A nuclear magnetic resonance image automatic analysis system and method

Abstract

The present invention discloses an automatic analysis system and method for nuclear magnetic resonance images, including a server and a mobile terminal connected to each other; the mobile terminal includes an image acquisition module; the server includes an image recognition module, a database, and a processor, which outputs a normal signal, first disease data, or second disease data to the mobile terminal according to a first vertebral body p through a first method. The present invention uses the processor to determine the bending direction and degree of the cervical spine according to the first method, and then can determine whether the patient's cervical spine is normal, and upload it to the mobile terminal, so that the patient can check the condition in time, which can improve the efficiency of seeing a doctor. The doctor can also use the results generated by the system as a reference to diagnose the patient, reduce the occurrence of misdiagnosis, and reduce the doctor's energy and improve work efficiency.

Description

Technical Field

[0001] The present invention relates to the field of image analysis, and in particular to a system and method for automatically analyzing nuclear magnetic resonance images. Background Art

[0002] Nuclear Magnetic Resonance Imaging (NMRI), also known as spin imaging (MRI), uses the principle of nuclear magnetic resonance (NMR) to detect the different attenuation of the released energy in different structural environments inside the material. The emitted electromagnetic waves are detected by applying an external gradient magnetic field to determine the position and type of the atomic nuclei that make up the object, and based on this, a structural image of the object can be drawn.

[0003] Applying this technology to imaging the internal structure of the human body has created a revolutionary medical diagnostic tool. The application of rapidly changing gradient magnetic fields has greatly accelerated the speed of MRI imaging, making the application of this technology in clinical diagnosis and scientific research a reality, and has greatly promoted the rapid development of medicine, neurophysiology and cognitive neuroscience.

[0004] At present, most people bend their heads to play with their mobile phones or work, which can cause cervical deformation. It is necessary to use medical means such as magnetic resonance imaging to detect whether the cervical spine is deformed. Because the internal structure images generated by magnetic resonance imaging require professional doctors to diagnose, patients need to spend a lot of time queuing and waiting for doctors, resulting in low medical efficiency, patients cannot consult doctors in time, and treatment delays may occur. At the same time, it will also increase the workload of doctors and increase the degree of fatigue, which can easily lead to misdiagnosis.

[0005] Therefore, there is an urgent need for an automatic analysis system and method for nuclear magnetic resonance images that can save patients' time in seeing a doctor, improve patients' efficiency in seeing a doctor, reduce doctors' workload, and improve diagnostic accuracy. Summary of the invention

[0006] The technical problem to be solved by the present invention is to provide a nuclear magnetic resonance image automatic analysis system and method which can save patients' medical consultation time, improve patients' medical consultation efficiency, reduce doctors' workload, and improve diagnostic accuracy.

[0007] The present invention provides an automatic nuclear magnetic resonance image analysis system, comprising a server and a mobile terminal connected to each other;

[0008] The mobile terminal includes an image acquisition module, and the image acquisition module is capable of acquiring a nuclear magnetic resonance image of a patient;

[0009] The server includes

[0010] An image recognition module capable of recognizing each first vertebral body p of the nuclear magnetic resonance image;

[0011] A database, which is used to store the first disease data and the second disease data;

[0012] A processor outputs a normal signal, first disease data, or second disease data to a mobile terminal through a first method according to the first vertebra p.

[0013] The present invention provides a nuclear magnetic resonance image automatic analysis system, wherein the first method is:

[0014] The database stores a first closed figure, which is a closed figure formed by a line connecting the geometric centers of each two adjacent first vertebrae p from the first first vertebrae p to the seventh first vertebrae p in a normal cervical spine and a line a1 connecting the first first vertebrae p and the seventh first vertebrae p, wherein the line a1 connecting the first first vertebrae p and the seventh first vertebrae p is a vertical line;

[0015] The processor outputs a closed figure formed by a line connecting the geometric centers of each two adjacent first vertebrae p from the first first vertebrae p to the seventh first vertebrae p in the patient's nuclear magnetic resonance image and a line a2 connecting the first first vertebrae p and the seventh first vertebrae p as a second closed figure, and the processor outputs a line a2 connecting the first first vertebrae p and the seventh first vertebrae p in the patient's nuclear magnetic resonance image as a first line;

[0016] The processor is used to determine whether the first connecting line a2 is in a vertical state;

