CN114943714A - Medical image processing system, device, electronic equipment and storage medium - Google Patents
Medical image processing system, device, electronic equipment and storage mediumDownload PDFInfo
- Publication number
- CN114943714A CN114943714ACN202210610225.3ACN202210610225ACN114943714ACN 114943714 ACN114943714 ACN 114943714ACN 202210610225 ACN202210610225 ACN 202210610225ACN 114943714 ACN114943714 ACN 114943714A
- Authority
- CN
- China
- Prior art keywords
- image
- body part
- interest
- metabolic
- organ
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
- G06T7/33—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/80—Fusion, i.e. combining data from various sources at the sensor level, preprocessing level, feature extraction level or classification level
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
- G06T2207/10081—Computed x-ray tomography [CT]
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
- G06T2207/10104—Positron emission tomography [PET]
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Medical Informatics (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Multimedia (AREA)
- Software Systems (AREA)
- Computing Systems (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Databases & Information Systems (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Quality & Reliability (AREA)
- Apparatus For Radiation Diagnosis (AREA)
- Image Processing (AREA)
Abstract
Translated fromChinese
Description
Translated fromChinese技术领域technical field
本发明涉及医疗技术领域，具体涉及一种医学图像处理系统、装置、电子设备及存储介质。The present invention relates to the field of medical technology, in particular to a medical image processing system, device, electronic device and storage medium.
背景技术Background technique
医学图像诊断特别是计算机断层图像诊断(CT，Computed Tomography)、正电子发射断层图像诊断(PET，Positron Emission Tomography)以及磁共振图像诊断(MR，Magnetic Resonance)均是非常重要的医学图像诊断方式，可以无创地提供患者的解剖结构图像，从而为相关疾病诊断提供有效的技术支撑。Medical image diagnosis, especially Computed Tomography (CT, Computed Tomography), Positron Emission Tomography (PET, Positron Emission Tomography) and Magnetic Resonance Image Diagnosis (MR, Magnetic Resonance) are all very important medical image diagnosis methods. The patient's anatomical structure images can be provided non-invasively, thereby providing effective technical support for the diagnosis of related diseases.
目前，经过PET-CT系统检查之后，PET-CT系统可根据扫描图像自动筛选出异常代谢的感兴趣区域，当医生获取标识有异常代谢感兴趣区域的图像后，通过人工识别，判断出感兴趣区域所属的身体部位以及其具体的器官，然后人工进行图像的身体部位和器官分割，费时费力，效率较低，而且只有专业的人员才能根据图像看出病人的病灶器官和病灶所属身体部位，对专业能力要求较高。At present, after being checked by the PET-CT system, the PET-CT system can automatically screen out the area of interest with abnormal metabolism based on the scanned image. It is time-consuming and labor-intensive, and the efficiency is low, and only professional personnel can see the patient's focal organ and the body part to which the lesion belongs based on the image. Professional competence is required.
发明内容SUMMARY OF THE INVENTION
本发明的目的在于克服上述技术不足，提供一种医学图像处理系统、装置、电子设备及存储介质，解决现有技术中通过人工进行异常代谢感兴趣区域所对应的器官及器官所属身体部位划分的技术问题。The purpose of the present invention is to overcome the above-mentioned technical deficiencies, and provide a medical image processing system, device, electronic equipment and storage medium, which solves the problem of manually dividing the organs corresponding to the abnormal metabolism region of interest and the body parts to which the organs belong in the prior art. technical problem.
为达到上述技术目的，本发明采取了以下技术方案：In order to achieve the above-mentioned technical purpose, the present invention has adopted the following technical solutions:
第一方面，本发明提供一种医学图像处理系统，In a first aspect, the present invention provides a medical image processing system,
包括处理器和存储器；including processor and memory;
所述存储器上存储有可被所述处理器执行的计算机可读程序；A computer-readable program executable by the processor is stored on the memory;
所述处理器执行所述计算机可读程序时实现如下步骤：When the processor executes the computer-readable program, the following steps are implemented:
获取目标对象的扫描图像；Obtain the scanned image of the target object;
基于所述扫描图像，获取所述目标对象的代谢检测图像以及器官分割图像和身体部位图像中的至少一个，其中，所述代谢检测图像标识有异常代谢感兴趣区域；obtaining, based on the scanned image, a metabolic detection image of the target object and at least one of an organ segmentation image and a body part image, wherein the metabolic detection image identifies a region of interest with abnormal metabolism;
根据所述代谢检测图像以及器官分割图像和身体部位图像中的至少一个，确定与所述异常代谢感兴趣区域对应的器官和/或身体部位。An organ and/or body part corresponding to the abnormal metabolic region of interest is determined according to the metabolic detection image and at least one of an organ segmentation image and a body part image.
在其中一些实施例中，所述代谢检测图像基于PET图像或CT图像获取，所述器官分割图像基于CT图像或者MR图像获取，所述身体部位图像基于定位片获取。In some of these embodiments, the metabolic detection image is acquired based on a PET image or a CT image, the organ segmentation image is acquired based on a CT image or an MR image, and the body part image is acquired based on a localization slice.
在其中一些实施例中，所述基于所述扫描图像，获取所述目标对象的代谢检测图像以及器官分割图像和身体部位图像中的至少一个，其中，所述代谢检测图像标识有异常代谢感兴趣区域，包括：In some of these embodiments, the metabolic detection image of the target object and at least one of an organ segmentation image and a body part image are obtained based on the scanned image, wherein the metabolic detection image is identified as having abnormal metabolism of interest area, including:
基于所述PET图像或CT图像，获取所述目标对象的异常代谢感兴趣区域，以得到标识有异常代谢感兴趣区域的代谢检测图像；以及Based on the PET image or CT image, acquiring the abnormal metabolic region of interest of the target object to obtain a metabolic detection image marked with the abnormal metabolic region of interest; and
基于所述CT图像或MR图像，对所述目标对象进行器官分割，以得到所述目标对象的器官分割图像和/或身体部位图像，和/或performing organ segmentation on the target object based on the CT image or the MR image to obtain an organ segmentation image and/or a body part image of the target object, and/or
基于所述定位片，对所述目标对象进行身体部位分割，以得到所述目标对象的身体部位图像。Based on the localization slice, body part segmentation is performed on the target object to obtain a body part image of the target object.
在其中一些实施例中，所述根据所述代谢检测图像以及器官分割图像和身体部位图像中的至少一个，确定与所述异常代谢感兴趣区域对应的器官和/或身体部位，包括：In some of these embodiments, the determining the organ and/or body part corresponding to the abnormal metabolism region of interest according to the metabolic detection image and at least one of the organ segmentation image and the body part image includes:
获取图像之间的转换关系；Get the conversion relationship between images;
基于所述图像之间的转换关系，将所述器官分割图像和/或所述身体部位图像与所述代谢检测图像进行配准，以确定与所述异常代谢感兴趣区域对应的器官和/或身体部位。registering the organ segmentation image and/or the body part image with the metabolic detection image based on the transformation relationship between the images to determine the organ and/or the organ corresponding to the abnormal metabolic region of interest body parts.
在其中一些实施例中，所述基于所述图像之间的转换关系，将所述器官分割图像和所述身体部位图像与所述代谢检测图像进行配准，以确定与所述异常代谢感兴趣区域对应的器官和身体部位，包括：In some of these embodiments, the organ segmentation image and the body part image are registered with the metabolic detection image based on the transformation relationship between the images to determine the abnormal metabolism of interest Regions correspond to organs and body parts, including:
将所述身体部位图像与所述器官分割图像进行融合，以生成第一图像，其中，所述第一图像为对目标对象的身体部位和器官进行分割后的图像；fusing the body part image and the organ segmentation image to generate a first image, wherein the first image is an image obtained by segmenting the body part and the organ of the target object;
基于所述图像之间的转换关系，将所述代谢检测图像与所述第一图像进行配准，以生成第二图像后，基于所述第二图像确定与所述异常代谢感兴趣区域对应的器官和身体部位，其中，所述第二图像为对目标对象进行了器官分割以及身体部位分割后，并标识有异常代谢感兴趣区域的图像。Based on the conversion relationship between the images, the metabolic detection image is registered with the first image to generate a second image, and then a region of interest corresponding to the abnormal metabolism is determined based on the second image. Organs and body parts, wherein the second image is an image marked with a region of interest for abnormal metabolism after organ segmentation and body part segmentation are performed on the target object.
在其中一些实施例中，所述基于所述图像之间的转换关系，将所述代谢检测图像与所述第一图像进行配准，以生成第二图像后，基于所述第二图像确定与所述异常代谢感兴趣区域对应的器官和身体部位，包括：In some of the embodiments, after the metabolic detection image is registered with the first image based on the conversion relationship between the images, to generate a second image, determining the relationship with the second image based on the second image. The organs and body parts corresponding to the abnormal metabolism region of interest include:
对所述第一图像进行预处理，得到一组代表所述扫描对象的体素；Preprocessing the first image to obtain a set of voxels representing the scanned object;
根据各个体素的图像坐标以及所述第一图像的标签数据，计算出每个体素在CT物理坐标系中的空间坐标，以得到一初始点云模型；According to the image coordinates of each voxel and the label data of the first image, the spatial coordinates of each voxel in the CT physical coordinate system are calculated to obtain an initial point cloud model;
根据所述CT物理坐标系和PET物理坐标系之间的转换关系，对所述初始点云模型进行转换，以生成第一点云模型后；After converting the initial point cloud model according to the conversion relationship between the CT physical coordinate system and the PET physical coordinate system, to generate a first point cloud model;
根据所述第一图像和所述第一点云模型生成配准图像；generating a registration image according to the first image and the first point cloud model;
将所述配准图像和所述代谢检测图像进行融合，以得到第二图像。The registration image and the metabolic detection image are fused to obtain a second image.
