CN-122023476-A - Image registration method, apparatus, computer device, storage medium, and program product

Abstract

The application relates to an image registration method, an image registration device, a computer device, a storage medium and a program product. The method is applied to a mobile PET system, and is used for obtaining reconstructed PET images by obtaining PET detection data based on particle beam irradiation and carrying out image reconstruction according to the PET detection data, and carrying out rigid registration on CT images to be registered based on the reconstructed PET images to obtain target images which are aligned with PET image space after registration. The original signal, namely the PET detection data based on particle beam irradiation, is used for image reconstruction, and then rigid registration is carried out on CT images to be registered, so that not only is no extra tracer injected required, but also the preparation time is reduced, the requirement on positioning is reduced, and the accuracy of subsequent image processing and final beam range verification is ensured.

Inventors

• LI JIAJIE
• WAN LIN
• Yuan Shuanghu
• QIU AO
• LIU QIBIN
• XIE QINGGUO
• LIU YANG
• HU WENTAO
• LI XIMENG
• LIU ZIXIAO
• LI LI

Assignees

• 苏州瑞派宁科技有限公司

Dates

Publication Date

20260512

Application Date

20251226

Claims (20)

1. 1. A method of image registration, the method being applied to a mobile PET system, the method comprising: acquiring PET detection data based on particle beam irradiation, and performing image reconstruction according to the PET detection data to obtain a reconstructed PET image; and carrying out rigid registration on CT images to be registered based on the reconstructed PET images to obtain target images which are spatially aligned with the PET images after registration.
2. 2. The image registration method of claim 1, wherein the PET detection data includes scour effect data, the scour effect data being first coincidence event data acquired after the end of particle beam irradiation.
3. 3. The image registration method of claim 2, wherein the PET detection data further comprises second coincidence event data acquired during particle beam irradiation.
4. 4. The image registration method of claim 1, wherein the PET detection data includes third coincidence event data acquired during any period of time from a start of particle beam irradiation to a preset period of time after an end of irradiation.
5. 5. The image registration method of claim 1, wherein the mobile PET system comprises: Any of a mobile open loop PET system, a mobile closed loop PET system, and a mobile flat panel PET system.
6. 6. The image registration method according to claim 1, wherein the performing image reconstruction from the PET detection data to obtain a reconstructed PET image comprises: correcting the PET detection data to obtain corrected event data; And (3) carrying out image reconstruction by adopting a preset reconstruction algorithm based on the corrected event data to obtain a reconstructed PET image.
7. 7. The image registration method of claim 6, wherein correcting the PET detection data to obtain target event data comprises: And carrying out normalization correction, dead time correction and random correction on the PET detection data to obtain corrected event data.
8. 8. The image registration method of claim 6, wherein the reconstruction algorithm comprises any one of a maximum likelihood expectation maximization and time-of-flight reconstruction algorithm, an ordered subset expectation maximization reconstruction algorithm, and a filtered back projection reconstruction algorithm.
9. 9. The image registration method according to claim 1, wherein the rigid registration of CT images to be registered based on reconstructed PET images comprises: and taking the reconstructed PET image as a fixed image, loading a CT image to be registered, and carrying out rigid registration on the CT image.
10. 10. The image registration method of claim 9, wherein a first feature is included in the PET image and a second feature corresponding to the first feature is included in the CT image; The rigid registration of the CT images includes: The CT image is rigidly transformed based on the first feature in the PET image to align the second feature in the CT image with the first feature in the PET image.
11. 11. The image registration method of claim 10, wherein the rigid transformation includes at least one of a translational operation and a rotational operation.
12. 12. The image registration method of claim 10, wherein the first feature and the second feature comprise: at least one of body contour features, bone features, organ features, and lesion features.
13. 13. The image registration method according to any one of claims 1 to 12, wherein the CT image to be registered comprises any one of a planning CT for planning a treatment and a localization CT for localization.
14. 14. An image registration apparatus, the apparatus being applied to a mobile PET system, the apparatus comprising: the data acquisition module is configured to acquire PET detection data based on particle beam irradiation, and perform image reconstruction according to the PET detection data to obtain a reconstructed PET image; and the registration module is configured to perform rigid registration on CT images to be registered based on the reconstructed PET images, and obtain target images which are spatially aligned with the PET images after registration.
15. 15. The image registration apparatus of claim 14, wherein the PET detection data includes scour effect data, the scour effect data being first coincidence event data acquired after the end of particle beam irradiation.
16. 16. The image registration apparatus of claim 15, wherein the PET detection data further comprises second coincidence event data acquired during particle beam irradiation.
17. 17. The image registration apparatus of claim 14, wherein the PET detection data includes third coincidence event data acquired during any period of time from a start of particle beam irradiation to a preset period of time after an end of irradiation.
18. 18. The image registration apparatus of claim 14, wherein the mobile PET system comprises: Any of a mobile open loop PET system, a mobile closed loop PET system, and a mobile flat panel PET system.
19. 19. The image registration apparatus of claim 14, wherein the data acquisition module further comprises: a correction unit configured to perform correction of the PET detection data to obtain corrected event data; and the reconstruction unit is configured to execute image reconstruction by adopting a preset reconstruction algorithm based on the corrected event data to obtain a reconstructed PET image.
20. 20. The image registration apparatus of claim 19, wherein the correction unit is further configured to perform normalization correction, dead time correction, and random correction on the PET detection data to obtain corrected event data.