[0017] If yes, then directly output the second closed figure as a third closed figure;

[0018] If not, the processor rotates the first connecting line a2 to a vertical state and outputs the second closed figure as a third closed figure;

[0019] The processor scales the third closed figure so that its highest point and lowest point can coincide with the highest point and lowest point of the first closed figure, and determines whether the third closed figure and the first closed figure have overlapping areas;

[0020] If so, determining whether the coverage area exceeds a first preset threshold of the area of ​​the first closed figure;

[0021] If it exceeds, the processor outputs a normal signal to the mobile terminal;

[0022] If not, the processor outputs the second disease data to the mobile terminal;

[0023] If not, the processor outputs the first disease data to the mobile terminal.

[0024] The present invention provides a nuclear magnetic resonance image automatic analysis system, wherein the first method is:

[0025] The processor is used to obtain the geometric center a of the first first vertebra p, the geometric center b of the fourth first vertebra p, and the geometric center c of the seventh first vertebra p, form a first vector m with a line connecting the geometric center a of the first first vertebra p and the geometric center b of the fourth first vertebra p, form a second vector n with a line connecting the geometric center c of the seventh first vertebra p and the geometric center b of the fourth first vertebra p, and output the angle formed by the first vector m and the second vector n as a first included angle α;

[0026] The processor is used to determine whether the magnitude of the first angle α exceeds a second angle threshold;

[0027] If yes, the processor outputs the second disease data to the mobile terminal;

[0028] If not, the processor is used to determine whether the orientation of the first angle α is the same as the orientation of the patient's face;

[0029] If they are the same, the processor outputs the first disease data to the mobile terminal and outputs the first disease level, the second disease level, or the third disease level to the mobile terminal through a third method;

[0030] If otherwise, the processor outputs a normal signal to the mobile terminal, and outputs the first prevention level, the second prevention level, or the third prevention level to the mobile terminal according to a third method.

[0031] The present invention provides an automatic nuclear magnetic resonance image analysis system, wherein the mobile terminal is connected to the server in an electrical manner.

[0032] The present invention provides an automatic nuclear magnetic resonance image analysis system, wherein the mobile terminal is a computer or a mobile device.

[0033] The present invention provides an automatic analysis system for nuclear magnetic resonance images, wherein the first disease data is cervical vertebrae kyphosis.

[0034] The present invention provides an automatic analysis system for nuclear magnetic resonance images, wherein the second disease data is the vertical curvature of the cervical spine.

[0035] The present invention provides an automatic analysis system for nuclear magnetic resonance images, wherein the severity of the disease levels is ranked as follows: first disease level < second disease level < third disease level, and the severity of the prevention levels is ranked as follows: first prevention level < second prevention level < third prevention level. When the patient's disease level is the third disease level.

[0036] The present invention provides an analysis method of an automatic nuclear magnetic resonance image analysis system, comprising the following steps:

[0037] obtaining MRI images of the patient;

[0038] Identifying each first vertebral body p of the nuclear magnetic resonance image;

[0039] storing first disease data and second disease data;

[0040] A normal signal, first disease data, or second disease data is output to a mobile terminal through a first method.

[0041] The difference between the automatic nuclear magnetic resonance image analysis system and method of the present invention and the prior art is that the automatic nuclear magnetic resonance image analysis system and method of the present invention uses the processor to judge the bending direction and degree of the cervical spine according to the first method, and then can judge whether the patient's cervical spine is normal, and upload it to the mobile terminal, so that the patient can check the condition in time, which can improve the efficiency of medical treatment. The doctor can also use the results generated by the system as a reference to diagnose the patient, reduce the occurrence of misdiagnosis, and at the same time reduce the doctor's energy and improve work efficiency.

[0042] The following is a further description of a nuclear magnetic resonance image automatic analysis system and method of the present invention in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 It is a flow chart of a system and method for automatic analysis of nuclear magnetic resonance images;

[0044] Figure 2 A contour map of a nuclear magnetic resonance image of a nuclear magnetic resonance image automatic analysis system and method;

[0045] Figure 3 is a schematic diagram of a first state of a second closed image;

[0046] Figure 4 is a schematic diagram of a second state of a second closed image;

[0047] Figure 5 It is a schematic diagram of a third closed figure;

[0048] Figure 6 It is a schematic diagram of the coverage area;

[0049] Figure 7 is a schematic diagram showing that the first angle α is opposite to the patient's face direction;