在其中一些实施例中，所述基于所述图像之间的转换关系，将所述器官分割图像或所述身体部位图像与所述代谢检测图像进行配准，以确定与所述异常代谢感兴趣区域对应的异常器官或身体部位，包括：In some of these embodiments, the organ segmentation image or the body part image is registered with the metabolic detection image based on the transformation relationship between the images to determine the abnormal metabolism of interest The abnormal organ or body part corresponding to the area, including:
基于所述器官分割图像与所述代谢检测图像之间的转换关系，将所述器官分割图像与所述代谢检测图像进行配准，以得到对所述异常代谢感兴趣区域进行了器官分割、并标识有异常代谢感兴趣区域的图像；或者Based on the conversion relationship between the organ segmentation image and the metabolism detection image, the organ segmentation image and the metabolism detection image are registered to obtain the organ segmentation for the abnormal metabolism region of interest, and An image that identifies a region of interest with abnormal metabolism; or
基于所述身体部位图像与所述代谢检测图像之间的转换关系，将所述身体部位图像与所述代谢检测图像进行配准，以得到对所述异常代谢感兴趣区域进行了身体部位分割、并标识有异常代谢感兴趣区域的图像。Based on the conversion relationship between the body part image and the metabolism detection image, the body part image and the metabolism detection image are registered to obtain the body part segmentation, And identify images with abnormal metabolic regions of interest.
第二方面，本发明还提供一种医学图像处理装置，包括：In a second aspect, the present invention also provides a medical image processing device, comprising:
图像获取模块，用于获取目标对象的扫描图像；The image acquisition module is used to acquire the scanned image of the target object;
异常检测模块，用于基于所述扫描图像，获取所述目标对象的代谢检测图像以及器官分割图像和身体部位图像中的至少一个，其中，所述代谢检测图像标识有异常代谢感兴趣区域；an abnormality detection module, configured to acquire, based on the scanned image, a metabolic detection image of the target object, and at least one of an organ segmentation image and a body part image, wherein the metabolic detection image is marked with a region of interest for abnormal metabolism;
位置确定模块，用于根据所述代谢检测图像以及所述器官分割图像和所述身体部位图像中的至少一个，确定与所述异常代谢感兴趣区域对应的器官和/或身体部位。A position determination module, configured to determine an organ and/or body part corresponding to the abnormal metabolism region of interest according to the metabolic detection image and at least one of the organ segmentation image and the body part image.
第三方面，本发明还提供一种电子设备，包括：处理器和存储器；In a third aspect, the present invention also provides an electronic device, comprising: a processor and a memory;
所述存储器上存储有可被所述处理器执行的计算机可读程序；A computer-readable program executable by the processor is stored on the memory;
所述处理器执行所述计算机可读程序时实现如下步骤：When the processor executes the computer-readable program, the following steps are implemented:
获取目标对象的拍摄图像，所述拍摄图像至少包括第一模态图像和第二模态图像；acquiring a captured image of the target object, where the captured image at least includes a first modality image and a second modality image;
基于所述第一模态图像获取所述目标对象的代谢检测图像，其中，所述代谢检测图像标识有异常代谢感兴趣区域；Acquiring a metabolic detection image of the target object based on the first modality image, wherein the metabolic detection image identifies a region of interest with abnormal metabolism;
基于所述第二模态图像确定所述器官分割图像和/或身体部位图像；determining the organ segmentation image and/or the body part image based on the second modality image;
根据所述代谢检测图像、所述器官分割图像和/或所述身体部位图像，确定与所述异常代谢感兴趣区域对应的器官和/或身体部位，并生成文本诊断报告。According to the metabolic detection image, the organ segmentation image and/or the body part image, the organ and/or the body part corresponding to the abnormal metabolism region of interest is determined, and a text diagnosis report is generated.
第四方面，本发明还提供了一种计算机可读存储介质，所述计算机可读存储介质存储有一个或者多个程序，所述一个或者多个程序可被一个或者多个处理器执行，以实现如下步骤：In a fourth aspect, the present invention further provides a computer-readable storage medium, where the computer-readable storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to Implement the following steps:
获取目标对象的扫描图像和光学图像，所述扫描图像至少包括第一模态图像和第二模态图像；acquiring a scanned image and an optical image of the target object, the scanned image including at least a first modality image and a second modality image;
基于所述第一模态图像获取所述目标对象的代谢检测图像，其中，所述代谢检测图像标识有异常代谢感兴趣区域；Acquiring a metabolic detection image of the target object based on the first modality image, wherein the metabolic detection image identifies a region of interest with abnormal metabolism;
基于所述第二模态图像确定所述器官分割图像；determining the organ segmentation image based on the second modality image;
基于所述光学图像确定所述身体部位图像；determining the body part image based on the optical image;
根据所述代谢检测图像、所述器官分割图像和所述身体部位图像，确定与所述异常代谢感兴趣区域对应的器官和身体部位，并生成文本诊断报告。According to the metabolic detection image, the organ segmentation image and the body part image, the organ and body part corresponding to the abnormal metabolism region of interest are determined, and a text diagnosis report is generated.
与现有技术相比，本发明提供的医学图像处理系统、装置、电子设备及存储介质，首先获取扫描图像，然后基于扫描图像，得到目标对象的代谢检测图像以及器官分割图像和身体部位图像中的至少一个，其中代谢检测图像标识有异常代谢感兴趣区域，器官分割图像对目标对象进行了器官分割，身体部位图像对目标对象进行了身体部位分割，最后将代谢检测图像与器官分割图像和身体部位图像中的至少一个进行融合处理，进而可以确定出与所述异常代谢感兴趣区域对应的器官和/或身体部位，无需医生进行人工判断感兴趣区域所属的身体部位以及其具体的器官后，手动进行图像中感兴趣区域所属的身体部位以及其具体的器官的划分，加快了病灶识别的过程，提高病灶识别效率，减轻了医生的工作，而且，无需专业人员进行图像识别判断，普通人员即可根据最后的融合图像看出目标对象的病灶器官和病灶所属身体部位，给病人也提供了方便。Compared with the prior art, the medical image processing system, device, electronic device and storage medium provided by the present invention first acquire a scanned image, and then, based on the scanned image, obtain a metabolic detection image of the target object, an organ segmentation image and a body part image. At least one of the target objects is identified by the metabolic detection image with abnormal metabolic regions of interest, the target object is segmented by the organ segmentation image, and the target object is segmented by the body part image, and finally the metabolic detection image is combined with the organ segmentation image and the body. At least one of the part images is fused, so that the organ and/or body part corresponding to the abnormal metabolism region of interest can be determined, without the need for a doctor to manually determine the body part to which the region of interest belongs and its specific organ, Manually divide the body parts and specific organs of the region of interest in the image, which speeds up the process of lesion identification, improves the efficiency of lesion identification, and reduces the work of doctors. The lesion organ of the target object and the body part to which the lesion belongs can be seen according to the final fusion image, which also provides convenience for the patient.
附图说明Description of drawings
图1是本发明提供的医学图像处理系统的一实施例的流程图；1 is a flowchart of an embodiment of a medical image processing system provided by the present invention;
图2是本发明提供的医学图像处理系统中，步骤S200的一实施例的流程图；2 is a flowchart of an embodiment of step S200 in the medical image processing system provided by the present invention;
图3是本发明提供的医学图像处理系统中，步骤S300的一实施例的流程图；3 is a flowchart of an embodiment of step S300 in the medical image processing system provided by the present invention;
图4是本发明提供的医学图像处理系统中，步骤S320的一实施例的流程图；4 is a flowchart of an embodiment of step S320 in the medical image processing system provided by the present invention;
图5本发明提供的医学图像处理装置的一实施例的示意图；5 is a schematic diagram of an embodiment of a medical image processing apparatus provided by the present invention;
图6是本发明医学图像处理程序的一实施例的运行环境示意图。FIG. 6 is a schematic diagram of a running environment of an embodiment of a medical image processing program of the present invention.
具体实施方式Detailed ways
为了使本发明的目的、技术方案及优点更加清楚明白，以下结合附图及实施例，对本发明进行进一步详细说明。应当理解，此处所描述的具体实施例仅仅用以解释本发明，并不用于限定本发明。In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention, but not to limit the present invention.
本发明所涉及的医学图像处理系统、装置、设备或者计算机可读存储介质可用于医学领域中的X射线计算机断层造影系统(CT系统)、正电子发射型断层显像系统(PET系统)、正电子发射计算机断层显像-X射线计算机断层造影多模态混合系统(PET-CT系统)、单光子发射断层造影系统(SPECT-CT系统)、正电子发射型断层显像-磁共振成像系统(PET-MRI系统)等医学成像系统中。本发明所涉及的系统、装置、设备或者计算机可读存储介质既可以与上述系统集成在一起，也可以是相对独立的。The medical image processing system, device, device or computer-readable storage medium involved in the present invention can be used in X-ray computed tomography systems (CT systems), positron emission tomography systems (PET systems), positive Electron emission computed tomography-X-ray computed tomography multimodal hybrid system (PET-CT system), single photon emission tomography system (SPECT-CT system), positron emission tomography-magnetic resonance imaging system ( PET-MRI system) and other medical imaging systems. The system, apparatus, device or computer-readable storage medium involved in the present invention can either be integrated with the above-mentioned system, or can be relatively independent.
本实施例提供了一种医学图像处理系统，可设置于医学成像系统中，医学成像系统可以为X射线计算机断层造影系统(CT系统)、正电子发射型断层显像系统(PET系统)、正电子发射计算机断层显像-X射线计算机断层造影多模态混合系统(PET-CT系统)、单光子发射断层造影系统(SPECT-CT系统)、正电子发射型断层显像-磁共振成像系统(PET-MRI系统)等，医学图像处理系统所执行的方法具体可由该系统的一个或者多个处理器执行。图1是本发明实施例提供的医学图像处理系统执行的方法的流程图，请参阅图1，医学图像处理系统处理器和存储器；存储器上存储有可被处理器执行的计算机可读程序；处理器执行计算机可读程序时实现如下步骤：This embodiment provides a medical image processing system, which can be set in a medical imaging system. The medical imaging system can be an X-ray computed tomography system (CT system), a positron emission tomography system (PET system), a positive Electron emission computed tomography-X-ray computed tomography multimodal hybrid system (PET-CT system), single photon emission tomography system (SPECT-CT system), positron emission tomography-magnetic resonance imaging system ( PET-MRI system), etc., the method performed by the medical image processing system can be specifically performed by one or more processors of the system. 1 is a flowchart of a method performed by a medical image processing system provided by an embodiment of the present invention. Please refer to FIG. 1, a medical image processing system processor and memory; the memory stores a computer-readable program that can be executed by the processor; processing When the computer executes the computer-readable program, it implements the following steps:
S100、获取目标对象的扫描图像；S100, acquiring a scanned image of the target object;
S200、基于扫描图像，获取目标对象的代谢检测图像以及器官分割图像和身体部位图像中的至少一个，其中，代谢检测图像标识有异常代谢感兴趣区域；S200. Based on the scanned image, obtain a metabolic detection image of the target object and at least one of an organ segmentation image and a body part image, wherein the metabolic detection image is marked with an abnormal metabolic region of interest;
S300、根据代谢检测图像以及器官分割图像和身体部位图像中的至少一个，确定与异常代谢感兴趣区域对应的器官和/或身体部位。S300. Determine the organ and/or body part corresponding to the abnormal metabolism region of interest according to the metabolic detection image and at least one of the organ segmentation image and the body part image.
