiple anatomical and functional images simultaneously when analyzing lesions. However, anatomical and functional image scanning cannot be performed simultaneously. In the case of BNCT, image fusion technology must be used to analyze the distribution of boron drugs in the human body during medical image modeling, matching the information from the anatomical and functional images. However, such image fusion technology is not perfect. Due to the relationship between the instrument settings and the patient's posture during imaging, the coordinate alignment of the human body contour boundaries is prone to misalignment, and the differences in contours of larger organs are easily diluted. This phenomenon may cause the boron concentration at the boundary between the human body and the air to be severely underestimated in the BNCT model, such as in the skin, oral mucosa, and sinus mucosa, which in turn affects the accuracy of boron drug quantification and makes the efficacy of BNCT treatment less than expected. [Summary of the Invention]

[0006] Due to the aforementioned reasons, it is urgent to improve the above-mentioned deficiencies at the present time. Therefore, the main objective of this invention is to provide a novel medical image analysis system and method. By fusing anatomical and functional images through an image fusion method based on geometric and distribution contours, this invention not only solves the problem of coordinate misalignment during image fusion, leading to inaccuracies in the quantitative analysis of boron concentration in human body contours and air boundaries, but also allows for automated analysis. This improves upon the traditional method of image fusion of anatomical and functional images, which requires skilled medical engineers or physicians to manually determine alignment, thus avoiding human error, lack of objective and unified standards for judgment, and prolonged preoperative planning time. For a detailed technical solution of this invention, please refer to the following description.

[0007] To achieve the purpose of this invention, in a broad embodiment, the present invention provides a medical image analysis system comprising the following system components: a contour module, which receives at least one anatomical image and at least one functional image; the contour module further includes a contour recognition unit and a contour definition unit; the contour recognition unit identifies the geometric contour in the anatomical image by pixel values, and the contour definition unit defines the distribution contour by the light signal intensity of boron drug or other calibrated substances in the functional image; an image fusion module, coupled to the contour module, superimposes the anatomical image and the functional image respectively, and outputs the intersection area of ​​the geometric contour and the distribution contour; and a distribution analysis module, coupled to the image fusion module, quantifies the distribution of boron drug or other calibrated substances on the tissue surface by using the intersection area of ​​the geometric contour and the distribution contour.

[0008] To achieve the purpose of this case, the present invention also provides a novel medical image analysis method, comprising the following procedures: preparing at least one functional image and at least one anatomical image; performing contour recognition in the anatomical image to obtain a geometric contour; defining the contour in the functional image to obtain a distribution contour; performing image fusion based on the geometric contour and the distribution contour, and taking their intersection area; and using the intersection area to perform quantitative analysis of the concentration of boron drugs or other calibrated substances.

Implementation Method

[0009] This invention will be described in detail with reference to preferred embodiments and viewpoints so that the reader may fully understand how these embodiments are implemented. However, those skilled in the art will understand that this invention can also be implemented without these details. Furthermore, this invention can also be used and implemented by other specific embodiments, and the various details set forth in this specification can be applied based on different needs, and various modifications or changes can be made without departing from the spirit of this invention. Therefore, this invention will be described with reference to preferred embodiments and viewpoints. Such descriptions are for explaining the structure of this invention and are only for illustration and not for limiting the scope of the patent application of this invention. The terminology used in the following description will be interpreted in the broadest reasonable sense. Those skilled in the art can adjust the structure of this invention to meet the needs of actual industry according to the requirements of manufacturing or therapeutic applications, which is hereby stated in advance. Furthermore, it should be noted that, for ease of explanation, boron drugs or other calibrating substances refer to substances used to calibrate specific tissues. Since the preferred embodiments of the present invention mainly utilize BNCT, boron drugs will be used as a representative in the following text. However, those skilled in the art will recognize after reading this specification that the system and method of the present invention can also be applied to substances other than boron drugs, such as fluorescent substances or isotopes, which will also be described here.

[0010] The terms "about," "approximately," or "substantially" as used in this invention generally refer to within 20 percent of a given value or range, preferably within 10 percent, and more preferably within 5 percent. The values ​​given herein are approximate, meaning that unless explicitly stated otherwise, the meaning of the terms "about," "approximately," or "substantially" can be inferred. Furthermore, the terms "front and back," "left and right," and "up and down" as used in this invention can be understood in conjunction with the accompanying drawings. Unless otherwise stated in this invention, the meaning can be obtained from the context of the description after reading it. Finally, for the sake of simplicity, some conventional structures and components are shown in a simplified schematic manner in the drawings, and for ease of viewing, the dimensions of the components in the drawings are not drawn to scale. Furthermore, in the description of this invention, the method procedure numbers described are only the procedure numbers of a certain method flow unless specifically indicated in the description of this invention. They are used only for clarity and convenience in the description, and are not used to describe the sequential order of the method flow. This is also stated in advance.