[0050] Figure 8 is a schematic diagram showing that the first angle α is in the same direction as the patient's face;

[0051] Fig. 9 is a schematic diagram of the third vector q. DETAILED DESCRIPTION

[0052] like Figures 1 to 9 As shown, see Figure 1 , 2 , inventing an automatic analysis system for nuclear magnetic resonance images, comprising a server and a mobile terminal connected to each other;

[0053] The mobile terminal includes an image acquisition module, and the image acquisition module is capable of acquiring a nuclear magnetic resonance image of a patient;

[0054] The server includes

[0055] An image recognition module capable of recognizing each first vertebral body p of the nuclear magnetic resonance image;

[0056] A database, which is used to store the first disease data and the second disease data;

[0057] A processor outputs a normal signal, first disease data, or second disease data to a mobile terminal through a first method according to the first vertebra p.

[0058] The present invention uses the processor to determine the bending direction and degree of the cervical spine according to the first method, and then can determine whether the patient's cervical spine is normal, and upload it to the mobile terminal, so that the patient can check the condition in time, which can improve the efficiency of medical treatment. Doctors can also use the results generated by the system as a reference to diagnose patients, reduce the occurrence of misdiagnosis, and at the same time reduce the doctor's energy and improve work efficiency.

[0059] The first disease data is the kyphosis of the cervical spine, and the second disease data is the verticality of the cervical spine.

[0060] Among them, the image recognition module can obtain the contour of the patient's nuclear magnetic resonance image by comparing the grayscale difference of the patient's nuclear magnetic resonance image, compare the acquired contour of the nuclear magnetic resonance image with the contour of the vertebra stored in the database, and extract the vertebral contour of the patient in the nuclear magnetic resonance image. The above content is the prior art.

[0061] The mobile terminal is a computer or a mobile device, and the mobile device can be a mobile phone or a tablet computer.

[0062] The present invention can facilitate users and medical personnel to view results through the above content.

[0063] The mobile terminal is connected to the server by electrical connection.

[0064] The present invention can enable the mobile terminal and the server to transmit signals to each other through electrical connection.

[0065] Preferably, see Figure 3 , 4 , 5, 6, the first method is:

[0066] The database stores a first closed figure, which is a closed figure formed by a line connecting the geometric centers of each two adjacent first vertebrae p from the first first vertebrae p to the seventh first vertebrae p in a normal cervical spine and a line a1 connecting the first first vertebrae p and the seventh first vertebrae p, wherein the line a1 connecting the first first vertebrae p and the seventh first vertebrae p is a vertical line;

[0067] The processor outputs a closed figure formed by a line connecting the geometric centers of each two adjacent first vertebrae p from the first first vertebrae p to the seventh first vertebrae p in the patient's nuclear magnetic resonance image and a line a2 connecting the first first vertebrae p and the seventh first vertebrae p as a second closed figure, and the processor outputs a line a2 connecting the first first vertebrae p and the seventh first vertebrae p in the patient's nuclear magnetic resonance image as a first line;

[0068] The processor is used to determine whether the first connecting line a2 is in a vertical state;

[0069] If yes, then directly output the second closed figure as a third closed figure;

[0070] If not, the processor rotates the first connecting line a2 to a vertical state and outputs the second closed figure as a third closed figure;

[0071] The processor scales the third closed figure so that its highest point and lowest point can coincide with the highest point and lowest point of the first closed figure, and determines whether the third closed figure and the first closed figure have overlapping areas;

[0072] If so, determining whether the coverage area exceeds a first preset threshold of the area of ​​the first closed figure;

[0073] If it exceeds, the processor outputs a normal signal to the mobile terminal;

[0074] If not, the processor outputs the second disease data to the mobile terminal;

[0075] If not, the processor outputs the first disease data to the mobile terminal;

[0076] The present invention compares the closed figure formed by the lines connecting each vertebra of the patient's cervical spine with the closed figure formed by the normal human cervical spine stored in the database. If they are covered, it proves that the curvature direction of the patient's cervical spine is consistent with that of the normal cervical spine. If they are not covered, it proves that the curvature direction of the patient's cervical spine is inconsistent with that of the normal cervical spine. Therefore, it is judged whether the patient's cervical spine has a disease, and the data is uploaded to the mobile terminal in time so that the patient and the patient's family can understand the condition in time, saving time in queuing for medical treatment, so that the patient can receive symptomatic treatment in time without delaying the condition. At the same time, it also provides a reference for doctors to prevent misdiagnosis, and can also reduce the workload of doctors and improve work efficiency.

