Bone density measurement method, method and device for constructing bone density database

Abstract

An embodiment of the present invention provides a bone density measurement method and apparatus. The method includes: obtaining a CT image of a target bone; performing image segmentation on the CT image to obtain a cancellous bone image therein; determining a probability distribution of CT values corresponding to pixel values according to the pixel values in the cancellous bone image; fitting the probability distribution based on a preset multi-parameter non-normal probability distribution function to obtain a first probability density function; determining CT values corresponding to probability values within a preset range according to the first probability density function, and determining a first bone density value according to the CT values.

Description

Technical Field

[0001] The present invention relates to the fields of physiological data measurement and computer medical image processing, and particularly to a bone density measurement method, a method and device for constructing a bone density database. Background Art

[0002] In recent years, with the continuous extension of human lifespan, osteoporosis, which is highly prevalent among the elderly, has received increasing attention from the medical community and the public. Bone density is an important indicator of human bone mass, reflecting the degree of osteoporosis and serving as the main technical indicator for predicting the likelihood of fractures. This indicator plays an important role in the prevention, diagnosis, and treatment of osteoporosis and other diseases. Therefore, current bone density measurement is the most fundamental means for diagnosing diseases such as osteoporosis.

[0003] Currently, the main bone density measurement method used in clinical diagnosis is dual-energy X-ray absorptiometry (DXA). Its principle is to set an X-ray generator on one side of the human bone to be measured and an X-ray receiver on the other side. The X-ray generator is controlled to emit two beams of X-rays with different energies to irradiate the human bone to be measured, and the X-ray receiver measures the energy of the received X-rays. The algorithm built into the device calculates the bone density of the bone to be measured based on the energy attenuation of the two beams of X-rays. Additionally, bone density measurement methods also include ultrasonic quantitative measurement, quantitative CT measurement based on a human bone density equivalent phantom, etc. However, the above methods also have problems with limited application scenarios. For example, in dual-energy X-ray absorptiometry, the bone to be measured needs to be placed between the X-ray generator and the receiver, that is, there is a on-site limitation that requires the bone to be measured to be present during measurement. For quantitative CT measurement based on a human bone density equivalent phantom, there is an equipment limitation that requires a specific equivalent phantom during measurement.

[0004] Therefore, a new bone density measurement method is needed. Summary of the Invention

[0005] Embodiments of the present invention provide a bone density measurement method, a method and device for constructing a bone density database. This method cuts out a cancellous bone image from the CT image of the target bone, then determines the probability density function of the CT value corresponding to the cancellous bone image based on the gray value corresponding to the pixel value of each pixel in the cancellous bone image and a preset Johnson function, and finally determines the bone density value based on the CT value corresponding to the probability maximum value of the probability density function. Using this method, the bone density value corresponding to the bone to be measured can be determined only based on the CT image of the bone to be measured, reducing the process and equipment limitations in bone density measurement and broadening the application scenarios of bone density measurement.

[0006] The technical solution adopted by the present invention to solve the above technical problems is to provide a bone density measurement method on the one hand, including:

[0007] Obtain the CT image of the target bone;

[0008] Perform image segmentation on the CT image to obtain the cancellous bone image therein;

[0009] According to the pixel values in the cancellous bone image, determine the probability distribution of the CT values corresponding to the pixel values;

[0010] Based on a preset multi-parameter non-normal probability distribution function, fit the probability distribution to obtain the first probability density function;

[0011] According to the first probability density function, determine the CT values corresponding to the probability values within a preset range, and according to the CT values, determine the first bone density value.

[0012] According to a possible implementation manner, the target bone includes the first lumbar vertebra and the second lumbar vertebra.

[0013] According to a possible implementation manner, performing segmentation on the CT image includes:

[0014] Segment the CT image by a preset image segmentation method based on machine learning.

[0015] According to a possible implementation manner, the preset multi-parameter non-normal probability distribution function is one of the Johnson SU, Johnson SB, and Johnson SL functions.