[0011] Referring to Figure 1, the present invention provides a medical image analysis system 100, comprising the following system components: a contour module 120, which receives at least one anatomical image and at least one functional image; the contour module 120 further comprising a contour recognition unit 121 and a contour definition unit 122; the contour recognition unit 121 identifies the geometric contour in the anatomical image by the pixel values ​​of the anatomical image; the contour definition unit 122 defines the distribution contour by the light signal intensity of boron drug or other calibrated substances (e.g., fluorescent substances or isotopes) in the functional image; an image fusion module 140, coupled to the contour module 120, overlays the geometric contour in the anatomical image and the distribution contour in the functional image, and outputs the intersection area; and a distribution analysis module 150, coupled to the image fusion module 140, quantifies the distribution of boron drug or other calibrated substances on the tissue surface by the intersection area of ​​the geometric contour and the distribution contour. According to one embodiment of the present invention, in addition to defining the distribution contour through the boron drug in the functional image, the contour definition unit 122 can also define the distribution contour on the functional image based on the geometric contour after the anatomical image and the functional image are superimposed by the image fusion module 140. That is, the source of the distribution contour can be defined by (1) the light signal intensity of the boron drug or other calibrated substances (e.g., fluorescent substances or isotopes), or (2) the anatomical contour is used as the basis for definition after the anatomical image and the functional image are superimposed, or (1) and (2) are executed, so that medical personnel can choose according to the application needs. The execution order of (1) and (2) can be (1) first and then (2), (2) first and then (1), or (1) and (2) are executed simultaneously. The boron drug can be, but is not limited to, p-Boronophenylalanine (BPA), sodium borothiol metal cluster (BSH), boronized amino acid drugs, or combinations thereof. On the other hand, the calibrating substance may be, but is not limited to, 18FBPA, FDG, fluorescent protein (e.g., Green Fluorescent Protein, GFP), etc.

[0012] According to one embodiment of the present invention, the medical image analysis system 100 may be software, hardware, or a combination of software and hardware. If it is software, when its operating process is installed on a computer system, the computer system immediately becomes a tool for executing the present invention; if it is hardware, the medical image analysis system 100 includes a typical processing chip, memory, temporary memory, display device, network communication interface, operating system, and application program, which are interconnected in a generally known manner to perform calculations, temporary storage, display, and data transmission. Each system component may also be a combination of hardware and software. Based on the above, which belong to a generally known architecture, it will not be described in detail here.

[0013] Referring to Figures 1 and 4, according to an embodiment of the present invention, the medical image analysis system 100 further includes a distribution analysis module 150 coupled to an image fusion module 140, which quantifies the distribution of boron drugs or other calibrated substances on the tissue surface based on the intersection area of ​​the geometric contour and the distribution contour. According to one embodiment of the present invention, referring to Figure 4, the aforementioned processing method of the intersection area can be precisely illustrated by a schematic diagram of the image processing process 400. In the process of acquiring the distribution contour 410, the state of the boundary between the skin S (or any human tissue surface) and the non-interest region NROI (e.g., air) in the functional image can be defined by the contour definition unit 122 using contour recognition technology to define its distribution contour. The specific source of the functional image can be, but is not limited to, positron emission tomography, gamma scintillation detector, or single-photon emission computed tomography. In the process of acquiring the geometric contour 420, the state of the boundary between the skin S and the non-interest region NROI (e.g., air) in the anatomical image can be defined by the contour recognition unit 121 from the skin S or human tissue to at least one voxel outside the non-interest region NROI. The specific source of the anatomical image can be, but is not limited to, computed tomography (CT image), magnetic resonance imaging (MRI image), etc. Finally, the intersection region 430 is used to acquire the intersection of the functional image and the anatomical image.