[0077] The first preset threshold is 20%-50%, preferably 25%.

[0078] Among them, the first one of the first vertebrae p is located at the highest position, and the seventh one of the first vertebrae p is located at the lowest position.

[0079] Among them, the state of the second closed figure after rotation is still that the first first vertebra p is at the highest position, and the seventh first vertebra p is at the lowest position.

[0080] The processor is capable of rotating the first connecting rod a2 to a vertical state with the highest point of the first connecting line a2 as the rotation center.

[0081] Among them, see Figure 6 , the coverage area is the sum of the shadow areas of two sections in different directions, and the area of ​​the shadow part on the right side of the first preset threshold.

[0082] Another variation of the present invention can be, see Figure 7 , 8 , 9, the first method is:

[0083] The processor is used to obtain the geometric center a of the first first vertebra p, the geometric center b of the fourth first vertebra p, and the geometric center c of the seventh first vertebra p, form a first vector m with a line connecting the geometric center a of the first first vertebra p and the geometric center b of the fourth first vertebra p, form a second vector n with a line connecting the geometric center c of the seventh first vertebra p and the geometric center b of the fourth first vertebra p, and output the angle formed by the first vector m and the second vector n as a first included angle α;

[0084] The processor is used to determine whether the magnitude of the first angle α exceeds a second angle threshold;

[0085] If so, the processor outputs the second disease data to the mobile terminal.

[0086] If not, the processor is used to determine whether the orientation of the first angle α is the same as the orientation of the patient's face;

[0087] If they are the same, the processor outputs the first disease data to the mobile terminal and outputs the first disease level, the second disease level, or the third disease level to the mobile terminal through a third method;

[0088] If otherwise, the processor outputs a normal signal to the mobile terminal, and outputs the first prevention level, the second prevention level, or the third prevention level to the mobile terminal according to a third method.

[0089] The present invention determines the direction and size of the first angle α through the processor to determine the patient's disease. If the direction of the first angle α is consistent with the patient's facial direction, it means that the patient's vertebra is at a normal curvature level. If the direction of the first angle α is opposite to the patient's facial direction, it means that the patient's vertebra is in a reverse arch state. If the angle of the first angle α is 180 degrees, it means that the patient's vertebra is in a vertical state, thereby outputting the disease data to the mobile terminal for the patient to understand, saving time in queuing for medical treatment, saving time for patients, and reducing pressure on doctors.

[0090] The second angle threshold is 170-180 degrees, preferably 175 degrees.

[0091] Preferably, see Figure 7 , 8 , 9, the third method is:

[0092] The database stores a first disease level, a second disease level, and a third disease level corresponding to the first disease data, and the database also stores a first prevention level, a second prevention level, and a third prevention level corresponding to the normal signal;

[0093] When the processor determines that the orientation of the first angle α is in the same direction as the patient's facial orientation, the processor forms a third vector q, a first obesity coefficient f, and a first obesity index w according to the following first formula based on the first vector m, the second vector n, and the line connecting the geometric center of the first first vertebra p and the geometric center of the seventh first vertebra p. The first obesity coefficient f and the first obesity index w are obtained by a second method.

[0094]

[0095] The processor outputs the first preset angle and the second preset angle by a fourth method;

[0096] If the first angle α is less than a first preset angle, the processor outputs a third disease level to the mobile terminal;

[0097] If the first angle α is greater than or equal to a first preset angle and less than a second preset angle, the processor outputs a second disease level to the mobile terminal;

[0098] If the first angle α is greater than or equal to a second preset angle, the processor outputs a first disease level to the mobile terminal;

[0099] When the processor determines that the orientation of the first angle α is opposite to the orientation of the patient's face;

[0100] If the first angle α is less than a first preset angle, the processor outputs a first prevention level to the mobile terminal;

[0101] If the first angle α is greater than or equal to a first preset angle and less than a second preset angle, the processor outputs a second prevention level to the mobile terminal;

[0102] If the first angle α is greater than or equal to a second preset angle, the processor outputs a third prevention level to the mobile terminal.