[0016] According to a possible implementation manner, the probability values within the preset range include the probability maximum value of the first probability density function.

[0017] In a second aspect, a method for constructing a bone density database is provided, including:

[0018] Obtain the attribute data of multiple target bones;

[0019] Based on the method described in the first aspect, obtain the bone density values of the multiple target bones;

[0020] Associate and save the bone density values of the target bones and their attribute data in a preset database.

[0021] According to a possible implementation manner, the attribute data includes one or more of age, gender, region, and bone name.

[0022] According to a possible implementation manner, the method further includes:

[0023] According to the attribute data of the multiple target bones, divide the bone density values of the multiple target bones into several groups;

[0024] According to the bone density values included in each group, determine the standard bone density value and variance value corresponding to each group, and save the standard bone density value and variance value in a preset database.

[0025] In a third aspect, a bone density measurement device is provided, including:

[0026] A CT image acquisition unit configured to acquire a CT image of a target bone;

[0027] A cancellous bone image acquisition unit configured to segment the CT image to obtain the cancellous bone image therein;

[0028] A probability distribution determination unit configured to determine the probability distribution of the CT values corresponding to the pixel values according to the pixel values in the cancellous bone image;

[0029] A probability density function determination unit configured to fit the probability distribution based on a preset multi-parameter non-normal probability distribution function to obtain a first probability density function;

[0030] A bone density value determination unit configured to determine the CT values corresponding to the probability values within a preset range, and determine a first bone density value according to the CT values.

[0031] In a fourth aspect, a device for constructing a bone density database is provided, including:

[0032] A bone attribute acquisition unit configured to acquire the attribute data of multiple target bones;

[0033] A bone density determination unit configured to obtain the bone density values of the multiple target bones based on the method described in the first aspect;

[0034] A storage unit configured to associatively store the bone density values of the target bones and their attribute data in a preset database. Description of the Drawings

[0035] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for description in the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0036] Figure 1 It is a flowchart of a bone density measurement method provided by an embodiment of the present invention;

[0037] Figure 2 Schematic diagram of obtaining cancellous bone part by segmentation in the lumbar spine diagram provided by the embodiment of the present invention;

[0038] Figure 3 Schematic diagram of probability density function fitted based on Johnson SU function provided by the embodiment of the present invention;

[0039] Figure 4 Flowchart of a method for constructing a bone density database provided by the embodiment of the present invention;

[0040] Figure 5 Structure diagram of a bone density measurement device provided by the embodiment of the present invention;

[0041] Figure 6 Structure diagram of a device for constructing a bone density database provided by the embodiment of the present invention. Detailed implementation manners

[0042] To make the objectives, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some but not all of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the scope of protection of the present invention.

[0043] As mentioned above, in recent years, with the continuous extension of human lifespan, osteoporosis, which is highly prevalent among the elderly, has received increasing attention from the medical community and the public. Bone density is an important indicator of human bone mass, reflecting the degree of osteoporosis and being the main technical indicator for predicting the likelihood of fractures. Bone density measurement plays an important role in the prevention, diagnosis and treatment of osteoporosis. Currently, the commonly used clinical diagnostic bone density measurement scheme includes Dual-energy X-ray Absorptiometry (DXA) measurement method. Its principle is to set an X-ray generator on one side of the human bone to be measured and an X-ray receiver on the other side. Control the X-ray generator to emit two beams of X-rays with different energies to irradiate the human bone to be measured, and the X-ray receiver measures the energy of the received X-rays. The algorithm built into the device calculates the bone density of the bone to be measured based on the energy attenuation of the two beams of X-rays. The problem with this scheme is that the operation process has specific on-site operation requirements. For example, the bone to be measured needs to be on-site and placed between the X-ray generator and the receiver.