本实施例中，首先获取扫描图像，然后基于扫描图像，得到目标对象的代谢检测图像以及器官分割图像和身体部位图像中的至少一个，其中代谢检测图像标识有异常代谢感兴趣区域，器官分割图像对目标对象进行了器官分割，身体部位图像对目标对象进行了身体部位分割，最后将代谢检测图像与器官分割图像和身体部位图像中的至少一个进行融合处理，进而可以确定出与所述异常代谢感兴趣区域对应的器官和/或身体部位，无需医生进行人工判断感兴趣区域所属的身体部位以及其具体的器官后，手动进行图像中感兴趣区域所属的身体部位以及其具体的器官的划分，加快了病灶识别的过程，提高病灶识别效率，减轻了医生的工作，而且，无需专业人员进行图像识别判断，普通人员即可根据最后的融合图像看出目标对象的病灶器官和病灶所属身体部位，给病人也提供了方便。In this embodiment, a scanned image is obtained first, and then based on the scanned image, a metabolic detection image of the target object and at least one of an organ segmentation image and a body part image are obtained, wherein the metabolic detection image is marked with an abnormal metabolic region of interest, and the organ segmentation image The target object is subjected to organ segmentation, the body part image is subjected to body part segmentation, and finally the metabolic detection image is fused with at least one of the organ segmentation image and the body part image, and then it can be determined that the abnormal metabolism is related to the abnormal metabolism. For the organs and/or body parts corresponding to the region of interest, there is no need for the doctor to manually determine the body parts and specific organs to which the region of interest belongs, and then manually classify the body parts and specific organs of the region of interest in the image. The process of lesion identification is accelerated, the efficiency of lesion identification is improved, and the work of doctors is reduced. Moreover, without the need for professionals to perform image recognition and judgment, ordinary personnel can see the target object's lesion organ and the body part to which the lesion belongs based on the final fusion image. It is also convenient for patients.
在一些实施例中，当代谢检测图像与器官分割图像和身体部位图像中的至少一个进行融合处理时，如果图像之间的成像区域不一致，即在一种情况下，PET图像和CT图像重建时，径向范围(FOV，Field ofView)不一致，例如PET扫描径向重建范围为700mm，能够覆盖人体全部区域，但CT径向重建范围为500mm，对于人体胳膊部分不能完全覆盖，但CT扫描仍能采集这一范围的数据，当感兴趣区域在上述CT径向重建范围之外时，可以使用扩展FOV的算法，预测CT扫描中500mm以外区域的重建图像，进而确定感兴趣区域所在的部位。在另一种情况下，当CT在轴向的扫描范围(AFOV，Axial Field of View)与PET扫描不一致时，例如PET扫描范围为头部到膝盖位置，而CT扫描范围为头部到大腿位置，此时扩展FOV功能很难预测到两种扫描方式下不重合区域的CT图像，可通过定位算法推测感兴趣区域所在的部位或器官；对于肝部等比较大的器官，可以仅通过PET图像做定位，确定感兴趣区域所在的器官。In some embodiments, when the metabolic detection image is fused with at least one of the organ segmentation image and the body part image, if the imaging area between the images is inconsistent, that is, in one case, when the PET image and the CT image are reconstructed , the radial range (FOV, Field of View) is inconsistent. For example, the radial reconstruction range of PET scan is 700mm, which can cover the whole area of the human body, but the radial reconstruction range of CT is 500mm, which cannot completely cover the arm part of the human body, but CT scan can still Collecting data in this range, when the region of interest is outside the CT radial reconstruction range, the extended FOV algorithm can be used to predict the reconstructed image of the region beyond 500mm in the CT scan, and then determine the location of the region of interest. In another case, when the scanning range of CT in the axial direction (AFOV, Axial Field of View) is inconsistent with that of PET scanning, for example, the scanning range of PET is from the head to the knee, while the scanning range of CT is from the head to the thigh. At this time, it is difficult to predict the CT images of the non-overlapping areas under the two scanning modes by the extended FOV function. The location or organ where the region of interest is located can be inferred by the positioning algorithm; for relatively large organs such as the liver, the PET image can be used only. Do localization to identify the organ where the region of interest is located.
在一些实施例中，获取目标对象的拍摄图像，拍摄图像包括第一模态图像和第二模态图像，其中，第一模态图像为PET图像，第二模态图像为CT图像。基于PET图像获取目标对象标识有异常代谢感兴趣区域的代谢检测图像，根据代谢检测图像与CT成像图像，确定异常代谢感兴趣区域对应的器官，并生成文本诊断报告；根据代谢检测图像与CT定位片图像，确定异常代谢感兴趣区域对应的身体部位，并生成文本诊断报告。In some embodiments, a captured image of the target object is acquired, and the captured image includes a first modality image and a second modality image, wherein the first modality image is a PET image, and the second modality image is a CT image. Based on the PET image, the target object is identified with a metabolic detection image with an abnormal metabolic region of interest, and according to the metabolic detection image and CT imaging image, the organ corresponding to the abnormal metabolic region of interest is determined, and a text diagnosis report is generated; according to the metabolic detection image and CT positioning Slice images, identify body parts corresponding to abnormal metabolic regions of interest, and generate textual diagnostic reports.
在一些实施例中，获取目标对象的扫描图像和光学图像，拍摄图像包括第一模态图像和第二模态图像，其中，第一模态图像为PET图像，第二模态图像为CT图像。基于PET图像获取目标对象标识有异常代谢感兴趣区域的代谢检测图像，根据代谢检测图像与CT成像图像，确定异常代谢感兴趣区域对应的器官，并生成文本诊断报告；根据代光学图像，确定异常代谢感兴趣区域对应的身体部位，并生成文本诊断报告。In some embodiments, a scanned image and an optical image of the target object are acquired, and the captured image includes a first modality image and a second modality image, wherein the first modality image is a PET image, and the second modality image is a CT image . Based on the PET image, the target object is marked with a metabolic detection image with an abnormal metabolic region of interest. According to the metabolic detection image and the CT imaging image, the organ corresponding to the abnormal metabolic region of interest is determined, and a text diagnosis report is generated; according to the optical image, the abnormality is determined. Metabolize body parts corresponding to regions of interest and generate textual diagnostic reports.
可选的，步骤S100中，目标对象包括但不限于人体、人体的组织器官、动物等。Optionally, in step S100, the target object includes but is not limited to the human body, tissues and organs of the human body, animals, and the like.
在一些实施例中，步骤S100中，扫描图像可以包括但不限于PET图像、CT图像、MR图像、定位片中的至少一个。其中，PET图像或CT图像可用于进行代谢检测图像的获取，CT图像或MR图像可用于进行器官分割图像，定位片可用于进行身体部位分割，例如，将CT图像进行器官分割，可得到器官分割图像，CT图像或者MR获取之前的第一张像图为定位片，定位片由于成像组织结构并没有那么清晰，只能看到轮廓，故可用于身体部位分割，进行身体部位图像的获取。PET图像通过PET设备获得，CT图像通过CT设备获得，MR图像通过MR设备获取。其中，定位片一般可以反映患者的整体概略信息，通常可以是患者的冠状位图像或者矢状位图像。定位片的扫描时间短，定位片上一般有定位框，借助定位框和平面图像可判断这一区域涵盖的结构。定位片能够明确扫描范围、角度、扫描参数、延迟时间等,制定出详细的扫描计划,为诊断的准确性提供保障。因此，通过定位片可以方便的对目标对象的身体部位进行分割，例如将人体部位分割为头部、颈部、胸部、腹部、腿部等等。CT图像反应了不同器官的CT值，不同器官的CT值不同，因此，可方便的进行器官分割，例如将人体器官分割为肝(具体到肝分段)、肾、肺(具体到肺分叶)、肋骨(具体到第几肋骨)、椎骨(具体到第几椎骨)等等。需要说明的是，定位片是在CT图像或者MR图像之前获取，定位片的获取方式包括但不限于CT设备拍摄、MR设备拍摄或者深度相机拍摄。In some embodiments, in step S100, the scanned image may include, but is not limited to, at least one of a PET image, a CT image, an MR image, and a localization slice. Among them, PET images or CT images can be used for obtaining metabolic detection images, CT images or MR images can be used for organ segmentation images, and localization slices can be used for body part segmentation. For example, organ segmentation can be obtained by performing organ segmentation on CT images. The first image before the image, CT image or MR acquisition is the positioning slice. Because the imaging tissue structure is not so clear, only the outline can be seen, so it can be used for body part segmentation and body part image acquisition. PET images are acquired by PET equipment, CT images are acquired by CT equipment, and MR images are acquired by MR equipment. Among them, the positioning slice can generally reflect the overall general information of the patient, and can usually be a coronal image or a sagittal image of the patient. The scanning time of the positioning sheet is short, and there is generally a positioning frame on the positioning sheet. With the help of the positioning frame and the plane image, the structure covered by this area can be determined. The positioning film can clarify the scanning range, angle, scanning parameters, delay time, etc., and formulate a detailed scanning plan to provide a guarantee for the accuracy of diagnosis. Therefore, the body parts of the target object can be easily segmented through the localization slice, for example, the body parts are segmented into head, neck, chest, abdomen, legs, and so on. CT images reflect the CT values of different organs, and the CT values of different organs are different. Therefore, organ segmentation can be easily performed. For example, human organs can be segmented into liver (specific to liver segments), kidneys, and lungs (specific to lung lobes). ), ribs (specific to which rib), vertebrae (specific to which vertebra), etc. It should be noted that the positioning slice is obtained before the CT image or the MR image, and the acquisition method of the positioning slice includes, but is not limited to, shooting with CT equipment, shooting with MR equipment, or shooting with a depth camera.