[0103] The present invention determines the patient's disease level by obtaining the angle value of the first angle α after the processor obtains the direction of the first angle α. When the processor outputs a normal signal, it also determines the curvature of the patient's vertebra according to the angle value of the first angle α, thereby outputting the first prevention level, the second prevention level, and the third prevention level, thereby informing the patient that he should pay attention to prevention. When the processor outputs the first disease data, it also determines the reverse curvature of the patient's vertebra according to the angle value of the first angle α, thereby outputting the first disease level, the second disease level, and the third disease level, thereby informing the patient of the disease level. The value of the first angle α is obtained according to the cosine theorem and the second method.

[0104] Wherein, the processor is capable of directly outputting the first preset angle and the second preset angle according to fixed values.

[0105] The units of the first vector m, the second vector n, and the third vector q are cm.

[0106] Among them, the first preset angle is 130-150 degrees, preferably 140 degrees, the second preset angle is 150-170 degrees, preferably 160 degrees, the angle less than the first preset angle is the first angle range, greater than or equal to the first preset angle and less than the second preset angle is the second angle range, and greater than or equal to the second preset angle is the third angle range.

[0107] First angle range [0，140) Third level of disease First level of prevention Second angle range [140，160) Second level of disease Second level of prevention The third angle range [160，∞) First disease level Third level of prevention

[0108] Among them, the severity of the disease levels is ranked as follows: first disease level < second disease level < third disease level; the severity of the prevention levels is ranked as follows: first prevention level < second prevention level < third prevention level. When the patient's disease level is the third disease level, it proves that the patient is currently in a severe kyphosis state and needs timely treatment. When the patient's disease level is the third prevention level, it proves that although the patient's cervical spine is in a normal bending direction, it is almost approaching a vertical state and more attention should be paid to maintenance.

[0109] For example, in the patient's MRI image, the orientation of the first angle α is the same as that of the face, the length of the first vector m is 100, the length of the second vector n is 80, and the length of the third vector is 165. Then, the first formula is used to calculate Where (-f) w The fourth method calculated 9, so the value of the first angle α is 133+9=142 degrees, which is within the second angle range. Therefore, it can be known that the patient's first disease data level is second-level data, that is, the patient suffers from cervical kyphosis, and the disease level is level two. In other words, the patient's current disease level is level two, but the final disease level must be output through the fourth method based on the patient's occupation.

[0110] Preferably, the second method is:

[0111] The database stores a first obesity coefficient f and a first obesity index w corresponding to a first BMI index value of the patient;

[0112] The mobile terminal includes a first input module, and the first input module can obtain the first height data and the first weight data of the patient;

[0113] The processor obtains a first BMI index value h of the patient according to the first height data l and the first weight data g and the following second formula;

[0114]

[0115] The processor outputs the first obesity coefficient f and the first obesity index w corresponding to the first BMI index value h into the first formula according to the first BMI index value h of the patient.

[0116] According to the present invention, patients with different weights have different degrees of curvature of the cervical spine when lying down, which will affect the first angle α in the detection result. For example, when a patient has a heavy weight, when lying down for an MRI, the back is thicker, but the pillow height is constant, such as 10 cm or 0 cm, which will make the angle of the cervical spine smaller, which can affect the first angle α. When a patient has a light weight, when lying down for an MRI, the back is thinner, but the pillow height is constant, such as 10 cm or 0 cm, which will make the angle of the cervical spine larger, which can also affect the first angle α. Therefore, the present invention can detect the first BMI index value h of the patient, thereby determining the patient's fatness or thinness, thereby selecting the corresponding first obesity coefficient f and first obesity index w, and bringing them into the first formula to finally obtain an accurate first angle α, thereby preventing doctors from making misdiagnoses and affecting the patient's condition.

[0117] Among them, the first BMI index is short for body mass index, which is an internationally commonly used standard to measure the fatness and health of the human body.

[0118] The first height data and the first weight data can be measured by a height and weight scale, the unit of the first height data is meter, and the unit of the first weight data is kilogram.

[0119] The first BMI index value h The first obesity coefficient f The first obesity index w (0，18.5] 5 1 (18.5，23.9] 0 0 (23.9，27.9] 3 2 (27.9，∞] 4 2

[0120] The reason why the first obesity coefficient f is smaller when the patient is thin is that when a thin patient lies down, the cervical vertebra angle is less affected.