[0044] Common bone density measurement methods also include ultrasonic bone density measurement methods, CT measurement methods based on human bone density equivalent phantoms, etc. Among them, ultrasonic bone density measurement methods mainly use ultrasonic waves to measure bone density. The measurement accuracy of this method is poor, so it is usually only used for self-measurement of personal bone density. For the CT measurement method based on the human bone density equivalent phantom, the bone density equivalent phantom is placed in the oral cavity, and then the bone density is calculated through oral CT scanning. The problem with this method is that it requires the assistance of a specific device, the equivalent phantom, during the measurement, which limits the applicable application scenarios.

[0045] In the prior art, there is also an automatic osteoporosis parameter measurement method based on CT images. According to this method, a clinical database needs to be constructed first, that is, by performing CT examinations and DXA bone density scans on the sampled population, and the time interval between the CT examination and the DXA bone density scan is less than 1 month. Then, the lumbar vertebrae of the CT scan image of the object to be measured are segmented, and the bone density is calculated. This method decomposes the simultaneous measurement of X-rays and an equivalent phantom into asynchronous measurements of constructing a clinical database and clinical bone density measurement, so that no equivalent phantom needs to be placed during actual clinical measurement. Essentially, it reduces the measurement accuracy through the estimation of the database in exchange for simplifying the measurement process. The problem with this method is that in actual applications, it is difficult to construct a clinical database, and the database needs to be updated as the CT equipment ages, or rebuilt as the CT equipment is upgraded, so it is difficult to implement.

[0046] In summary, various existing bone density measurement methods often have restrictions in terms of on-site processes, required equipment, and post-maintenance, which limit the scope and effectiveness of their applications.

[0047] To solve the above technical problems, this specification proposes a bone density measurement method. Using this method, the bone density data of the human bone can be obtained only based on the CT scan image of the human bone, greatly reducing the restrictions on bone density measurement in terms of process or equipment. Thus, while ensuring the measurement effect, it is more convenient to measure bone density and solve the deficiencies of the prior art.

[0048] Figure 1 It is a flowchart of a bone density measurement method provided by an embodiment of the present invention. As Figure 1 shown, this method at least includes the following steps:

[0049] Step 11, obtain a CT image of the target bone.

[0050] A CT image is an image obtained by scanning a target bone with a CT device. CT (Computed Tomography) is electron computed tomography. Generally, in CT scanning, highly collimated X-ray beams, γ rays, ultrasonic waves, etc., together with highly sensitive detectors, can be used to perform cross-sectional scanning around the part of the human body to be measured to obtain a CT image of the part to be measured. In different embodiments, the CT image of the target bone can be obtained by different specific methods. For example, in one embodiment, the target bone can be scanned on-site with a CT device (such as a CT imaging device commonly used in ordinary hospitals) to obtain the CT image of the target bone. In another embodiment, it is also possible to receive a CT image of the target bone that has been pre-obtained at other locations through off-site scanning. It should be noted that this specification does not focus on the specific methods of how to obtain the CT image of the target bone, but mainly focuses on the processing process of the obtained CT image.

[0051] The target bone is the human bone corresponding to the obtained CT image. In different embodiments, bones of different specific parts of the human body can be obtained. In one embodiment, in order to measure the bone density data of a potentially diseased bone and provide a basis for diagnosing whether there is actually a disease in the potentially diseased bone, the target bone can be a potentially diseased bone. In another embodiment, a predetermined target bone can be used as the object for bone density detection in a routine physical examination. In a specific embodiment, the predetermined target bones can be the first lumbar vertebra and the second lumbar vertebra.

[0052] Step 12: Segment the CT image to obtain the cancellous bone image therein.