[0014] Please refer to Figure 5. As mentioned above, according to one aspect of the present invention, taking positron emission tomography (PET) as an example of a functional imaging application scenario, since PET imaging identifies lesions by ingesting isotopes by the patient, its principle is based on the annihilation effect produced by the combination of positrons emitted during the β decay of the isotope with surrounding electrons, detected by a detector. After the reaction, the emitted photons locate the signal source (i.e., the lesion, such as a tumor). Because the isotope tracker accumulates at specific marked locations or areas with high blood flow and metabolism, PET images are mainly used to diagnose pathological or physiological functional changes, and provide limited information on the anatomical geometry of the human body. Furthermore, taking computed tomography (CT) as an example of anatomical imaging, since CT images primarily use X-rays as the light source to irradiate the patient, it utilizes the differences in X-ray attenuation of internal tissues. The signals measured after different irradiation angles through the body are combined for imaging. When the attenuation of substances in the irradiation path exhibits ideal linear attenuation, with no scattering or noise, tomographic images can be obtained through processing. However, these ideal conditions are difficult to achieve; the resolution for soft tissues is poor, and the high dose generated during the imaging process may increase the risk of harm to the patient. Especially for lesions in soft tissues with low thickness, or areas with low boron drug absorption (such as normal tissue), the present invention improves the accuracy of quantitative analysis of boron drug concentration in human body contours (e.g., skin S or human tissue) and air boundaries by processing the intersection region of the two. In one embodiment of the present invention, the intersection region is calculated as shown in Figure 4, by taking at least one voxel protruding from the skin S or human tissue outward from the non-interest area (NROI) in the geometric contour, that is, the average, maximum, and minimum values ​​of the volumes of the n intersection regions from the first intersection region 1, the second intersection region 2, ... the nth intersection region n, or a threshold set according to the application needs (e.g., the weight value of each region of interest). In other words, the intersection region can be processed as the average, maximum, and minimum values ​​of the volumes of the n intersection regions protruding outward from the geometric contour, or an appropriate threshold.

[0015] According to the present invention, the medical image analysis system 100 further includes a reading module 110 coupled to a contour module 120 for reading and transmitting functional images and anatomical images. The reading module 110 further includes an anatomical image unit 111 and a functional image unit 112, respectively used to execute the aforementioned procedures for reading functional images and anatomical images. According to embodiments of the present invention, the reading module 110 may be, but is not limited to, software or hardware, or a combination of hardware and software, and may be installed or deployed at a remote end (e.g., in the cloud) or a near end (local end).

[0016] According to the present invention, the image fusion module 140 further includes a coordinate processing unit 141. According to an embodiment of the present invention, the image fusion module 140 uses an algorithm for image fusion, which may be, but is not limited to, high-pass filtering, pyramid method, wavelet transforms, curvelet transforms, principal component analysis, HIS (hue, saturation, brightness) transform fusion, or any combination thereof. When the image fusion module 140 fuses functional images and anatomical images, since, as mentioned above, the scanning of functional images and anatomical images is separate, the angles and positions when capturing the affected area may be significantly different. Therefore, in one embodiment of the present invention, the coordinate processing unit 141 is used to align at least one functional image and anatomical image, specifically including angle transposition processing and coordinate translation processing of the functional image and anatomical image. In one embodiment, the functional image can be aligned with the anatomical image through the aforementioned angle transposition and coordinate translation processing; in another embodiment, the anatomical image can also be aligned with the functional image through angle transposition and coordinate translation processing; in yet another embodiment, the functional image and the anatomical image can be simultaneously processed by angle transposition and coordinate translation processing to synchronize to the required coordinate system, so that the coordinate contour and geometric contour can be more accurate in the process of image fusion.

[0017] Please refer to Figure 2. According to the present invention, the image fusion module 140 further includes a resolution correction unit 142 to improve the resolution of functional images and / or anatomical images. According to one aspect of the present invention, generally speaking, since functional images and anatomical images have different sources and different scanning and imaging times, their resolutions are usually different. Therefore, before the image fusion module 140 performs image fusion, the low-resolution image must be corrected to a high-resolution image to increase the accuracy of the image fusion process. The resolution correction unit 142 can select to correct functional images, anatomical images, or simultaneously correct functional images and anatomical images according to the application requirements. The correction condition depends on whether the resolution of each image meets the requirements of image fusion. Please refer to Figure 6. In one embodiment of the present invention, one method of the resolution correction unit 142 in correcting image resolution is to use an interpolation method. Specifically, the method can be to calculate the value of any point between two points through linear interpolation or non-linear interpolation. Using the linear interpolation method in Figure 6, we can assume that the four pixels surrounding the coordinate point F(x, y) in the image are F(x 0, y 0), F(x 0, y 1), F(x 1, y 0), and F(x 1, y 1). When calculating the values ​​of coordinate points Fu(x, y 0), Fm(x, y), and Fd(x, y 0), we can first calculate the values ​​of Fu(x, y 0) and Fm(x, y 1) using the following two formulas: Fu(x, y 0) = F(x 0, y 0) * (x 1-x) + (x 1, y 0) * (xx 0) -----(1); Fm(x, y 1) = F(x 0, y 1) * (x 1-x) + (x 1, y 1) * (xx 0) -----(2); Finally, based on the results of the above two equations, calculate the following value of F(x, y): F(x, y) = F(x, y0) * (y1-y) + F(x, y1) * (y-y0) -----(3). Thus, the resolution correction unit 142 can correct the functional image or anatomical image to the required resolution using the above method. It should be noted that those skilled in the art, after reading this specification, can also select different resolution correction methods or various parameters required for modification correction based on the needs of different applications, such as edge details, the amount of information to be retained in the image, color difference, edge smoothness, etc. The image resolution correction method in Figure 6 is only an example and is not intended to limit the scope of the claims of this case.