在一些实施例中，为了增加图像识别的准确率，在获取了扫描图像后，对扫描图像进行预处理，预处理的过程包括但不限于降噪、滤波、灰度二值化处理、归一化增强处理等处理过程，以得到更为清晰的图像，以方便后续的代谢异常区域识别、器官分割以及身体部位分割。In some embodiments, in order to increase the accuracy of image recognition, after the scanned image is acquired, the scanned image is preprocessed, and the preprocessing process includes but is not limited to noise reduction, filtering, grayscale binarization, normalization Processes such as chemical enhancement processing are used to obtain clearer images, which are convenient for subsequent identification of abnormal metabolic regions, organ segmentation, and body part segmentation.
步骤S200是为了基于PET图像、CT图像、MR图像和/或定位片来得到目标对象的代谢检测图像以及器官分割图像和身体部位图像中的至少一个。在一些实施例中，代谢检测图像基于PET图像或CT图像获取，器官分割图像基于CT图像或者MR图像获取，身体部位图像基于定位片获取。Step S200 is to obtain at least one of a metabolic detection image and an organ segmentation image and a body part image of the target object based on the PET image, CT image, MR image and/or localization slice. In some embodiments, metabolic detection images are acquired based on PET images or CT images, organ segmentation images are acquired based on CT images or MR images, and body part images are acquired based on localization slices.
在一些实施例中，请参阅图2，步骤S200具体包括：In some embodiments, referring to FIG. 2 , step S200 specifically includes:
S210、基于PET图像或CT图像，获取目标对象的异常代谢感兴趣区域，以得到标识有异常代谢感兴趣区域的代谢检测图像；S210, based on the PET image or the CT image, obtain the abnormal metabolism region of interest of the target object, so as to obtain a metabolic detection image marked with the abnormal metabolism region of interest;
S220、基于CT图像或MR图像，对目标对象进行器官分割，获取目标对象的器官分割图像和/或身体部位图像；S220. Perform organ segmentation on the target object based on the CT image or MR image, and obtain the organ segmentation image and/or the body part image of the target object;
S230、基于定位片，对目标对象进行身体部位分割，以得到所述目标对象的身体部位图像。S230. Based on the positioning slice, perform body part segmentation on the target object to obtain a body part image of the target object.
具体的，步骤S210至步骤S230可以同时或者依次进行，本发明实施例对此三个步骤的执行顺序不做限定，图2仅仅示出了一种执行的方式，并不代表本发明实施例必须按照该执行顺序执行。此外，除步骤S210外，步骤S220和步骤S230可以只执行其中任意一个。Specifically, steps S210 to S230 may be performed simultaneously or sequentially. The embodiment of the present invention does not limit the execution order of the three steps. FIG. 2 only shows an execution method, which does not mean that the embodiment of the present invention must Execute in this execution order. In addition, in addition to step S210, only any one of steps S220 and S230 may be performed.
步骤S210中，根据PET图像或CT图像检测出目标对象的异常代谢感兴趣区域，异常代谢感兴趣区域可以为发生病变的异常代谢区域，例如肿瘤病灶；异常代谢感兴趣区域还可以为对示踪剂摄取量较高的正常身体组织，如肝、脑、膀胱、肾、心肌等。在药物代谢过程中，异常区域与周围正常的身体组织相比，对药物反应不同。本发明检测异常代谢感兴趣区域的方式有多种，例如，通过药物代谢指标来判断出异常代谢感兴趣区域，其具体过程为：根据药物代谢情况设置阈值，从而得到预设的医学参数阈值，然后根据身体部位对药物的代谢情况与医学参数阈值的对比结果识别出异常代谢感兴趣区域。此外，还可以通过深度学习算法对PET图像或CT图像中的异常位置进行识别。其具体过程为：深度学习算法在经过训练后，可以直接从输入的PET图像或CT图像中识别出异常代谢感兴趣区域，具体的，深度学习算法的训练过程为：获取训练集并构建深度学习模型，其中，训练集包括输入参数和输出参数，输入参数可以为PET图像或CT图像，输出参数可以为经过人工进行异常代谢感兴趣区域划分的代谢检测图像，然后采用所述训练集对深度学习模型进行训练，以得到训练完备的深度学习模型。检测异常代谢感兴趣区域的方式可根据实际情况自由选择，本发明实施例对此不做限定。In step S210, a region of interest for abnormal metabolism of the target object is detected according to the PET image or CT image, and the region of interest for abnormal metabolism may be an abnormal metabolism region with lesions, such as tumor lesions; the region of interest for abnormal metabolism may also be a tracer. Normal body tissues with high dose intake, such as liver, brain, bladder, kidney, myocardium, etc. During drug metabolism, abnormal areas respond differently to drugs than surrounding normal body tissue. There are various ways of detecting the abnormal metabolism region of interest in the present invention. For example, the abnormal metabolism region of interest is determined by the drug metabolism index, and the specific process is: setting a threshold value according to the drug metabolism situation, so as to obtain a preset medical parameter threshold value, Regions of interest for abnormal metabolism are then identified based on the comparison of the metabolism of the drug by the body part with the thresholds of medical parameters. In addition, abnormal locations in PET images or CT images can also be identified by deep learning algorithms. The specific process is as follows: after the deep learning algorithm is trained, it can directly identify the abnormal metabolic region of interest from the input PET image or CT image. Specifically, the training process of the deep learning algorithm is: obtaining a training set and constructing a deep learning algorithm. model, wherein the training set includes input parameters and output parameters, the input parameters can be PET images or CT images, and the output parameters can be metabolic detection images that have been manually divided into regions of interest for abnormal metabolism, and then the training set is used for deep learning. The model is trained to obtain a fully trained deep learning model. The method of detecting the abnormal metabolism region of interest can be freely selected according to the actual situation, which is not limited in the embodiment of the present invention.
步骤S220中，当采用CT图像进行器官分割时，由于人体器官或组织是由多种物质成分和不同的密度构成的，因此在X线穿透人体器官或组织时，各点对X线的吸收系数是不同的。因此反映在CT图像上时不同器官会有不同的HU值，加之CT的分辨率相对较高，因此可以把各个器官分割出来，其具体实现方式有很多，例如，根据预设的CT值阈值来进行器官分割，亦可以通过深度学习算法来进行器官分割，其具体过程为：深度学习算法在经过训练后，可以直接从输入的CT图像中识别出器官。器官分割的方式可根据实际情况自由选择，本发明实施例对此不做限定。In step S220, when the CT image is used for organ segmentation, since human organs or tissues are composed of a variety of material components and different densities, when the X-ray penetrates the human organ or tissue, the absorption of X-rays at each point will be reduced. The coefficients are different. Therefore, different organs will have different HU values when reflected on the CT image. In addition, the resolution of CT is relatively high, so each organ can be segmented. There are many specific implementation methods. For example, according to the preset CT value threshold Organ segmentation can also be performed through a deep learning algorithm. The specific process is as follows: after the deep learning algorithm is trained, it can directly identify the organ from the input CT image. The manner of organ segmentation can be freely selected according to the actual situation, which is not limited in this embodiment of the present invention.
步骤S220中，当采用MR图像进行器官分割时，由于人体不同的组织在不同的磁场中表现出来的信号不一致，因此呈现在MR图像上就会有差异，MR可以设置很多序列，很多参数，在不同的序列和参数中组织体现在MR图像上也不一样，故也可以利用MR图像进行器官分割。In step S220, when the MR image is used for organ segmentation, since different tissues of the human body show inconsistent signals in different magnetic fields, there will be differences in the MR images. Many sequences and parameters can be set for MR. Different sequences and parameters show different tissues in MR images, so MR images can also be used for organ segmentation.
步骤S230中，定位片用于进行身体部位分割。在一些实施例中，可采用图像语义分割模型实现身体部位的分割，例如将目标物体分割为头部、胸部、腹部、盆腔部和下肢部，将定位片根据身体部位分开，其中，图像语义分割模型可采用FCN模型(全卷积网络模型)、U-NET神经网络模型等，本发明对其实现方式不做限定。In step S230, the positioning patch is used for body part segmentation. In some embodiments, an image semantic segmentation model can be used to achieve the segmentation of body parts, for example, the target object is divided into head, chest, abdomen, pelvis and lower limbs, and the localization slices are separated according to body parts, wherein the image semantic segmentation The model may adopt an FCN model (full convolution network model), a U-NET neural network model, etc., and the present invention does not limit its implementation.
步骤S300是为了将代谢检测图像与器官分割图像和/或身体部位图像融合，进而确定出与异常代谢感兴趣区域对应的异常器官和/或身体部位。在一些实施例中，请参阅图3，步骤S300具体包括：Step S300 is to fuse the metabolic detection image with the organ segmentation image and/or the body part image, so as to determine the abnormal organ and/or body part corresponding to the abnormal metabolism region of interest. In some embodiments, referring to FIG. 3 , step S300 specifically includes:
S310、获取图像之间的转换关系；S310, acquiring the conversion relationship between the images;
S320、基于图像之间的转换关系，将器官分割图像和/或身体部位图像与代谢检测图像进行配准，以确定与异常代谢感兴趣区域对应的器官和/或身体部位。S320. Based on the conversion relationship between the images, register the organ segmentation image and/or the body part image with the metabolic detection image to determine the organ and/or body part corresponding to the abnormal metabolic region of interest.
本实施例中，由于代谢检测图像与器官分割图像和/或身体部位图像是由不同的设备获取，因此，在将图像进行融合时，需要先获取图像之间的配准关系，才能在代谢检测图像中标识出目标对象的器官以及身体部位。通过将代谢检测图像与器官分割图像和/或身体部位图像进行配准后，即可得到进行了器官分割和/或身体部位分割并标识有异常代谢感兴趣区域的图像，通过此图像即可快速的确定异常代谢感兴趣区域对应的异常器官，更进一步，可以快速确定异常器官所属的身体部位，例如确定目标对象的异常位置在胸部的左肺上叶上或者在腹部的第五腰椎上。In this embodiment, since the metabolic detection image and the organ segmentation image and/or the body part image are acquired by different devices, when the images are fused, the registration relationship between the images needs to be acquired first before the metabolic detection image can be used in the metabolic detection. Organs and body parts of the target object are identified in the image. By registering the metabolic detection image with the organ segmentation image and/or the body part image, an image with organ segmentation and/or body part segmentation and marked with abnormal metabolic regions of interest can be obtained. By determining the abnormal organ corresponding to the abnormal metabolism region of interest, it can quickly determine the body part to which the abnormal organ belongs.