[0053] Generally speaking, bones are composed of two parts: cancellous bone and compact bone. Compacted bone is mainly distributed on the surface of long bone shafts and other types of bones, and cancellous bone is mainly distributed on the two ends of long bones, vertebral surfaces and ribs. Compacted bone is usually dense and hard in structure, with strong compression and torsion strength. Cancellous bone is often sea-like and has a loose structure. In this step, from the bone CT image, the image of cancellous bone is segmented. The reason is that the cancellous bone image segmented is mainly used for bone density determination in subsequent steps, and the determination of bone density is often used for the determination of human bone state, and then determines whether the human body suffers from a disease, such as whether it suffers from osteoporosis. However, after the human skeleton grows up, the change of bone density of its compact bone is very small, and it is also very small to be affected by the state of the human body, that is, even if the bone to be tested (for example, the target bone) has osteoporosis, the bone density of its cortical bone changes very little, so it is difficult to determine the current state of the bone according to the bone density of its cortical bone, whether there is osteoporosis. The bone density of cancellous bone is greatly affected by the condition of the human body. That is to say, the current condition of the bone to be tested can be better determined based on the bone density of the cancellous bone, and then it can be judged whether the human body suffers from osteoporosis.

[0054] In different embodiments, the CT image can be segmented based on different specific methods to obtain the cancellous bone image therein, which is not limited in this specification. For example, in one embodiment, the CT image can be segmented by a preset image segmentation method based on machine learning. Figure 2 A schematic diagram of segmenting and obtaining the cancellous bone part in a lumbar vertebra image provided by an embodiment of the present invention, such as Figure 2 As shown, after the image is cut, a cancellous bone image 1 wrapped by a cancellous bone edge 2 is obtained.

[0055] Step 13: Determine the probability distribution of the CT value corresponding to the pixel value according to the pixel value in the cancellous bone image.

[0056] In this step, the CT value corresponding to the pixel value can be determined according to the pixel value of each pixel in the cancellous bone image. According to one embodiment, the CT value corresponding to the pixel can be determined according to the gray value of each pixel in the cancellous bone image. Then, the distribution of the probability of occurrence of each CT value is determined according to the CT value corresponding to all pixels in the cancellous bone image. In one embodiment, the sum (or integral) of the probability of occurrence of each CT value after determination is 1.

[0057] Step 14: Fit the probability distribution based on a preset multi-parameter non-normal probability distribution function to obtain a first probability density function.

[0058] In this step, a multi-parameter non-normal probability distribution function, such as the Johnson distribution function, can be used to fit the probability distribution of the CT values obtained in step 13. Specifically, by adjusting the parameters of the Johnson distribution function, the y-dimensional values corresponding to the x-dimensional values (corresponding to the respective CT values) of the points on the Johnson distribution function curve can be made to approach the occurrence probabilities of the respective CT values. In different embodiments, different specific Johnson distribution functions can be employed. In one embodiment, the preset multi-parameter non-normal probability distribution function can be one of the unbounded Johnson distribution (Johnson SU), the bounded Johnson distribution (Johnson SB), and the semi-bounded Johnson distribution (Johnson SL) functions. In different embodiments, different methods for adjusting the parameters of the Johnson distribution function and for determining that the parameter adjustment is completed (i.e., the function fitting is completed) can be adopted, and this specification places no restrictions thereon.

[0059] After the above-mentioned fitting is completed, a first probability density function can be obtained, that is, the probability density function of the CT values corresponding to the respective pixels in the cancellous bone image obtained in step 12. In this step, by using the Johnson distribution function to fit the CT value probability distribution data, its greatest advantage lies in that, compared with fitting by a conventional function such as the Normal distribution function (usually with 2 variables), a better fitting effect for the data can be obtained, so that in subsequent steps, more accurate bone density values can be determined based on the obtained probability density function. Figure 3 Schematic diagram of the probability density function fitted based on the Johnson SU function provided by an embodiment of the present invention, as Figure 3 shown, the probability density function 4 of the lumbar cancellous bone is obtained.

[0060] Step 15, according to the first probability density function, determine the CT values corresponding to the probability values within a preset range therein, and according to the CT values, determine the first bone density value.

