PCCT-based vBMD detection method and system

Abstract

The invention relates to a PCCT-based vBMD detection method and system, and the method comprises the following steps: collecting an original image through PCCT scanning, and reconstructing the original image to obtain a spectral image in an SPP format; inputting the spectral image in the SPP format into a post-processing workstation to generate a calcium image, and obtaining a CaCT value; according to a conversion coefficient for converting the CaCT value into a vBMD value, obtaining the vBMD value through the CaCT value; wherein the establishment of the conversion coefficient for converting the CaCT value into the vBMD value comprises the following steps: taking the real hydroxyapatite concentration of the phantom as the vBMD value; an ROI is set in the phantom, the ROI is scanned through PCCT, and a spectral image in an SPP format is obtained through reconstruction; inputting the spectral image in the SPP format into the post-processing workstation to obtain a calcium image, and obtaining an average CaCT value of the ROI; and obtaining a conversion coefficient for converting the CaCT value into the vBMD value according to the vBMD value and the average CaCT value. Compared with the prior art, the labor force is saved, and the cost benefit is improved.

Description

Technical Field

[0001] The present invention relates to the field of bone density detection, and in particular to a PCCT-based vBMD detection method and system. Background Art

[0002] In recent years, with the development of population aging, the incidence of osteoporosis, characterized by bone loss and bone tissue microstructure disorder, has gradually increased, increasing the risk of fractures and socioeconomic burden on patients. Bone mineral density (BMD) value is the most commonly used quantitative indicator of osteoporosis and is considered an independent predictor of fracture risk and all-cause mortality.

[0003] Dual-energy X-ray absorptiometry (DXA) generates areal BMD (aBMD) values ​​for the entire vertebrae and is a convenient and cost-effective method for detecting osteoporosis. However, the accuracy of this two-dimensional measurement method is limited by spinal degeneration, vascular calcification, and scoliosis. Quantitative computed tomography (QCT) uses a three-dimensional approach to delineate regions of interest in cancellous bone, which can obtain more realistic volumetric BMD (vBMD) values ​​without artifacts caused by tissue overlap. However, QCT requires regular phantom calibration to maintain the accuracy of the measurements, which is time-consuming and labor-intensive. Dual-energy CT (DECT) can distinguish materials using the energy dependence of the photoelectric effect under different X-ray spectra. Compared with QCT, DECT-based BMD quantification based on hydroxyapatite specificity has comparable diagnostic accuracy in detecting abnormal BMD in vivo. However, with the exception of dual-detector CT, currently commonly used DECT methods all require preset sequences and cannot obtain vBMD values ​​in conventional CT scans, making it difficult to perform opportunistic screening for osteoporosis. Summary of the invention

[0004] The purpose of the present invention is to overcome the defects of the above-mentioned prior art and provide a vBMD detection method and system based on PCCT, which saves labor and improves cost-effectiveness.

[0005] The purpose of the present invention can be achieved by the following technical solutions:

[0006] A vBMD detection method based on PCCT comprises the following steps:

[0007] Collecting an original image through PCCT scanning, and reconstructing the original image to obtain a spectral image in SPP format;

[0008] Inputting the spectral image in the SPP format into a post-processing workstation to generate a calcium map and obtain a CaCT value;

[0009] Obtaining the vBMD value using the CaCT value according to a conversion coefficient for converting the CaCT value into the vBMD value;

[0010] The steps for establishing the conversion coefficient of the CaCT value to the vBMD value are as follows:

[0011] The actual hydroxyapatite concentration of the phantom was taken as the vBMD value;

[0012] Setting a ROI in the phantom, scanning the ROI by PCCT, and reconstructing and acquiring a spectral image in an SPP format;

[0013] Inputting the spectral image in the SPP format into the post-processing workstation to obtain a calcium map, and obtaining an average CaCT value of the ROI;

[0014] A conversion coefficient for converting the CaCT value into the vBMD value is obtained according to the vBMD value and the average CaCT value.

[0015] Furthermore, the reconstruction parameters include multiple ones of layer thickness, layer spacing, quantitative kernel, quantum iterative reconstruction level, matrix size and scanning field of view.

[0016] Furthermore, the phantom is a standardized European spine phantom, comprising three artificial vertebrae.