[0018] According to one embodiment of the present invention, the medical image analysis system 100 further includes an intelligent module 130, coupled to a contour module 120, an image fusion module 140, and a distribution analysis module 150. The intelligent module monitors the contour recognition unit 121 and the contour definition unit 122 for defining the geometric contour and the recognized contour, the image fusion of the functional image and the anatomical image, and the distribution analysis module 150 for evaluating the accuracy of the boron drug in the human body contour and air boundary, or the required quantitative analysis of a specific area of ​​the human body. When the intelligent module 130 determines that the accuracy of the boron drug meets the required detection standard in the training data, the intelligent module 130 controls the distribution analysis module 150 to output the detection result of the boron drug. However, when the intelligent module 130 determines that the accuracy of the boron drug does not meet the required detection standard in the training data, the intelligent module 130 controls the contour module 120, the image fusion module 140, and the distribution analysis module to adjust the operating mode or related parameters according to the accuracy of the boron drug, until the detection result meets the preset detection standard.

[0019] As described above, taking one embodiment of the present invention as an example, the adjustment of the above-mentioned operating mode, in terms of the contour definition unit 122 in the contour module 120, the intelligent module 130 can adjust the source of the distribution contour based on the detection result, that is, as mentioned above: (1) the light signal intensity definition of boron drug or other calibrating substances, or (2) the definition based on the anatomical contour after the superposition of anatomical images and functional images. Based on the foregoing description, it will not be repeated here; taking another embodiment of the present invention as an example, the adjustment of the above-mentioned operating mode, in terms of the image fusion module For 140, the algorithm used in the image fusion process can be adjusted (switching from one algorithm to another for trial), such as the aforementioned high-pass filtering method, pyramid method, wavelet transform method, or curvelet transform method, principal component analysis method, HIS transform fusion, and even the coordinate transformation and resolution correction methods in coordinate processing unit 141 and resolution correction unit 142; and taking another embodiment of the present invention as an example, the adjustment of the above-mentioned operating mode, for the distribution analysis module 150, can adjust the filtering parameters of the optical signal intensity, such as signal intensity, signal to noise ratio, full width at half maximum, signal integration method and related parameters, etc. The algorithms required for the operation of the intelligent module 130 can be, but are not limited to, supervised learning network, unsupervised learning network, hybrid learning network, associative learning network, optimization application network, or any combination thereof.

[0020] Please refer to Figure 3A. The present invention further proposes a medical image analysis method 200, which includes the following procedures: In procedure S1, at least one functional image and at least one anatomical image are prepared so that the contour module 120 can receive the aforementioned functional image and anatomical image; In procedure S2, the anatomical image is contour identified by the contour recognition unit 121 to obtain a geometric contour; In procedure S3, the functional image is contour defined by the contour definition unit 122 to obtain a distribution contour; In procedure S4, the image fusion module 140 performs image fusion based on the geometric contour and the distribution contour, and takes their intersection area. The image fusion algorithm may be, but is not limited to, high-pass filtering, pyramid method, wavelet transform method, or curvelet transform method, principal component analysis method, HIS transform fusion, or a combination of the above; In procedure S5, the distribution analysis module 150 uses the intersection area to perform quantitative analysis of the concentration of boron drugs or other calibrated substances. In one embodiment of the present invention, in addition to defining the distribution contour through the boron drug in the functional image, the contour definition unit 122 can also define the distribution contour on the functional image based on the geometric contour in the program S4 through the image fusion module 140 after superimposing the anatomical image and the functional image. That is, the source of the distribution contour can be defined by (1) the light signal intensity of the boron drug or other calibrated substances (such as fluorescent substances or isotopes), or (2) the anatomical contour is used as the basis for definition after superimposing the anatomical image and the functional image. The execution methods of (1) and (2) are the same as those in the previous paragraph, so they will not be repeated.