在一些实施例中，请参阅图4，步骤S320具体包括：In some embodiments, referring to FIG. 4 , step S320 specifically includes:
S321、将身体部位图像与器官分割图像进行配准，以生成第一图像，其中，第一图像为对目标对象的身体部位和器官进行分割后的图像；S321, registering the body part image and the organ segmentation image to generate a first image, wherein the first image is an image obtained by segmenting the body part and the organ of the target object;
S322、基于图像之间的转换关系，将代谢检测图像与第一图像进行配准，以生成第二图像后，基于第二图像确定与异常代谢感兴趣区域对应的器官和身体部位，其中，第二图像为对目标对象进行了器官分割以及身体部位分割后，并标识有异常代谢感兴趣区域的图像。S322. Based on the conversion relationship between the images, register the metabolic detection image with the first image to generate the second image, and then determine the organ and body part corresponding to the abnormal metabolic region of interest based on the second image, wherein the first The second image is an image of the target object after organ segmentation and body part segmentation, and marked with abnormal metabolism regions of interest.
本实施例中，在进行配准融合时，由于身体部位图像和器官分割图像为同一设备获取，因此，可直接进行融合得到第一图像然后将第一图像与代谢检测图像进行配准融合得到第二图像，当然，在其它的实施例中，还可以先将器官分割图像与代谢检测图像融合得到一中间图像后，然后将中间图像与身体部位图像进行融合得到第二图像，本发明实施例对其具体融合的顺序不作限定，只要能够生成第二图像的技术方案均在本发明的保护范围中。In this embodiment, when performing registration and fusion, since the body part image and the organ segmentation image are obtained by the same device, the first image can be obtained by direct fusion, and then the first image and the metabolic detection image can be registered and fused to obtain the first image. Two images, of course, in other embodiments, an intermediate image may be obtained by first fusing the organ segmentation image and the metabolic detection image, and then the intermediate image and the body part image may be fused to obtain the second image. The specific fusion sequence is not limited, as long as the technical solution capable of generating the second image is within the protection scope of the present invention.
在一些实施例中，步骤S322具体包括：In some embodiments, step S322 specifically includes:
对第一图像进行预处理，得到一组代表所述扫描对象的体素；Preprocessing the first image to obtain a set of voxels representing the scanned object;
根据各个体素的图像坐标以及第一图像的标签数据，计算出每个体素在CT物理坐标系中的空间坐标，以得到一初始点云模型；According to the image coordinates of each voxel and the label data of the first image, the spatial coordinates of each voxel in the CT physical coordinate system are calculated to obtain an initial point cloud model;
根据CT物理坐标系和PET物理坐标系之间的转换关系，对初始点云模型进行转换，以生成第一点云模型后；According to the conversion relationship between the CT physical coordinate system and the PET physical coordinate system, the initial point cloud model is converted to generate the first point cloud model;
根据第一图像和第一点云模型生成配准图像；generating a registration image according to the first image and the first point cloud model;
将配准图像和代谢检测图像进行融合，以得到第二图像。The registration image and the metabolic detection image are fused to obtain a second image.
本实施例中，第一图像为通过CT图像获取，为了得到代表扫描对象的点云数据，需先将第一图像进行预处理，然后再通过多个坐标转换后，得到第一点云模型。其中，预处理的过程可以是为了更精准的提取出一组代表扫描对象的体素，具体的，预处理的过程可以为：对器官分割图像至少进行二值化和/或轮廓提取处理，以得到代表扫描对象的体素。其中，二值化处理的过程是为了实现对器官分割图像的灰度处理，轮廓提取处理是为了提取出器官分割图像中扫描对象的轮廓，在具体实施时，可采用图像梯度算法提取出扫描对象的轮廓，进而可以更精准的得到代表扫描对象的体素。In this embodiment, the first image is obtained from a CT image. In order to obtain point cloud data representing the scanned object, the first image needs to be preprocessed, and then a first point cloud model is obtained after multiple coordinate transformations. The preprocessing process may be to more accurately extract a set of voxels representing the scanned object. Specifically, the preprocessing process may be: at least perform binarization and/or contour extraction processing on the organ segmentation image, so as to Get the voxels representing the scanned object. Among them, the process of binarization processing is to realize the grayscale processing of the organ segmentation image, and the contour extraction processing is to extract the contour of the scanned object in the organ segmentation image. In the specific implementation, the image gradient algorithm can be used to extract the scanned object. The contour of the scanned object can be obtained more accurately.
每个体素都有代表其位置的坐标，此坐标为图像坐标系下的坐标，为了和PET图像进行配准，需要先将图像坐标系下的图像坐标转换为PET物理坐标系下的坐标，其需要经过两次坐标转换，即图像坐标—CT坐标—PET坐标。Each voxel has a coordinate representing its position. This coordinate is the coordinate in the image coordinate system. In order to register with the PET image, it is necessary to convert the image coordinate in the image coordinate system into the coordinate in the PET physical coordinate system. It needs to go through two coordinate transformations, that is, image coordinates - CT coordinates - PET coordinates.
具体的，图像坐标转换为CT物理坐标系下的坐标可利用第一图像的标签数据进行，第一图像一般以DICOM格式保存，DICOM数据具有标签数据，本实施例采用标签数据中的Patient Position-(0018,5100)、Image Position(Patient)-(0020,0032)和ImageOrientation(Patient)-(0020,0037)三个字段来确定进行坐标的转换，以实现扫描对象的空间定位。其中，Image Position表示图像的左上角在患者坐标系下的空间坐标，ImageOrientation表示图像坐标系与患者坐标系对应坐标轴的夹角余弦值，Image Orientation有6个参数，前三个为图像坐标系的X轴与患者坐标系的三个轴之间的夹角余弦，后三个为图像坐标系的Y轴与患者坐标系的三个轴之间的夹角余弦，如果6个参数里边只有0和1或-1，则图像一定与患者坐标系某个平面平行；如果出现小数，则表示不是完全是3个位面中的一个，会有一定夹角。Patient Position描述患者相对于成像设备的位置，说明了患者摆位和进床方式，患者坐标系和CT物理坐标系的原点重合，但是各坐标轴方向会根据患者摆位和进床方式不同而有区别。Patient Position提供了患者坐标系和CT物理坐标系间的转换关系。通过上述三个字段即可将体素的图像坐标转换为体素在CT物理坐标系下的空间坐标，得到一个初始点云模型。Specifically, the transformation of the image coordinates into the coordinates in the CT physical coordinate system can be performed by using the label data of the first image. The first image is generally saved in DICOM format, and the DICOM data has label data. In this embodiment, the Patient Position- (0018,5100), Image Position(Patient)-(0020,0032) and ImageOrientation(Patient)-(0020,0037) are three fields to determine the coordinate transformation to realize the spatial positioning of the scanned object. Among them, Image Position represents the spatial coordinate of the upper left corner of the image in the patient coordinate system, ImageOrientation represents the cosine value of the angle between the image coordinate system and the corresponding coordinate axis of the patient coordinate system, and Image Orientation has 6 parameters, the first three are the image coordinate system. The cosine of the angle between the X axis and the three axes of the patient coordinate system, the last three are the cosine of the angle between the Y axis of the image coordinate system and the three axes of the patient coordinate system, if there are only 0 in the 6 parameters and 1 or -1, the image must be parallel to a certain plane of the patient coordinate system; if there is a decimal, it means that it is not exactly one of the three planes, and there will be a certain angle. Patient Position describes the position of the patient relative to the imaging device, and describes the patient placement and bed approach. The origin of the patient coordinate system and the CT physical coordinate system coincide, but the direction of each coordinate axis will vary according to the patient placement and bed approach. the difference. Patient Position provides the transformation relationship between the patient coordinate system and the CT physical coordinate system. Through the above three fields, the image coordinates of the voxels can be converted into the spatial coordinates of the voxels in the CT physical coordinate system, and an initial point cloud model can be obtained.
进一步的，当得到了初始点云模型后，即可进一步根据CT物理坐标系和PET物理坐标系之间的转换关系，将初始点云模型的各个坐标转换为PET物理坐标系下的坐标，得到第一点云模型。其中，CT物理坐标系和PET物理坐标系之间的转换关系为已知，在进行CT设备和PET设备的系统校准时，即可获取，例如可通过模体的校准得到该转化关系。本实施例中，通过CT物理坐标系和PET物理坐标系之间的旋转矩阵R和平移矩阵T来进行坐标的转换，具体的，CT物理坐标系和PET物理坐标系的坐标转换关系可通过如下公式获取：Further, when the initial point cloud model is obtained, each coordinate of the initial point cloud model can be converted into the coordinates under the PET physical coordinate system according to the conversion relationship between the CT physical coordinate system and the PET physical coordinate system, to obtain: The first point cloud model. The conversion relationship between the CT physical coordinate system and the PET physical coordinate system is known, and can be obtained during system calibration of CT equipment and PET equipment. For example, the conversion relationship can be obtained by phantom calibration. In this embodiment, the coordinate conversion is performed by the rotation matrix R and the translation matrix T between the CT physical coordinate system and the PET physical coordinate system. Specifically, the coordinate conversion relationship between the CT physical coordinate system and the PET physical coordinate system can be obtained through the following Formula to get:
XPET＝R*XCT+T，XPET =R*XCT +T,
其中，XPET表示空间中任一坐标点在PET物理坐标系下的三维坐标，XCT表示空间中同一坐标点在CT物理坐标系下的三维坐标，R表示旋转矩阵，T表示平移矩阵，举例来说，某一体素的空间坐标为(x1,y1,z1)，经过上述公式进行坐标转换后，其被转化为(x2,y2,z2)，多个体素转化后的坐标集合即构成了第一点云模型。Among them, XPET represents the three-dimensional coordinates of any coordinate point in the space under the PET physical coordinate system, XCT represents the three-dimensional coordinates of the same coordinate point in the space under the CT physical coordinate system, R represents the rotation matrix, and T represents the translation matrix. For example For example, the spatial coordinates of a voxel are (x1 , y1 , z1 ), and after the coordinate transformation of the above formula, it is transformed into (x2 , y2 , z2 ), and the transformed The set of coordinates constitutes the first point cloud model.