[0017] Furthermore, ROIs are drawn at the center, upper third, and lower third levels of each of the artificial vertebrae.

[0018] Furthermore, the conversion coefficient of the CaCT value to the vBMD value is:

[0019] vBMD=CaCT×0.82

[0020] Wherein, vBMD is the vBMD value, and CaCT is the CaCT value.

[0021] Furthermore, the calcium map is generated based on the decomposition of three substances, which include water, air and calcium.

[0022] Furthermore, the spectral image in the SPP format is used to separate water, air and calcium based on spectral analysis on the post-processing workstation.

[0023] Furthermore, the QCT value obtained by QCT scanning is compared with the vBMD value obtained by the conversion coefficient of converting the CaCT value into the vBMD value to verify the feasibility of the conversion coefficient.

[0024] Furthermore, the QCT scanned phantom was calibrated every three days.

[0025] According to another aspect of the present invention, there is provided a PCCT-based vBMD detection system, comprising:

[0026] A PCCT scanning module, used for acquiring a spectral image through PCCT scanning, and reconstructing the spectral image to obtain a spectral image in an SPP format;

[0027] A CaCT value acquisition module, which generates a calcium map by inputting the spectral image in the SPP format into a post-processing workstation to obtain a CaCT value;

[0028] A vBMD value acquisition module, which obtains the vBMD value through the CaCT value according to a conversion coefficient of converting the CaCT value into the vBMD value;

[0029] The steps for establishing the conversion coefficient of the CaCT value to the vBMD value are as follows:

[0030] The actual hydroxyapatite concentration of the phantom was taken as the vBMD value;

[0031] Setting a ROI in the phantom, scanning the ROI by PCCT, and reconstructing and acquiring a spectral image in an SPP format;

[0032] Inputting the spectral image in the SPP format into the post-processing workstation to obtain a calcium map, and obtaining an average CaCT value of the ROI;

[0033] A conversion coefficient for converting the CaCT value into the vBMD value is obtained according to the vBMD value and the average CaCT value.

[0034] Compared with the prior art, the present invention has the following beneficial effects:

[0035] 1. The present invention uses the actual hydroxyapatite concentration of the phantom as the vBMD value; sets a ROI in the phantom, scans the ROI through PCCT, and reconstructs and obtains an SPP map; obtains a calcium map based on the SPP map to obtain an average CaCT value of the ROI; and obtains a conversion coefficient for converting the CaCT value to the vBMD value based on the vBMD value and the average CaCT value. This avoids errors in obtaining area BMD values ​​using two-dimensional images, such as interference from overlapping artifacts such as spinal degeneration, scoliosis, and vascular sclerosis.

[0036] 2. The present invention establishes a conversion coefficient for converting CaCT values ​​into vBMD values, obtains a spectral image based on PCCT scanning, reconstructs the spectral image to generate an SPP format image, generates a calcium map through a post-processing workstation to obtain a CaCT value, and uses the conversion coefficient for converting CaCT values ​​into vBMD values ​​to obtain the human body vBMD value. This can be performed without using a calibrated phantom and a relatively expensive QCT post-processing workstation, thereby saving labor and improving cost-effectiveness. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1This is a schematic diagram of the process of the PCCT-based vBMD detection method proposed in the present invention;

[0038] Figure 2 This is a flow chart of the conversion coefficient of converting CaCT value into vBMD value in the PCCT-based vBMD detection method proposed in the present invention. DETAILED DESCRIPTION

[0039] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. This embodiment is implemented based on the technical solution of the present invention, and provides a detailed implementation method and specific operation process, but the protection scope of the present invention is not limited to the following embodiments.

[0040] Abbreviations involved:

[0041] Photon Counting Computed Tomography (PCCT)

[0042] Volumetric Bone Mineral Density (vBMD)

[0043] Region of Interest, ROI

[0044] Volume of Interest, VOI

[0045] Spectral Processing Package: Spectral Processing Package, SPP

[0046] Quantitative Computed Tomography, QCT

[0047] Hounsfield unit: Hounsfield unit, HU

[0048] Calcium CT attenuation, CaCT

[0049] Example 1

[0050] This embodiment provides a vBMD detection method based on PCCT, such as Figure 1 As shown, the following steps are included:

[0051] S1. Collect the original image through PCCT scanning, and reconstruct the original image to obtain the spectral image in SPP format.