[0121] For example, if the patient is 1.8 meters tall and weighs 85 kg, the patient's first BMI index h=26.23 is obtained according to the second formula, and the patient's first obesity coefficient f is 3 and the first obesity index w is 2 when the table is looked up.

[0122] Preferably, the fourth method is:

[0123] The first input module is capable of acquiring the first occupation of the patient;

[0124] The database stores occupations susceptible to disease risk and occupations not susceptible to disease risk corresponding to the first occupation;

[0125] The processor is used to determine whether the first occupation is an occupation not prone to disease risk;

[0126] If the patient is in an occupation with low risk of disease, determining whether the patient's nuclear magnetic resonance image is first disease data;

[0127] If it is the first disease data, then the first preset angle and the second preset angle are increased by the first variable;

[0128] The processor is used to determine whether the first occupation is an occupation prone to disease risk;

[0129] If the patient is in a high-risk occupation, determining whether the patient's nuclear magnetic resonance image is a normal signal;

[0130] If it is a normal signal, the first preset angle and the second preset angle are increased by a first variable.

[0131] The present invention divides the patient's occupation into a high-risk occupation and a low-risk occupation. If the patient's occupation is a low-risk occupation but cervical spondylosis is detected, the angle threshold is increased to enhance the disease level; if the patient's occupation is a high-risk occupation but cervical spondylosis is detected, the angle threshold is increased to reduce the prevention level.

[0132] Among them, the occupations prone to disease risks include: intellectuals, surgeons, teachers, office staff, financial personnel, computer office workers, designers, draftsmen, welders, instrument maintenance workers, hand embroidery workers and seamstresses.

[0133] Among them, the occupations with low risk of disease include: athletes, fitness coaches, couriers, and food delivery people.

[0134] Wherein, the first variable is 10-20 degrees, preferably 15 degrees.

[0135] For example, the patient's occupation is a courier, and he suffers from cervical kyphosis, and the first angle α is 142 degrees, and the disease level is the second disease level, then the first preset angle 140 is increased to 155, and the second preset angle 160 is increased to 175, so the first angle range is (0, 155], so the level of the patient's first disease data is level three, that is, as a courier, which has a low risk of cervical disease, but he does have cervical spondylosis, the patient's disease level is increased according to the system settings, so as to attract the patient's attention and protect the cervical spine better.

[0136] See also Figure 1 , 2 The present invention provides an analysis method of an automatic nuclear magnetic resonance image analysis system, which comprises the following steps:

[0137] Step 1, obtaining a magnetic resonance imaging image of the patient;

[0138] Step 2, identifying each first vertebral body p of the nuclear magnetic resonance image;

[0139] Step 3, storing the first disease data and the second disease data;

[0140] Step 4: output the normal signal, the first disease data, or the second disease data to the mobile terminal through the first method.

[0141] The embodiments described above are merely descriptions of preferred implementation modes of the present invention and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary technicians in this field should all fall within the protection scope determined by the claims of the present invention.

Claims

1. A nuclear magnetic resonance image automatic analysis system, characterized in that: including interconnected servers and mobile terminals; The mobile terminal includes an image acquisition module, and the image acquisition module is capable of acquiring a nuclear magnetic resonance image of a patient; The server includes An image recognition module capable of recognizing each first vertebral body p in the nuclear magnetic resonance image; A database, which is used to store the first disease data and the second disease data; A processor, which outputs a normal signal, first disease data, or second disease data to a mobile terminal through a first method according to the first vertebra p; The first method is: The database stores a first closed figure, which is a closed figure formed by a line connecting the geometric centers of each two adjacent first vertebrae p from the first first vertebrae p to the seventh first vertebrae p in a normal cervical spine and a line a1 connecting the first first vertebrae p and the seventh first vertebrae p, wherein the line a1 connecting the first first vertebrae p and the seventh first vertebrae p is a vertical line; The processor outputs a closed figure formed by a line connecting the geometric centers of each two adjacent first vertebrae p from the first first vertebrae p to the seventh first vertebrae p in the patient's nuclear magnetic resonance image and a line a2 connecting the first first vertebrae p and the seventh first vertebrae p as a second closed figure, and the processor outputs a line a2 connecting the first first vertebrae p and the seventh first vertebrae p in the patient's nuclear magnetic resonance image as a first line; The processor is used to determine whether the first connecting line a2 is in a vertical state; If yes, then directly output the second closed figure as a third closed figure; If not, the processor rotates the first connecting line a2 to a vertical state and outputs the second closed figure as a third closed figure; The processor scales the third closed figure so that its highest point and lowest point can coincide with the highest point and lowest point of the first closed figure, and determines whether the third closed figure and the first closed figure have overlapping areas; If so, determining whether the coverage area exceeds a first preset threshold of the area of ​​the first closed figure; If it exceeds, the processor outputs a normal signal to the mobile terminal; If not, the processor outputs the second disease data to the mobile terminal; If not, the processor outputs the first disease data to the mobile terminal.