[0021] Please continue referring to Figure 3B. According to one embodiment of the present invention, the procedure S4 of the medical image analysis method 200 further includes the following subroutine: In procedure S4a, the coordinate processing unit 141 performs coordinate transformation on the coordinates of the functional image or anatomical image to synchronize the coordinate systems of the functional image and the anatomical image, so that the functional image and the anatomical image can be aligned with each other. Specifically, the coordinate transformation includes angle transposition processing and coordinate translation processing of the functional image and the anatomical image. In one embodiment, the functional image can be aligned with the anatomical image through the above-mentioned angle transposition processing and coordinate translation processing; in another embodiment, the anatomical image can also be aligned with the functional image through angle transposition processing and coordinate translation processing; in yet another embodiment, the functional image and the anatomical image can be simultaneously processed by angle transposition processing and coordinate translation processing to the desired coordinate system, so that the coordinate contour and geometric contour can be more accurate in the process of image fusion. Furthermore, the procedure S4 of the medical image analysis method 200 includes the following subroutine: In procedure S4c 1, the image fusion module 140 superimposes the coordinate-transformed functional images and / or anatomical images according to the positions of the geometric contours and distribution contours.

[0022] Please continue to refer to Figure 3B. According to an embodiment of the present invention, the procedure S4 of the medical image analysis method 200 further includes the following subroutines: In procedure S4b, the resolution correction unit 142 performs resolution correction on the functional image or anatomical image, wherein the resolution correction in procedure S4b is to correct the low-resolution image to a high-resolution image, and the specific method can be based on linear interpolation or nonlinear interpolation; and, in procedure S4c 2, the resolution-corrected functional image and / or anatomical image are superimposed according to the position of the geometric contour and the distribution contour.

[0023] Please refer to Figure 3A. According to one embodiment of the present invention, the medical image analysis method 200 further includes the following procedure: In procedure S6, the smart module 130 determines whether the quantitative analysis of boron drugs or calibrated substances on the human body or tissue surface by the distribution analysis module 150 meets the detection standards required for the accuracy of BNCT. If it meets the standards, procedure S9 is executed, and the distribution analysis module 150 outputs the detection results of boron drugs. If it does not meet the standards, procedure S7 is executed, and the smart module 130 adjusts the operating modes of the contour module 120, the image fusion module 140, and the distribution analysis module 150 in the medical image analysis method 200. In procedure S8, procedure S6 is executed again until the output detection results meet the detection standards required for the accuracy of BNCT. In the embodiments of the present invention, the algorithms required for the operation of the smart module 130 can be, but are not limited to, supervised learning networks, unsupervised learning networks, hybrid learning networks, associative learning networks, optimization learning networks, etc., or any combination thereof.

[0024] In summary, the medical image analysis system and method proposed in this invention improve upon conventional techniques where functional and anatomical images are discontinuous, resulting in some areas failing to fuse or exhibiting phase differences after image fusion, leading to poor accuracy in boron concentration calculation. Therefore, this invention utilizes the geometric contour definition of anatomical images and the identification of the distribution contours of functional images to correct this error, thereby obtaining a more accurate boron concentration. Thus, this invention not only possesses significant industrial applicability but also exhibits a novel structure and improved performance, hence this application. It should be noted that those skilled in the art are free to make various modifications after reading this invention, but all such modifications will not depart from the scope of protection sought in the appended claims. [Simplified Explanation of the Diagram]

[0025] The following detailed description of the invention and the schematic diagrams of the embodiments should enable the invention to be more fully understood; however, it should be understood that this is limited to the reference for understanding the application of the invention and not to limiting the invention to a particular embodiment.

[0026] Figure 1 illustrates the system architecture of the medical image analysis system. Figure 2 further illustrates the detailed architecture of the image fusion module. Figure 3A illustrates the procedure of the medical image analysis method. Figure 3B illustrates the detailed procedure of image fusion. Figure 4 further illustrates the method of taking the intersection area when geometric contours and distribution contours are superimposed. Figure 5 shows the actual shooting situation using anatomical images (illustrated as CT images) and functional images (illustrated as PET images). It is clear that the geometric contours of the anatomical images are clearer, but the characteristics of the marker substances are more clearly visible in the distribution contours. Figure 6 illustrates one method of the resolution correction unit in an embodiment of the present invention for correcting image resolution.