[0052] The patient is placed in the supine position for PCCT examination, with arms straightened above the head, and the chest and abdomen or (ultra-high resolution) lumbar spine scanning mode is selected for scanning. The ultra-high resolution lumbar spine scanning ranges from the 11th thoracic vertebra to the first sacral vertebra, and the chest and abdomen scanning ranges from 2 cm above the diaphragm to the bottom of the pelvis, both using a spiral scanning mode.

[0053] The specific scanning parameters for chest and abdomen scans were as follows: tube voltage, 120 kV; CARE keV image quality grade, 145; automatic tube current adjustment (CareDose 4D), optimized for non-contrast enhancement; collimation, 144 × 0.4 mm; pitch, 0.8; and rotation time, 0.25 s.

[0054] The specific scanning parameters of the ultra-high-resolution lumbar spine scan were as follows: tube voltage, 120 kV; CARE keV image quality grade, 285; CareDose 4D, optimized for bone / calcium; collimation, 120 × 0.2 mm; pitch, 0.8; and rotation time, 0.50 s.

[0055] The specific reconstruction parameters of the SPP image are as follows: the layer thickness / layer spacing is 1.0 / 0.7 mm, the quantitative kernel is Qr40, the quantum iterative reconstruction level is 3, and the matrix size is selected as 512×512pix 2 The field of view of chest and abdomen scan and lumbar spine scan were set to 413×413mm respectively. 2 and 185×185mm 2 Reconstruction generates threshold 3D images for assessment of QCT-derived vBMD (threshold 3D images are comparable to those obtained from standard energy-integrating detector CT scans performed at equivalent X-ray tube voltages and are reconstructed from a data stream containing all X-ray quanta with energies above a minimum threshold energy).

[0056] S2. Input the spectral image in SPP format into the Syngovia VB60 workstation to generate a calcium map and obtain the CaCT value.

[0057] On the Syngovia VB60 workstation, the "Calcium Quantification" module in the "Dual Energy" module was used to generate calcium maps using SPP format images. The calcium map was generated based on the decomposition of three substances, including tissue, fat, and calcium. PCCT can set different energy thresholds to divide photons into two or more energy boxes according to the size of the energy for comprehensive energy analysis. In the low-energy box, the Hounsfield unit (HU) value of the tissue is 55, the HU value of fat is -110, and the relative contrast material value is 1.55; in the high-energy box, the HU value of the tissue remains at 55, the HU value of fat also remains at -96, and the relative contrast material value is 1.67.

[0058] In the target vertebra, the volume region of interest (VOI) generated by the QCT post-processing workstation (QCT Pro V6.1, Mindways Software Inc.) was simulated, and the manual function was selected in the Syngovia VB60 workstation to draw the VOI, so that its size and position were as consistent as possible with the VOI derived from QCT Pro. The specific steps for drawing the VOI are as follows: First, a 9mm straight line parallel to the longitudinal axis of the vertebra was drawn in the sagittal plane, and the largest possible ROI was drawn on the cross-sections at both ends of the straight line, while avoiding the surrounding cortical bone and the posterior vertebral venous plexus. Then, with a height of 9mm and two ROIs as the upper and lower bottom surfaces, click "createVOI" to create a cylindrical VOI, and the workstation automatically generates the CT value of the VOI, which is the CaCT value of the vertebra.

[0059] S3. Obtain the vBMD value using the CaCT value according to the conversion coefficient of the CaCT value into the vBMD value.

[0060] The steps for establishing the conversion coefficient from CaCT value to vBMD value are as follows: Figure 2 As shown, the following steps are included:

[0061] A standardized European spine phantom (ESP, QRM GmbH) containing three artificial vertebrae with real hydroxyapatite concentrations of 49, 102, and 198 mg / mL, representing vBMD, was scanned three times using six different scanning protocols at a tube voltage of 120 kV on a PCCT system (NAEOTOM Alpha, VA50, Siemens Healthcare). The ESP was scanned in ultra-high-resolution lumbar spine scanning mode with the following scanning parameters: voltage: 120 kV; CARE keV image quality level: 285; CareDose 4D: optimized for bone / calcium; collimation, 120 × 0.2 mm; spacing, 0.8 mm; rotation time, 0.50 s.