当得到第一点云模型后，即可对第一图像进行转换，根据第一点云模型和第一图像，生成配准图像，此配准图像的各个像素点的坐标即为PET物理坐标系下的坐标，因此可以直接与代谢检测图像进行融合，基于配准图像的身体部位和器官结果，可以相应的在代谢检测图像中实现器官和身体部位的分割，最终得到对目标对象进行了器官分割以及身体部位分割后，并标识有异常代谢感兴趣区域的第二图像。After the first point cloud model is obtained, the first image can be converted, and a registration image is generated according to the first point cloud model and the first image. The coordinates of each pixel of the registration image are the PET physical coordinate system Therefore, it can be directly fused with the metabolic detection image. Based on the body parts and organ results of the registration image, the organs and body parts can be segmented correspondingly in the metabolic detection image, and finally the target object has been segmented. and a second image after segmentation of body parts and identifying regions of interest with abnormal metabolism.
本发明无需医生进行人工进行异常代谢感兴趣区域所属的身体部位以及其具体的器官的划分，加快了病灶识别的过程，提高病灶识别效率，减轻了医生的工作，而且，无需专业人员进行图像识别判断，普通人员即可根据最后的融合图像看出目标对象的病灶器官和病灶所属身体部位，给病人也提供了方便。The present invention does not require a doctor to manually classify the body part to which the abnormal metabolic interest region belongs and its specific organs, accelerates the process of lesion identification, improves the efficiency of lesion identification, and reduces the work of doctors, and does not require professionals to perform image identification Judgment, ordinary personnel can see the target object's lesion organ and the body part to which the lesion belongs according to the final fusion image, which also provides convenience for the patient.
当然，需要说明的是，当第一图像通过MR图像获取时，可直接通过获取PET坐标系和MR坐标系后，将第一图像和代谢检测图像进行配准融合，进而得到对目标对象进行了器官分割以及身体部位分割后，并标识有异常代谢感兴趣区域的第二图像。Of course, it should be noted that when the first image is obtained through the MR image, the first image and the metabolic detection image can be registered and fused directly after obtaining the PET coordinate system and the MR coordinate system, and then the target object can be obtained by registration and fusion. After organ segmentation and body part segmentation, a second image with regions of interest for abnormal metabolism is identified.
由于代谢检测图像还可以直接与器官分割图像或者身体部位图像直接进行融合，得到对异常代谢感兴趣区域进行了器官分割、并标识有异常代谢感兴趣区域的图像或者对异常代谢感兴趣区域进行了身体部位分割、并标识有异常代谢感兴趣区域的图像，因此，在一些实施例中，步骤S320包括：Since the metabolic detection image can also be directly fused with the organ segmentation image or the body part image, an image with the abnormal metabolism region of interest has been segmented and marked with the abnormal metabolic region of interest, or the abnormal metabolic region of interest has been identified. Body parts are segmented, and images with abnormal metabolic regions of interest are identified. Therefore, in some embodiments, step S320 includes:
基于器官分割图像与代谢检测图像之间的转换关系，将器官分割图像与代谢检测图像进行配准，以得到对目标对象进行了器官分割、并标识有异常代谢感兴趣区域的图像；或者Based on the transformation relationship between the organ segmentation image and the metabolic detection image, register the organ segmentation image with the metabolic detection image to obtain an image in which the target object has been organ-segmented and identified with abnormal metabolic regions of interest; or
基于身体部位图像与代谢检测图像之间的转换关系，将身体部位图像与代谢检测图像进行配准，以得到对目标对象进行了身体部位分割、并标识有异常代谢感兴趣区域的图像。Based on the conversion relationship between the body part image and the metabolic detection image, the body part image and the metabolic detection image are registered to obtain an image that segmented the body part of the target object and marked the abnormal metabolic region of interest.
本实施例中，得到的图像有两张，分别是进行器官分割后的异常代谢感兴趣区域图像和进行身体部位分割后的异常代谢感兴趣区域图像。即通过两张图像来显示异常代谢感兴趣区域对应的器官和身体部位。其中，器官分割图像与代谢检测图像的配准方法、身体部位图像与代谢检测图像的配准方法与上述第一图像和代谢检测图像的配准方法类似，在此不再赘述。In this embodiment, two images are obtained, which are an abnormal metabolism region of interest image after organ segmentation and an abnormal metabolism region of interest image after body part segmentation. That is, two images are used to display the organs and body parts corresponding to the region of interest with abnormal metabolism. The registration method of the organ segmentation image and the metabolic detection image, and the registration method of the body part image and the metabolic detection image are similar to the above-mentioned registration method of the first image and the metabolic detection image, and will not be repeated here.
应该理解的是，虽然图1至图4的流程图中的各个步骤按照箭头的指示依次显示，但是这些步骤并不是必然按照箭头指示的顺序依次执行。除非本文中有明确的说明，这些步骤的执行并没有严格的顺序限制，这些步骤可以以其它的顺序执行。It should be understood that although the steps in the flowcharts of FIG. 1 to FIG. 4 are shown in sequence according to the arrows, these steps are not necessarily executed in the sequence shown by the arrows. Unless explicitly stated herein, the execution of these steps is not strictly limited to the order, and these steps may be performed in other orders.
基于上述医学图像处理系统，本发明实施例还相应的提供一种医学图像处理装置400，请参阅图5，该医学图像处理装置400包括图像获取模块410、异常检测模块420和位置确定模块430。Based on the above medical image processing system, an embodiment of the present invention also provides a medical
图像获取模块410用于获取目标对象的扫描图像。The
异常检测模块420用于基于扫描图像，获取目标对象的代谢检测图像以及器官分割图像和身体部位图像中的至少一个，其中，代谢检测图像标识有异常代谢感兴趣区域。The
位置确定模块430用于根据代谢检测图像以及器官分割图像和身体部位图像中的至少一个，确定与异常代谢感兴趣区域对应的异常器官和/或身体部位。The
本实施例中，首先获取扫描图像，然后基于扫描图像，得到目标对象的代谢检测图像以及器官分割图像和身体部位图像中的至少一个，其中代谢检测图像标识有异常代谢感兴趣区域，器官分割图像对目标对象进行了器官分割，身体部位图像对目标对象进行了身体部位分割，最后将代谢检测图像与器官分割图像和身体部位图像中的至少一个进行融合处理，进而可以确定出与异常代谢感兴趣区域对应的器官和/或身体部位，无需医生进行人工判断感兴趣区域所属的身体部位以及其具体的器官后，手动进行图像中感兴趣区域所属的身体部位以及其具体的器官的划分，加快了病灶识别的过程，提高病灶识别效率，减轻了医生的工作，而且，无需专业人员进行图像识别判断，普通人员即可根据最后的融合图像看出目标对象的病灶器官和病灶所属身体部位，给病人也提供了方便。In this embodiment, a scanned image is obtained first, and then based on the scanned image, a metabolic detection image of the target object and at least one of an organ segmentation image and a body part image are obtained, wherein the metabolic detection image is marked with an abnormal metabolic region of interest, and the organ segmentation image Organ segmentation is performed on the target object, and the body part image is performed on the target object. Finally, the metabolic detection image is fused with at least one of the organ segmentation image and the body part image, and then it can be determined to be interested in abnormal metabolism. The organs and/or body parts corresponding to the regions, without the need for doctors to manually determine the body parts and specific organs of the region of interest, and then manually classify the body parts and specific organs of the region of interest in the image, which speeds up the process. The process of lesion identification improves the efficiency of lesion identification and relieves the doctor's work. Moreover, without the need for professionals to perform image recognition and judgment, ordinary personnel can see the target object's lesion organ and the body part to which the lesion belongs based on the final fusion image. Convenience is also provided.
在一些实施例中，代谢检测图像基于PET图像或CT图像获取，器官分割图像基于CT图像或者MR图像获取，身体部位图像基于定位片获取。In some embodiments, metabolic detection images are acquired based on PET images or CT images, organ segmentation images are acquired based on CT images or MR images, and body part images are acquired based on localization slices.
在一些实施例中，异常检测模块420包括第一图像获取单元、第二图像获取单元以及第三图像获取单元。In some embodiments, the
第一图像获取单元用于基于PET图像或CT图像，获取目标对象的异常代谢感兴趣区域，以得到标识有异常代谢感兴趣区域的代谢检测图像。The first image acquisition unit is configured to acquire the abnormal metabolism region of interest of the target object based on the PET image or the CT image, so as to obtain a metabolic detection image marked with the abnormal metabolism region of interest.
第二图像获取单元用于基于CT图像或MR图像，对所述目标对象进行器官分割，以得到目标对象的器官分割图像；The second image acquisition unit is configured to perform organ segmentation on the target object based on the CT image or the MR image, so as to obtain an organ segmentation image of the target object;
第三图像获取单元用于基于定位片，对目标对象进行身体部位分割，以得到目标对象的身体部位图像。The third image acquisition unit is configured to perform body part segmentation on the target object based on the positioning slice, so as to obtain a body part image of the target object.
在一些实施例中，位置确定模块430包括转换关系获取单元以及确定单元。In some embodiments, the
转换关系获取单元用于获取图像之间的转换关系；The conversion relationship acquiring unit is used to acquire the conversion relationship between the images;
确定单元用于基于图像之间的转换关系，将器官分割图像和/或身体部位图像与代谢检测图像进行配准，以确定与异常代谢感兴趣区域对应的器官和/或身体部位。The determining unit is configured to register the organ segmentation image and/or the body part image with the metabolism detection image based on the conversion relationship between the images, so as to determine the organ and/or body part corresponding to the abnormal metabolic region of interest.
在一些实施例中，确定单元包括第一图像生成子单元和第二图像生成子单元。In some embodiments, the determination unit includes a first image generation subunit and a second image generation subunit.
第一图像生成子单元用于将身体部位图像与器官分割图像进行融合，以生成第一图像，其中，第一图像为对目标对象的身体部位和器官进行分割后的图像；The first image generation subunit is used to fuse the body part image and the organ segmentation image to generate a first image, wherein the first image is an image obtained by segmenting the body parts and organs of the target object;
第二图像生成子单元用于基于图像之间的转换关系，将代谢检测图像与第一图像进行配准，以生成第二图像后，基于第二图像确定与异常代谢感兴趣区域对应的器官和身体部位，其中，第二图像为对目标对象进行了器官分割以及身体部位分割后，并标识有异常代谢感兴趣区域的图像。The second image generation subunit is used for registering the metabolic detection image with the first image based on the conversion relationship between the images, so as to generate the second image, and then determine the organs and regions of interest corresponding to the abnormal metabolism based on the second image. A body part, wherein the second image is an image marked with an abnormal metabolism region of interest after organ segmentation and body part segmentation are performed on the target object.