Description

Image registration method, apparatus, computer device, storage medium, and program product Technical Field The present application relates to the field of image data processing technology, and in particular, to an image registration method, an image registration apparatus, a computer device, a computer readable storage medium, and a computer program product. Background Particle beam therapy (such as proton or heavy ion therapy) has become an important development of advanced radiation therapy because of its unique bragg peak physical properties, enabling precise deposition of high dose radiation at a tumor target region, significantly reducing damage to surrounding normal tissues. However, this technique is highly sensitive to patient anatomy changes, placement errors, and tissue density heterogeneity, and minor deviations can lead to range shifts, causing target under-irradiation or organ-jeopardy overexposure, severely threatening therapeutic safety and efficacy. Therefore, it is important to realize real-time and accurate range verification of the particle beam in the body. The online positron emission tomography (in-beam PET) technology can indirectly reflect the actual stop position of the beam current by detecting positron annihilation signals generated by the reaction of the particle beam and tissue nuclei, and is regarded as a verification means with the most clinical transformation potential. In the related art, in-beam PET systems are mainly divided into two types, i.e., integrated type and non-integrated type. Although the integration scheme can realize the registration of the natural coordinate system, complicated mechanical transformation is required to be carried out on the treatment equipment, and the problems of high cost, influence on the authentication of the original equipment, clinical operation and the like exist. In contrast, non-integrated, mobile PET systems have a more promising deployment prospect due to the lack of need to retrofit treatment devices, flexibility of deployment. However, such systems face a critical bottleneck in clinical applications, in that they lack an inherent alignment relationship with the coordinate system of the particle therapy system, requiring hardware alignment prior to each treatment relying on high precision external positioning devices to acquire coordinate transformation parameters for inter-image registration, or PET images acquired by injection of fluorodeoxyglucose (Fluoro Deoxy Glucose, abbreviated as FDG). Not only does this approach significantly extend the preparation time, increasing the labor burden, but the additional injection of FDG also results in increased costs and radiation risks. Disclosure of Invention The present application provides an image registration method, apparatus, computer device, computer-readable storage medium, and computer program product to solve at least the above-mentioned problems in the related art. The technical scheme of the application is as follows: According to a first aspect of an embodiment of the present application, there is provided an image registration method, the method being applied to a mobile PET system, the method comprising: acquiring PET detection data based on particle beam irradiation, and performing image reconstruction according to the PET detection data to obtain a reconstructed PET image; and carrying out rigid registration on CT images to be registered based on the reconstructed PET images to obtain target images which are spatially aligned with the PET images after registration. In one embodiment, the PET detection data includes scour effect data, which is first coincidence event data acquired after the end of particle beam irradiation. In one embodiment, the PET detection data further comprises second coincidence event data acquired during particle beam irradiation. In one embodiment, the PET detection data includes third coincidence event data acquired during any period from the start of irradiation of the particle beam to a preset time period after the end of irradiation. In one embodiment, the mobile PET system includes any of a mobile open loop PET system, a mobile closed loop PET system, and a mobile flat panel PET system. In one embodiment, the image reconstruction is performed according to the PET detection data to obtain a reconstructed PET image, and the image reconstruction comprises the steps of correcting the PET detection data to obtain corrected event data, and performing image reconstruction by adopting a preset reconstruction algorithm based on the corrected event data to obtain the reconstructed PET image. In one embodiment, the correcting the PET detection data to obtain target event data includes normalizing correction, dead time correction and random correction on the PET detection data to obtain corrected event data. In one embodiment, the reconstruction algorithm includes any one of a maximum likelihood expectation maximization and time-of-flight reconstruction a