[0062] After the PCCT scanner acquires binary raw data, the system reconstructs images in spectral post-processing (SPP) format for vBMD measurement according to the following parameters: slice thickness: 1.0 mm; slice spacing: 0.8 mm; quantitative kernel: Qr40; quantum iterative reconstruction level: 3; matrix size, 512 × 512 pix 2 ; Scanning field size: 336×336mm 2 ; Collimator width: 144 × 0.4 mm. All images were transferred to a post-processing workstation (Syngovia VB60, Siemens Healthcare) to achieve species separation based on spectral analysis and thus calculate calcium concentration.

[0063] The above scanning and reconstruction were repeated twice to verify the repeatability, and the average of the three measurements was taken as the result to reduce the measurement error.

[0064] On the Syngovia VB60 workstation, the calcium quantification module in the "dual energy" module was used to generate a calcium map using the SPP image. The calcium map was generated based on the decomposition of three substances, including water, air, and calcium. PCCT can set different energy thresholds to divide photons into two or more energy boxes according to the size of the energy for comprehensive energy analysis. In the low-energy box, the HU value of water is 0, the HU value of air is -1000, and the relative contrast material (Rel.CM) value is 1.55; in the high-energy box, the HU value of water remains at 0, the HU value of air also remains at -1000, and the relative contrast material value is 1.67.

[0065] Bone density was measured by a radiologist with 12 years of experience in musculoskeletal radiology. Three ROIs were drawn at the central, upper third, and lower third levels of each vertebra. They were as large as possible to avoid contact with the bone cortex. After the ROIs were drawn, they were scanned by PCCT and reconstructed to obtain spectral images in SPP format. The spectral images in SPP format were input into the post-processing workstation to obtain the calcium map and the CaCT value of each ROI. The average of the CaCT values ​​of the three ROIs was the CaCT value of the artificial vertebra that was finally included in the calculation of the conversion coefficient.

[0066] Calcium Hydroxyapatite Ca 10 (PO4)6(OH)2 is the main mineral component in the ESP phantom and the human body, with calcium accounting for 39.9% of its total weight. The calcium map can measure the CaCT value. Therefore, based on the actual concentration of the phantom, the conversion coefficient and fitting curve of the CaCT value to the vBMD value can be calculated for the acquisition of the vBMD value in the human body. The conversion coefficient of the CaCT value to the vBMD value is:

[0067] vBMD=CaCT×0.82

[0068] Wherein, vBMD is the vBMD value, and CaCT is the CaCT value.

[0069] Thirty adult patients who underwent lumbar spine or thoracoabdominal scans on PCCT were included in the study. The QCT values ​​of the 30 patients were used as the reference standard. QCT requires regular phantom calibration to maintain the accuracy of the measured values. The QCT phantom in this example was calibrated once every three days. The QCT values ​​were compared with the vBMD values ​​obtained by the conversion formula of CaCT values ​​to vBMD values. The correlation and consistency between the two were compared. The results showed that there was a good correlation between the two (r s=0.797, p<0.001) and very high agreement (bias, -2.4; 95% LOA, -26.7 to 21.8 mg / cm 3 ), indicating that the conversion coefficient of CaCT value to vBMD value is feasible for application in human body.

[0070] Example 2

[0071] This embodiment provides a vBMD detection system based on PCCT, including:

[0072] A PCCT scanning module is used to collect original images through PCCT scanning, and reconstruct the original images to obtain spectral images in SPP format;

[0073] The CaCT value acquisition module generates a calcium map by inputting the spectral image in SPP format into the post-processing workstation to obtain the CaCT value;

[0074] The vBMD value acquisition module obtains the vBMD value through the CaCT value according to the conversion coefficient of the CaCT value to the vBMD value;

[0075] The steps for establishing the conversion coefficient from CaCT value to vBMD value are as follows:

[0076] The actual hydroxyapatite concentration of the phantom was taken as the vBMD value;

[0077] Set ROI in the phantom, scan the ROI through PCCT, and reconstruct the spectral image in SPP format;

[0078] The spectral image in SPP format was input into the post-processing workstation to obtain the calcium map and the average CaCT value of the ROI;

[0079] The conversion coefficient of CaCT value to vBMD value was obtained based on the vBMD value and the average CaCT value.