[0061] In different embodiments, the bone density value of the target bone can be determined according to the CT values corresponding to the probability values within different preset ranges. In one embodiment, the probability value within the preset range can be the probability maximum value of the first probability density function. That is to say, according to the CT value with the maximum occurrence probability among all the CT values in the first probability density function, the bone density value of the target bone is determined. For example, Figure 3In the illustrated embodiment, the CT value corresponding to the position 3 where the probability maximum value is located can be determined according to the probability density function 4, for example, it is 1350, and the bone density value of the target bone can be determined. In another embodiment, the probability value within the preset range can be any probability value whose difference from the illustrated probability maximum value is less than the predetermined threshold, and the bone density value of the target bone can be determined according to the CT value corresponding to this probability value.

[0062] According to an embodiment of another aspect of this specification, a method for constructing a bone density database is provided. Figure 5 The flowchart of a method for constructing a bone density database provided by an embodiment of the present invention is as Figure 5 shown, and this method includes the following steps:

[0063] Step 51, obtain the attribute data of multiple target bones.

[0064] In different embodiments, the attribute data can be different types of attribute data. In one embodiment, the attribute data can include one or more of age, gender, region, bone name.

[0065] Step 52, based on the bone density measurement method described in the above embodiments, obtain the bone density values of the multiple target bones.

[0066] In this step, based on Figure 1 the method shown, determine the bone density values of the multiple target bones.

[0067] Step 53, associatively save the bone density values of the target bones and their attribute data in a preset database.

[0068] According to one implementation manner, the bone density values of the multiple target bones can also be divided into several groups according to the attribute data of the multiple target bones;

[0069] According to the bone density values included in each group, determine the standard bone density value and variance value corresponding to each group, and save the standard bone density value and variance value in a preset database.

[0070] In different specific embodiments, the manner of determining the standard bone density value corresponding to the group can be different, and this specification does not limit this. For example, the standard bone density value corresponding to it can be determined according to the mean value of the bone density values included in the group.

[0071] In different embodiments, the multiple target bones may be grouped according to different one or more attribute data. For example, the multiple target bones may be grouped at one or multiple levels according to one or more of age, gender, region, etc. Furthermore, the standard bone density value and variance of the one or multiple level groupings may also be determined. In one example, grouping may be performed according to male, female, and different age groups, so as to obtain the standard bone density and variance of different genders and different age groups.

[0072] In one embodiment, in the above database, the standard bone density and variance of different age groups and different genders saved therein can be used to provide a basis for disease diagnosis including osteoporosis.

[0073] Using the bone density measurement method provided by the embodiments of the present specification, the bone density data of the human bones can be determined by a computer only based on the CT scan images of the human bones. And CT equipment is a conventional equipment widely owned by hospitals, and CT scan images are relatively easy to obtain. Therefore, this method greatly reduces the restrictions on bone density measurement in terms of process or equipment, and can more conveniently perform bone density measurement while ensuring the measurement effect. On this basis, using the method for constructing a bone density database provided by the embodiments of the present specification, a database storing the bone density data of different types of users and their statistical indicators can be constructed, so as to provide a diagnosis basis for medical diagnosis for different types of users.

[0074] According to an embodiment of another aspect of the present specification, a bone density measurement device is provided. Figure 6 The following is a structural diagram of a bone density measurement device provided by an embodiment of the present invention, as Figure 6 shown, the device includes:

[0075] A CT image acquisition unit 61, configured to acquire a CT image of a target bone;

[0076] A cancellous bone image acquisition unit 62, configured to segment the CT image to acquire a cancellous bone image therein;

[0077] A probability distribution determination unit 63, configured to determine the probability distribution of the CT values corresponding to the pixel values according to the pixel values in the cancellous bone image;

[0078] A probability density function determination unit 64, configured to fit the probability distribution based on a preset multi-parameter non-normal probability distribution function to obtain a first probability density function;

[0079] A bone density value determination unit 65, configured to determine the CT value corresponding to the probability value within a preset range, and determine a first bone density value according to the CT value.

[0080] According to an embodiment of another aspect of the present specification, a device for constructing a bone density database is provided. Figure 6 The following is a structural diagram of a device for constructing a bone density database provided by an embodiment of the present invention. As Figure 6 shown, the device includes:

[0081] A bone attribute acquisition unit 71, configured to acquire attribute data of a plurality of target bones;

[0082] A bone density determination unit 72, configured to obtain bone density values of the plurality of target bones based on the method described in claim 1;

[0083] A storage unit 73, configured to associatively store the bone density values of the target bones and their attribute data in a preset database.