Claims

1. A medical image analysis system, comprising: a contour module for receiving an anatomical image and a functional image, the contour module further comprising: a contour recognition unit for recognizing a geometric contour from the anatomical image; a contour definition unit for recognizing a distribution contour from the functional image; an image fusion module coupled to the contour module for overlaying the anatomical image and the functional image, and outputting at least one intersection region of the geometric contour and the distribution contour; and a distribution analysis module for performing quantitative analysis from the at least one intersection region.

2. The medical image analysis system as claimed in claim 1, wherein the contour definition unit defines the distribution contour by means of an anatomical contour through the superimposed anatomical image and the functional image.

3. The medical image analysis system as described in claim 1, wherein the contour definition unit defines the distribution contour by means of the light signal intensity of a boron drug or other calibrating substance.

4. The medical image analysis system as claimed in claim 1, wherein the contour recognition unit defines the geometric contour by protruding at least one pixel from the skin or human tissue in a direction outside the non-interest area.

5. The medical image analysis system as described in claim 1, wherein the image fusion module performs image fusion using an algorithm selected from high-pass filtering, pyramid method, wavelet transform method, or curvelet transform method, principal component analysis method, HIS transform fusion, or any combination thereof.

6. The medical image analysis system as described in claim 1, wherein the image fusion module further includes a coordinate processing unit that aligns the functional image and the anatomical image by angle transposition processing and / or coordinate translation processing.

7. The medical image analysis system as described in claim 1, wherein the image fusion module further includes a resolution correction unit to improve the resolution of functional images and / or anatomical images.

8. The medical image analysis system as described in claim 7, wherein the resolution correction unit improves the resolution of the image through linear interpolation or nonlinear interpolation.

9. The medical image analysis system as claimed in claim 1 further comprises: a reading module coupled to the contour module for transmitting the functional image and the anatomical image to the contour module, wherein the reading module further comprises: an anatomical image unit for reading the anatomical image; and a functional image unit for reading the functional image.

10. The medical image analysis system as described in claim 1 further includes an intelligent module that uses artificial intelligence to monitor and adjust the operation modes of the contour module, the image fusion module, and the distributed analysis module, wherein the algorithm of the artificial intelligence is selected from supervised learning networks, unsupervised learning networks, hybrid learning networks, associative learning networks, optimization learning networks, or any combination thereof.

11. A medical image analysis method, comprising the following procedures: preparing a functional image and an anatomical image; identifying the anatomical image using a contour module to obtain a geometric contour; defining the functional image using the contour module to obtain a distribution contour; performing image fusion of the anatomical image and the functional image using an image fusion module, and taking at least one intersection region of the geometric contour and the distribution contour; and performing quantitative analysis based on the at least one intersection region using a distribution analysis module.

12. The medical image analysis method as described in claim 11, further comprising the contour module defining the distribution contour by means of an anatomical contour through the superimposed anatomical image and the functional image.

13. The medical image analysis method as described in claim 11, further comprising defining the distribution profile by means of the light signal intensity of a boron drug or other calibrating substance through the profile module.

14. The medical image analysis method as described in claim 11, wherein the image fusion algorithm is selected from high-pass filtering, pyramid method, wavelet transform method, or curvelet transform method, principal component analysis method, HIS transform fusion, or any combination thereof.

15. The medical image analysis method as described in claim 11, further comprising using a coordinate processing unit to perform coordinate transformation on the coordinates of functional images and / or anatomical images, through angle transposition processing, coordinate translation processing, or a combination of the above with a synchronized coordinate system.

16. The medical image analysis method as described in claim 11, further comprising performing resolution correction of the functional image and / or the anatomical image by means of a resolution correction unit.

17. The medical image analysis method as described in claim 11, further comprising using a smart module to determine whether the accuracy of the detection results of the quantitative analysis performed by the distribution analysis module meets the required detection standards, and if so, outputting the detection results.

18. The medical image analysis method as described in claim 11, further comprising using a smart module to determine whether the accuracy of the detection results of the quantitative analysis performed by the distribution analysis module meets the required detection standards; if not, the smart module adjusts the operation mode of the contour module, the image fusion module, the distribution analysis module, or the above components.

Images