[0080] The rest is the same as in Example 1.

[0081] The preferred specific embodiments of the present invention are described in detail above. It should be understood that a person skilled in the art can make many modifications and changes based on the concept of the present invention without creative work. Therefore, any technical solution that can be obtained by a person skilled in the art through logical analysis, reasoning or limited experiments based on the concept of the present invention on the basis of the prior art should be within the scope of protection determined by the claims.

Claims

1. A vBMD detection method based on PCCT, characterized in that: The following steps are involved: Collecting an original image through PCCT scanning, and reconstructing the original image to obtain a spectral image in SPP format; Inputting the spectral image in the SPP format into a post-processing workstation to generate a calcium map and obtain a CaCT value; Obtaining a vBMD value using the CaCT value according to a conversion coefficient for converting the CaCT value into the vBMD value; The steps for establishing the conversion coefficient of the CaCT value to the vBMD value are as follows: The actual hydroxyapatite concentration of the phantom was taken as the vBMD value; Setting a ROI in the phantom, scanning the ROI by PCCT, and reconstructing and acquiring a spectral image in an SPP format; Inputting the spectral image in the SPP format into the post-processing workstation to obtain a calcium map, and obtaining an average CaCT value of the ROI; A conversion coefficient for converting the CaCT value into the vBMD value is obtained according to the vBMD value and the average CaCT value.

2. The PCCT-based vBMD detection method according to claim 1, characterized in that: The reconstruction parameters include various ones of layer thickness, layer spacing, quantitative kernel, quantum iterative reconstruction level, matrix size and scanning field of view.

3. The PCCT-based vBMD detection method according to claim 1, characterized in that: The phantom was a standardized European spine phantom, comprising three artificial vertebrae.

4. The PCCT-based vBMD detection method according to claim 3, characterized in that: ROIs were drawn at the central, upper third, and lower third levels of each of the artificial vertebrae.

5. The PCCT-based vBMD detection method according to claim 1, characterized in that: The conversion coefficient of the CaCT value to the vBMD value is: vBMD=CaCT×0.82 Wherein, vBMD is the vBMD value, and CaCT is the CaCT value.

6. The PCCT-based vBMD detection method according to claim 1, characterized in that: The calcium map is generated based on a three-substance decomposition, including water, air, and calcium.

7. The PCCT-based vBMD detection method according to claim 1, characterized in that: The spectral image in the SPP format is used to separate water, air and calcium based on spectral analysis on the post-processing workstation.

8. The PCCT-based vBMD detection method according to claim 1, characterized in that: The QCT value obtained by QCT scanning is compared with the vBMD value obtained by the conversion coefficient of converting the CaCT value into the vBMD value to verify the feasibility of the conversion coefficient.

9. The PCCT-based vBMD detection method according to claim 8, characterized in that: The QCT scanned phantom was calibrated every three days.

10. A vBMD detection system based on PCCT, characterized in that: include: A PCCT scanning module, used for acquiring a spectral image through PCCT scanning, and reconstructing the spectral image to obtain a spectral image in an SPP format; A CaCT value acquisition module, which generates a calcium map by inputting the spectral image in the SPP format into a post-processing workstation to obtain a CaCT value; A vBMD value acquisition module, which obtains the vBMD value through the CaCT value according to a conversion coefficient of converting the CaCT value into the vBMD value; The steps for establishing the conversion coefficient of the CaCT value to the vBMD value are as follows: The actual hydroxyapatite concentration of the phantom was taken as the vBMD value; Setting a ROI in the phantom, scanning the ROI by PCCT, and reconstructing and acquiring a spectral image in an SPP format; Inputting the spectral image in the SPP format into the post-processing workstation to obtain a calcium map, and obtaining an average CaCT value of the ROI; A conversion coefficient for converting the CaCT value into the vBMD value is obtained according to the vBMD value and the average CaCT value.

Images