[0084] According to an embodiment of another aspect, a computer-readable medium is further provided, including a computer program stored thereon, and the computer executes the above method when running.

[0085] According to an embodiment of another aspect, a computing device is further provided, including a memory and a processor, where an executable code is stored in the memory, and when the processor executes the executable code, the above method is implemented.

[0086] The above describes specific embodiments of the present specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than in the embodiments and still achieve the desired results. Additionally, the processes depicted in the drawings do not necessarily require the particular order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0087] Those skilled in the art should further realize that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, computer software, or a combination of the two. To clearly illustrate the interchangeability of hardware and software, the components and steps of each example have been generally described according to their functions in the above description. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Skilled artisans can use different methods for each specific application to implement the described functions, but such implementation should not be considered to exceed the scope of the present invention.

[0088] The steps of the methods or algorithms described in connection with the embodiments disclosed herein may be implemented by hardware, software modules executed by a processor, or a combination of both. The software modules may be placed in a random access memory (RAM), memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium well-known in the art.

[0089] The specific embodiments described above have further elaborated on the purpose, technical solutions, and beneficial effects of the present invention. It should be understood that the above description is only the specific embodiments of the present invention and is not used to limit the protection scope of the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A bone density measurement method, comprising: Obtaining a CT image of a target bone; Performing image segmentation on the CT image to obtain a cancellous bone image therein; Determining the probability distribution of the CT values corresponding to the pixel values according to the pixel values in the cancellous bone image; Fitting the probability distribution based on a preset multi-parameter non-normal probability distribution function to obtain a first probability density function; Determining the CT value corresponding to the probability value within a preset range according to the first probability density function, and determining a first bone density value according to the CT value.

2. The method according to claim 1, wherein, The target bone includes the first lumbar vertebra and the second lumbar vertebra.

3. The method according to claim 1, wherein, Performing segmentation on the CT image includes: Performing segmentation on the CT image by a preset machine learning-based image segmentation method.

4. The method according to claim 1, wherein The preset multi-parameter non-normal probability distribution function is one of Johnson SU, Johnson SB, and Johnson SL functions.

5. The method according to claim 1, wherein, The probability value within the preset range includes the probability maximum value of the first probability density function.

6. A method for constructing a bone density database, comprising: Obtaining attribute data of multiple target bones; Obtaining the bone density values of the multiple target bones based on the method described in claim 1; Associatively storing the bone density values of the target bones and their attribute data in a preset database.

7. The method according to claim 6, wherein The attribute data includes one or more of age, gender, region, and bone name.

8. The method according to claim 6, further comprising: Dividing the bone density values of the multiple target bones into several groups according to the attribute data of the multiple target bones; Determining the standard bone density value and variance value corresponding to each group according to the bone density values included in each group, and storing the standard bone density value and variance value in a preset database.

9. A bone density measurement device, comprising: A CT image acquisition unit configured to obtain a CT image of a target bone; A cancellous bone image acquisition unit configured to perform segmentation on the CT image to obtain a cancellous bone image therein; A probability distribution determination unit configured to determine the probability distribution of the CT values corresponding to the pixel values according to the pixel values in the cancellous bone image; A probability density function determination unit configured to fit the probability distribution based on a preset multi-parameter non-normal probability distribution function to obtain a first probability density function; A bone density value determination unit configured to determine the CT value corresponding to the probability value within a preset range, and determine a first bone density value according to the CT value.

10. A device for constructing a bone density database, comprising: A bone attribute acquisition unit configured to obtain attribute data of multiple target bones; A bone density determination unit configured to obtain the bone density values of the multiple target bones based on the method described in claim 1; A storage unit configured to associatively store the bone density values of the target bones and their attribute data in a preset database.

Images