2. The automatic nuclear magnetic resonance image analysis system according to claim 1, characterized in that: The first method may also be: The processor is used to obtain the geometric center a of the first first vertebra p, the geometric center b of the fourth first vertebra p, and the geometric center c of the seventh first vertebra p, form a first vector m with a line connecting the geometric center a of the first first vertebra p and the geometric center b of the fourth first vertebra p, form a second vector n with a line connecting the geometric center c of the seventh first vertebra p and the geometric center b of the fourth first vertebra p, and output the angle formed by the first vector m and the second vector n as a first included angle α; The processor is used to determine whether the magnitude of the first angle α exceeds a second angle threshold; If yes, the processor outputs the second disease data to the mobile terminal; If not, the processor is used to determine whether the orientation of the first angle α is the same as the orientation of the patient's face; If they are the same, the processor outputs the first disease data to the mobile terminal and outputs the first disease level, the second disease level, or the third disease level to the mobile terminal through a third method; If otherwise, the processor outputs a normal signal to the mobile terminal, and outputs the first prevention level, the second prevention level, or the third prevention level to the mobile terminal according to a third method.

3. The automatic nuclear magnetic resonance image analysis system according to claim 2, characterized in that: The mobile terminal is connected to the server in an electrical manner.

4. The automatic nuclear magnetic resonance image analysis system according to claim 3, characterized in that: The mobile terminal is a computer or a mobile device.

5. The automatic nuclear magnetic resonance image analysis system according to claim 4, characterized in that: The first disease data is cervical kyphosis.

6. The automatic nuclear magnetic resonance image analysis system according to claim 5, characterized in that: The second disease data is the vertical curvature of the cervical spine.

7. The automatic nuclear magnetic resonance image analysis system and method according to claim 6, characterized in that: The severity of the disease levels is ranked as follows: first disease level < second disease level < third disease level; the severity of the prevention levels is ranked as follows: first prevention level < second prevention level < third prevention level. When the patient's disease level is the third disease level.

8. An analysis method according to any one of claims 1 to 7, characterized in that: The following steps are included obtaining MRI images of the patient; Identifying each first vertebral body p of the nuclear magnetic resonance image; storing first disease data and second disease data; Outputting a normal signal, first disease data, or second disease data to a mobile terminal by a first method; The first method is: The database stores a first closed figure, which is a closed figure formed by a line connecting the geometric centers of each two adjacent first vertebrae p from the first first vertebrae p to the seventh first vertebrae p in a normal cervical spine and a line a1 connecting the first first vertebrae p and the seventh first vertebrae p, wherein the line a1 connecting the first first vertebrae p and the seventh first vertebrae p is a vertical line; The processor outputs a closed figure formed by a line connecting the geometric centers of each two adjacent first vertebrae p from the first first vertebrae p to the seventh first vertebrae p in the patient's nuclear magnetic resonance image and a line a2 connecting the first first vertebrae p and the seventh first vertebrae p as a second closed figure, and the processor outputs a line a2 connecting the first first vertebrae p and the seventh first vertebrae p in the patient's nuclear magnetic resonance image as a first line; The processor is used to determine whether the first connecting line a2 is in a vertical state; If yes, then directly output the second closed figure as a third closed figure; If not, the processor rotates the first connecting line a2 to a vertical state and outputs the second closed figure as a third closed figure; The processor scales the third closed figure so that its highest point and lowest point can coincide with the highest point and lowest point of the first closed figure, and determines whether the third closed figure and the first closed figure have overlapping areas; If so, determining whether the coverage area exceeds a first preset threshold of the area of ​​the first closed figure; If it exceeds, the processor outputs a normal signal to the mobile terminal; If not, the processor outputs the second disease data to the mobile terminal; If not, the processor outputs the first disease data to the mobile terminal.

Images