在一些实施例中，第二图像生成子单元具体用于：In some embodiments, the second image generation subunit is specifically used for:
对第一图像进行预处理，得到一组代表扫描对象的体素；Preprocessing the first image to obtain a set of voxels representing the scanned object;
根据各个体素的图像坐标以及器官分割图像的标签数据，计算出每个体素在CT物理坐标系中的空间坐标，以得到一初始点云模型；According to the image coordinates of each voxel and the label data of the organ segmentation image, the spatial coordinates of each voxel in the CT physical coordinate system are calculated to obtain an initial point cloud model;
根据CT物理坐标系和PET物理坐标系之间的转换关系，对初始点云模型进行转换，以生成第一点云模型后；According to the conversion relationship between the CT physical coordinate system and the PET physical coordinate system, the initial point cloud model is converted to generate the first point cloud model;
根据第一图像和第一点云模型生成配准图像；generating a registration image according to the first image and the first point cloud model;
将配准图像和代谢检测图像进行融合，以得到第二图像。The registration image and the metabolic detection image are fused to obtain a second image.
在一些实施例中，确定单元具体用于：In some embodiments, the determining unit is specifically used to:
基于器官分割图像与代谢检测图像之间的转换关系，将器官分割图像与代谢检测图像进行配准，以得到对异常代谢感兴趣区域进行了器官分割、并标识有异常代谢感兴趣区域的图像；或者Based on the conversion relationship between the organ segmentation image and the metabolic detection image, the organ segmentation image and the metabolic detection image are registered to obtain an image with the abnormal metabolism region of interest segmented and marked with the abnormal metabolism region of interest; or
基于身体部位图像与代谢检测图像之间的转换关系，将身体部位图像与代谢检测图像进行配准，以得到对异常代谢感兴趣区域进行了身体部位分割、并标识有异常代谢感兴趣区域的图像。Based on the conversion relationship between the body part image and the metabolic detection image, the body part image and the metabolic detection image are registered to obtain the image with the body part segmentation of the abnormal metabolic interest area and the abnormal metabolic interest area marked. .
如图6所示，基于上述医学图像处理方法，本发明还相应提供了一种电子设备，该电子设备可以是医学成像系统的控制装置、移动终端、桌上型计算机、笔记本、掌上电脑及服务器等计算设备。该电子设备包括处理器10、存储器20及显示器30。图6仅示出了电子设备的部分组件，但是应理解的是，并不要求实施所有示出的组件，可以替代的实施更多或者更少的组件。As shown in FIG. 6 , based on the above medical image processing method, the present invention also provides an electronic device correspondingly, and the electronic device can be a control device of a medical imaging system, a mobile terminal, a desktop computer, a notebook, a palmtop computer and a server. and other computing equipment. The electronic device includes a
存储器20在一些实施例中可以是该电子设备的内部存储单元，例如电子设备的硬盘或内存。存储器20在另一些实施例中也可以是电子设备的外部存储设备，例如电子设备上配备的插接式硬盘，智能存储卡(Smart Media Card,SMC)，安全数字(Secure Digital,SD)卡，闪存卡(Flash Card)等。进一步地，存储器20还可以既包括电子设备的内部存储单元也包括外部存储设备。存储器20用于存储安装于电子设备的应用软件及各类数据，例如安装电子设备的程序代码等。存储器20还可以用于暂时地存储已经输出或者将要输出的数据。在一实施例中，存储器20上存储有医学图像处理程序40，该医学图像处理程序40可被处理器10所执行，从而实现本申请各实施例的医学图像处理方法。The
处理器10在一些实施例中可以是一中央处理器(Central Processing Unit,CPU)，微处理器或其他数据处理芯片，用于运行存储器20中存储的程序代码或处理数据，例如执行医学图像处理方法等。In some embodiments, the
显示器30在一些实施例中可以是LED显示器、液晶显示器、触控式液晶显示器以及OLED(Organic Light-Emitting Diode，有机发光二极管)触摸器等。显示器30用于显示在所述医学图像异常识别设备的信息以及用于显示可视化的用户界面。电子设备的处理器10、存储器20及显示器30通过系统总线相互通信。In some embodiments, the
在一实施例中，当处理器10执行存储器20中医学图像处理程序40时实现以下步骤：In one embodiment, when the
获取目标对象的扫描图像；Obtain the scanned image of the target object;
基于扫描图像，获取目标对象的代谢检测图像以及器官分割图像和身体部位图像中的至少一个，其中，代谢检测图像标识有异常代谢感兴趣区域；obtaining, based on the scanned image, a metabolic detection image of the target object, and at least one of an organ segmentation image and a body part image, wherein the metabolic detection image identifies an abnormal metabolic region of interest;
根据代谢检测图像以及器官分割图像和身体部位图像中的至少一个，确定与异常代谢感兴趣区域对应的器官和/或身体部位。An organ and/or body part corresponding to the abnormal metabolic region of interest is determined from the metabolic detection image and at least one of the organ segmentation image and the body part image.
在一些实施例中，代谢检测图像基于PET图像或CT图像获取，器官分割图像基于CT图像或者MR图像获取，身体部位图像基于定位片获取。In some embodiments, metabolic detection images are acquired based on PET images or CT images, organ segmentation images are acquired based on CT images or MR images, and body part images are acquired based on localization slices.
在一些实施例中，处理器10执行存储器20中医学图像处理程序40时还实现如下步骤：In some embodiments, the
基于PET图像或CT图像，获取目标对象的异常代谢感兴趣区域，以得到标识有异常代谢感兴趣区域的代谢检测图像；以及Based on the PET image or CT image, the abnormal metabolic region of interest of the target object is acquired to obtain a metabolic detection image with the abnormal metabolic region of interest identified; and
基于CT图像或MR图像对目标对象进行器官分割和/或身体部位分割，以得到目标对象的器官分割图像和/或身体部位图像，和/或Perform organ segmentation and/or body part segmentation on the target object based on the CT image or MR image to obtain an organ segmentation image and/or body part image of the target object, and/or
基于所述定位片，对所述目标对象进行身体部位分割，以得到所述目标对象的身体部位图像。Based on the localization slice, body part segmentation is performed on the target object to obtain a body part image of the target object.
在一些实施例中，处理器10执行存储器20中医学图像处理程序40时还实现如下步骤：In some embodiments, the
获取图像之间的转换关系；Get the conversion relationship between images;
基于图像之间的转换关系，将器官分割图像和/或身体部位图像与代谢检测图像进行配准，以确定与异常代谢感兴趣区域对应的器官和/或身体部位。Based on the transformation relationship between the images, the organ segmentation image and/or the body part image is registered with the metabolic detection image to determine the organ and/or body part corresponding to the abnormal metabolic region of interest.
在一些实施例中，处理器10执行存储器20中医学图像异常识别代谢程序40时还实现如下步骤：In some embodiments, the
将身体部位图像与器官分割图像进行配准，以生成第一图像，其中，第一图像为对目标对象的身体部位和器官进行分割后的图像；registering the body part image and the organ segmentation image to generate a first image, wherein the first image is an image obtained by segmenting the body part and the organ of the target object;
基于图像之间的转换关系，将代谢检测图像与第一图像进行配准，以生成第二图像后，基于第二图像确定与异常代谢感兴趣区域对应的器官和身体部位，其中，第二图像为对目标对象进行了器官分割以及身体部位分割后，并标识有异常代谢感兴趣区域的图像。Based on the conversion relationship between the images, after registering the metabolic detection image with the first image to generate a second image, the organs and body parts corresponding to the abnormal metabolic region of interest are determined based on the second image, wherein the second image After organ segmentation and body part segmentation are performed on the target object, and images with abnormal metabolism regions of interest are identified.
在一些实施例中，处理器10执行存储器20中医学图像处理程序40时还实现如下步骤：In some embodiments, the
对第一图像进行预处理，得到一组代表扫描对象的体素；Preprocessing the first image to obtain a set of voxels representing the scanned object;
根据各个体素的图像坐标以及第一图像的标签数据，计算出每个体素在CT物理坐标系中的空间坐标，以得到一初始点云模型；According to the image coordinates of each voxel and the label data of the first image, the spatial coordinates of each voxel in the CT physical coordinate system are calculated to obtain an initial point cloud model;
根据CT物理坐标系和PET物理坐标系之间的转换关系，对初始点云模型进行转换，以生成第一点云模型后；According to the conversion relationship between the CT physical coordinate system and the PET physical coordinate system, the initial point cloud model is converted to generate the first point cloud model;
根据第一图像和第一点云模型生成配准图像；generating a registration image according to the first image and the first point cloud model;
将配准图像和所述代谢检测图像进行融合，以得到第二图像。The registration image and the metabolic detection image are fused to obtain a second image.
在一些实施例中，处理器10执行存储器20中医学图像处理程序40时还实现如下步骤：In some embodiments, the
基于器官分割图像与代谢检测图像之间的转换关系，将器官分割图像与代谢检测图像进行配准，以得到对异常代谢感兴趣区域进行了器官分割、并标识有异常代谢感兴趣区域的图像；或者Based on the conversion relationship between the organ segmentation image and the metabolic detection image, the organ segmentation image and the metabolic detection image are registered to obtain an image with the abnormal metabolism region of interest segmented and marked with the abnormal metabolism region of interest; or
基于身体部位图像与代谢检测图像之间的转换关系，将身体部位图像与代谢检测图像进行配准，以得到对异常代谢感兴趣区域进行了身体部位分割、并标识有异常代谢感兴趣区域的图像。Based on the conversion relationship between the body part image and the metabolic detection image, the body part image and the metabolic detection image are registered to obtain the image with the body part segmentation of the abnormal metabolic interest area and the abnormal metabolic interest area marked. .
综上所述，本发明提供的医学图像处理系统、装置、电子设备及存储介质，首先获取扫描图像，然后基于扫描图像，得到目标对象的代谢检测图像以及器官分割图像和身体部位图像中的至少一个，其中代谢检测图像标识有异常代谢感兴趣区域，器官分割图像对目标对象进行了器官分割，身体部位图像对目标对象进行了身体部位分割，最后将代谢检测图像与器官分割图像和身体部位图像中的至少一个进行融合处理，进而可以确定出与异常代谢感兴趣区域对应的器官和/或身体部位，无需医生进行人工判断感兴趣区域所属的身体部位以及其具体的器官后，手动进行图像中感兴趣区域所属的身体部位以及其具体的器官的划分，加快了病灶识别的过程，提高病灶识别效率，减轻了医生的工作，而且，无需专业人员进行图像识别判断，普通人员即可根据最后的融合图像看出目标对象的病灶器官和病灶所属身体部位，给病人也提供了方便。To sum up, the medical image processing system, device, electronic device and storage medium provided by the present invention firstly obtains a scanned image, and then, based on the scanned image, obtains at least one of the target object's metabolic detection image, organ segmentation image and body part image. One, in which the metabolic detection image identifies abnormal metabolic regions of interest, the organ segmentation image performs organ segmentation on the target object, the body part image performs body part segmentation on the target object, and finally the metabolic detection image is combined with the organ segmentation image and the body part image. At least one of them is fused to determine the organ and/or body part corresponding to the region of interest with abnormal metabolism, without the need for doctors to manually determine the body part to which the region of interest belongs and its specific organ, and then manually perform the image processing. The body part to which the region of interest belongs and its specific organs are divided, which speeds up the process of lesion identification, improves the efficiency of lesion identification, and reduces the work of doctors. Moreover, without the need for professionals to perform image identification and judgment, ordinary personnel can use the final results to identify the lesions. The fusion image shows the target object's focal organ and the body part to which the focal target belongs, which also provides convenience for the patient.
当然，本领域普通技术人员可以理解实现上述实施例方法中的全部或部分流程，是可以通过计算机程序来指令相关硬件(如处理器，控制器等)来完成，所述的程序可存储于一计算机可读取的存储介质中，该程序在执行时可包括如上述各方法实施例的流程。其中所述的存储介质可为存储器、磁碟、光盘等。Of course, those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be implemented by instructing relevant hardware (such as processors, controllers, etc.) through a computer program, and the programs can be stored in a In the computer-readable storage medium, when the program is executed, the processes of the above-mentioned method embodiments may be included. The storage medium may be a memory, a magnetic disk, an optical disk, or the like.
以上所述本发明的具体实施方式，并不构成对本发明保护范围的限定。任何根据本发明的技术构思所做出的各种其他相应的改变与变形，均应包含在本发明权利要求的保护范围内。The specific embodiments of the present invention described above do not limit the protection scope of the present invention. Any other corresponding changes and modifications made according to the technical concept of the present invention shall be included in the protection scope of the claims of the present invention.
Claims (10)
Translated fromChinesePriority Applications (4)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210610225.3ACN114943714A (en) | 2022-05-31 | 2022-05-31 | Medical image processing system, device, electronic equipment and storage medium |
| PCT/CN2023/097379WO2023232067A1 (en) | 2022-05-31 | 2023-05-31 | Systems and methods for lesion region identification |
| EP23815244.1AEP4508597A4 (en) | 2022-05-31 | 2023-05-31 | SYSTEMS AND METHODS FOR IDENTIFYING LESION AREAS |
| US18/942,756US20250069228A1 (en) | 2022-05-31 | 2024-11-10 | Systems and methods for lesion region identification |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210610225.3ACN114943714A (en) | 2022-05-31 | 2022-05-31 | Medical image processing system, device, electronic equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114943714Atrue CN114943714A (en) | 2022-08-26 |
Family
ID=82909233
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210610225.3APendingCN114943714A (en) | 2022-05-31 | 2022-05-31 | Medical image processing system, device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114943714A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114202516A (en)* | 2021-11-29 | 2022-03-18 | 上海联影医疗科技股份有限公司 | Foreign matter detection method and device, electronic equipment and storage medium |
| CN115311252A (en)* | 2022-09-02 | 2022-11-08 | 上海联影医疗科技股份有限公司 | Medical image processing method, system and storage medium |
| CN115984220A (en)* | 2022-12-30 | 2023-04-18 | 上海联影智能医疗科技有限公司 | Metabolic level evaluation method, computer device, and storage medium |
| WO2023232067A1 (en)* | 2022-05-31 | 2023-12-07 | Shanghai United Imaging Healthcare Co., Ltd. | Systems and methods for lesion region identification |
| CN117689567A (en)* | 2024-01-31 | 2024-03-12 | 广州索诺康医疗科技有限公司 | Ultrasonic image scanning method and device |
| CN118380113A (en)* | 2024-05-15 | 2024-07-23 | 东莞市东南部中心医院(东莞市东南部中医医疗服务中心) | Auxiliary data processing method and system suitable for neurology image |
| CN118691806A (en)* | 2024-04-10 | 2024-09-24 | 湖南医药学院 | Diseased lymph node image segmentation method and lymphoma auxiliary diagnosis method |
| WO2025123216A1 (en)* | 2023-12-12 | 2025-06-19 | 中国科学院深圳先进技术研究院 | Medical image recognition method and apparatus, device and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103942785A (en)* | 2014-04-09 | 2014-07-23 | 苏州大学 | Lung tumor segmentation method based on PET and CT images of image segmentation |
| CN109308728A (en)* | 2018-10-25 | 2019-02-05 | 上海联影医疗科技有限公司 | PET-Positron emission computed tomography scan image processing method and processing device |
| US20190188870A1 (en)* | 2017-12-20 | 2019-06-20 | International Business Machines Corporation | Medical image registration guided by target lesion |
| CN113711271A (en)* | 2019-03-15 | 2021-11-26 | 豪夫迈·罗氏有限公司 | Deep convolutional neural network for tumor segmentation by positron emission tomography |
| CN114159085A (en)* | 2021-12-06 | 2022-03-11 | 深圳市联影高端医疗装备创新研究院 | PET image attenuation correction method and device, electronic equipment and storage medium |
- 2022
- 2022-05-31CNCN202210610225.3Apatent/CN114943714A/enactivePending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103942785A (en)* | 2014-04-09 | 2014-07-23 | 苏州大学 | Lung tumor segmentation method based on PET and CT images of image segmentation |
| US20190188870A1 (en)* | 2017-12-20 | 2019-06-20 | International Business Machines Corporation | Medical image registration guided by target lesion |
| CN109308728A (en)* | 2018-10-25 | 2019-02-05 | 上海联影医疗科技有限公司 | PET-Positron emission computed tomography scan image processing method and processing device |
| CN113711271A (en)* | 2019-03-15 | 2021-11-26 | 豪夫迈·罗氏有限公司 | Deep convolutional neural network for tumor segmentation by positron emission tomography |
| CN114159085A (en)* | 2021-12-06 | 2022-03-11 | 深圳市联影高端医疗装备创新研究院 | PET image attenuation correction method and device, electronic equipment and storage medium |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114202516A (en)* | 2021-11-29 | 2022-03-18 | 上海联影医疗科技股份有限公司 | Foreign matter detection method and device, electronic equipment and storage medium |
| WO2023232067A1 (en)* | 2022-05-31 | 2023-12-07 | Shanghai United Imaging Healthcare Co., Ltd. | Systems and methods for lesion region identification |
| CN115311252A (en)* | 2022-09-02 | 2022-11-08 | 上海联影医疗科技股份有限公司 | Medical image processing method, system and storage medium |
| CN115984220A (en)* | 2022-12-30 | 2023-04-18 | 上海联影智能医疗科技有限公司 | Metabolic level evaluation method, computer device, and storage medium |
| WO2025123216A1 (en)* | 2023-12-12 | 2025-06-19 | 中国科学院深圳先进技术研究院 | Medical image recognition method and apparatus, device and storage medium |
| CN117689567A (en)* | 2024-01-31 | 2024-03-12 | 广州索诺康医疗科技有限公司 | Ultrasonic image scanning method and device |
| CN117689567B (en)* | 2024-01-31 | 2024-05-24 | 广州索诺康医疗科技有限公司 | Ultrasonic image scanning method and device |
| CN118691806A (en)* | 2024-04-10 | 2024-09-24 | 湖南医药学院 | Diseased lymph node image segmentation method and lymphoma auxiliary diagnosis method |
| CN118691806B (en)* | 2024-04-10 | 2025-03-28 | 湖南医药学院 | Diseased lymph node image segmentation method and lymphoma auxiliary diagnosis method |
| CN118380113A (en)* | 2024-05-15 | 2024-07-23 | 东莞市东南部中心医院(东莞市东南部中医医疗服务中心) | Auxiliary data processing method and system suitable for neurology image |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114943714A (en) | Medical image processing system, device, electronic equipment and storage medium | |
| US9741131B2 (en) | Anatomy aware articulated registration for image segmentation | |
| US9342885B2 (en) | Method of generating a multi-modality anatomical atlas | |
| Linguraru et al. | Automated segmentation and quantification of liver and spleen from CT images using normalized probabilistic atlases and enhancement estimation | |
| US10803354B2 (en) | Cross-modality image synthesis | |
| US9471987B2 (en) | Automatic planning for medical imaging | |
| EP4266252A2 (en) | Systems and methods for image generation | |
| US9218542B2 (en) | Localization of anatomical structures using learning-based regression and efficient searching or deformation strategy | |
| US9082231B2 (en) | Symmetry-based visualization for enhancing anomaly detection | |
| US10796464B2 (en) | Selective image reconstruction | |
| US12423838B2 (en) | Medical image registration method and apparatus | |
| US20100128953A1 (en) | Method and system for registering a medical image | |
| US11327773B2 (en) | Anatomy-aware adaptation of graphical user interface | |
| US8588498B2 (en) | System and method for segmenting bones on MR images | |
| US9336457B2 (en) | Adaptive anatomical region prediction | |
| US10460508B2 (en) | Visualization with anatomical intelligence | |
| US9020215B2 (en) | Systems and methods for detecting and visualizing correspondence corridors on two-dimensional and volumetric medical images | |
| US20190012805A1 (en) | Automatic detection of an artifact in patient image data | |
| US20250069228A1 (en) | Systems and methods for lesion region identification | |
| US11495346B2 (en) | External device-enabled imaging support | |
| US9286688B2 (en) | Automatic segmentation of articulated structures | |
| US11823399B2 (en) | Multi-scan image processing | |
| Palma et al. | Deformable registration of 3D CT images with partial liver | |
| WO2024253692A1 (en) | Attenuation correction for medical imaging | |
| Pradhan | Development of Efficient Intensity Based Registration Techniques for Multi-modal Brain Images |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |