CN112263236A

CN112263236A - Whole-body tumor MRI intelligent evaluation system and method

Info

Publication number: CN112263236A
Application number: CN202010999233.2A
Authority: CN
Inventors: 刘想; 岳新; 王霄英; 贺长征; 张虽虽; 刘伟鹏
Original assignee: Beijing Smarttree Medical Technology Co Ltd
Current assignee: Beijing Smarttree Medical Technology Co Ltd
Priority date: 2020-09-22
Filing date: 2020-09-22
Publication date: 2021-01-26
Anticipated expiration: 2040-09-22
Also published as: CN112263236B

Abstract

The invention provides a whole-body tumor MRI intelligent evaluation system, which comprises: when a patient finishes shooting a whole body magnetic resonance examination project, the image identification module identifies a DICOM image sequence related to a DICOM image, the image quality judgment module analyzes and judges the quality of the required DICOM image sequence, the variation judgment module analyzes the postoperative change of the DICOM image sequence with qualified quality, the anatomical segmentation module segments a target organ and/or a target tissue of the DICOM image without postoperative change, the target focus identification module segments and positions a focus, and the structural report module integrates all processed data and images, stores the data and the images and outputs a diagnosis impression for reference of a doctor. The invention also discloses a whole-body tumor MRI intelligent evaluation method. According to the invention, the evaluation parts and the part evaluation methods are associated with a plurality of AI diagnosis models, and the related evaluation methods are built in the structured report module, so that the recording strength is greatly reduced, the diagnosis efficiency of doctors is improved, and the examination cost of patients is reduced.

Description

Whole-body tumor MRI intelligent evaluation system and method

Technical Field

The invention relates to the field of medical information, in particular to a whole-body tumor MRI intelligent evaluation system and a whole-body tumor MRI intelligent evaluation method.

Background

The imaging baseline examination and imaging follow-up examination of tumor patients are of paramount importance in assessing the benefit of a treatment regimen. Patients with advanced tumors undergo various treatment regimens and are often evaluated for imaging efficacy at 8 week, 12 week, and 16 week intervals. If PET-CT is adopted, the price is high, and radiation damage is caused, so that the PET-CT is difficult to apply to real services. For example, with the increase of treatment methods for advanced prostate cancer, accurate assessment of advanced metastatic prostate cancer becomes one of the important tasks for image examination of prostate cancer patients. In recent years, multi-parameter magnetic resonance imaging (mpMRI) is widely applied to diagnosis of prostate cancer, wherein DWI imaging can detect not only intraglandular tumors, but also lymph nodes and bone metastasis foci, and can be further used for evaluating the whole body tumor load; qualitative diagnosis is performed based on DWI image representation, and quantitative measurement can also be performed on ADC maps. Several studies have demonstrated that the diagnostic efficacy of whole body magnetic resonance imaging (WB-MRI) on systemic metastasis in patients with advanced prostate cancer is comparable to PET/CT, and can be used not only for disease diagnosis, but also for evaluating post-treatment response. Since the complexity of the WB-MRI technique is high, a standardized image acquisition scheme and report standards need to be formulated, and a method for diagnosing whole-body magnetic resonance imaging is not created in the existing medical institution, in the prior art, the complexity exceeds the memory capacity of people, the time for collecting and collating data is too long, a single pathology needs 2-3 hours to complete a diagnosis report, and no real operability exists, so that the WB-MRI technique cannot be put into clinical use, the examination cost of a patient is increased, and the diagnosis efficiency of a clinician is reduced.

Disclosure of Invention

In view of the above, the main objective of the present invention is to provide an MRI-evaluation intelligent reporting system and method for a whole-body tumor, which can solve the problems of increasing the examination cost of a patient and reducing the diagnosis efficiency of a clinician due to the inability to diagnose the whole-body magnetic resonance imaging in the prior art.

In order to achieve the purpose, the technical scheme of the invention is realized as follows:

on one hand, the invention provides a whole-body tumor MRI intelligent evaluation system, which comprises an image information management module, an image identification module, an image quality judgment module, a variation judgment module, an anatomical segmentation module, a target focus identification module and a structural report module, wherein the image information management module is connected with the image identification module and is used for transmitting a DICOM image of a patient to the image identification module through a DICOM protocol when the patient finishes shooting a whole-body magnetic resonance examination item; the image identification module is connected with the image information management module, the image quality judgment module and the structural report module and used for identifying DICOM images matched with the whole-body magnetic resonance examination items, extracting all part information based on DICOM image header file information, extracting a DICOM image sequence required by each part based on a first preset rule, defining the extracted DICOM image sequence as a first image, sending the first image to the image quality judgment module, and sending the part information, the sequence type of the first image corresponding to the part and the first image to the structural report module; when the DICOM image is not matched with the whole-body magnetic resonance examination item, the diagnosis process is stopped and first prompt information is sent to the structured report module; the image quality judging module is respectively connected with the image identification module, the variation judging module and the structural reporting module and used for analyzing the quality of the first image based on preset conditions, respectively sending the first image meeting the preset conditions to the variation judging module and the structural reporting module and defining the first image meeting the preset conditions as a second image; if the quality of the first image does not meet the preset condition, stopping the diagnosis process; sending the judgment result and the second prompt message to a structured report module; the variation judging module is connected with the image quality judging module, the anatomical segmentation module and the structural report module and is used for judging whether the second image has postoperative change and/or congenital development variation or not, if the second image does not have the postoperative change and/or the congenital development variation, the second image is respectively sent to the anatomical segmentation module and the structural report module, if the second image has the postoperative change and/or the congenital development variation, the type of the postoperative change and/or the congenital development variation is identified, each parameter of a postoperative residual structure is measured, and the type of the postoperative change and/or the congenital development variation and all the parameters are sent to the structural report module; the anatomical segmentation module is respectively connected with the variation judgment module, the target lesion identification module and the structured report module, and is used for segmenting all target organs and all target tissues on the second image based on a second preset rule, setting anatomical coordinates of each target organ and each target tissue, outputting first diagnosis data, setting an anatomical label for each anatomical coordinate, outputting a region of each anatomical label, namely a third image, and respectively sending the first diagnosis data and the third image to the target lesion identification module and the structured report module; the target focus identification module is respectively connected with the dissection segmentation module and the structured report module and used for segmenting all focuses on a third image based on a third preset rule and first diagnosis data, setting focus coordinates for each focus, namely second diagnosis data, setting focus labels for each focus coordinate, outputting the region of each focus label, namely a fourth image, comparing the second diagnosis data with the first diagnosis data, positioning and measuring each focus, and sending the positioning result, focus measurement value, key image, second diagnosis data and fourth image of each focus to the structured report module; the structural report module is respectively connected with the image identification module, the image quality judgment module, the variation judgment module, the anatomical segmentation module and the target focus identification module and is used for automatically generating a diagnosis impression for a doctor to check the positioning result and the focus measurement value of the focus based on a built-in rule; and stores all the received data and all the images.

Preferably, the anatomical segmentation module further comprises a label judgment unit, configured to judge whether the anatomical label is compliant based on the first diagnostic data, and the judgment rule is: judging whether the radial line volume of the maximum connected domain of each anatomical label is within a preset threshold value, judging whether the spatial relationship of adjacent anatomical labels is correct, judging the shapes of different anatomical labels, and sending the judgment result to a structured report module; and if the anatomical label is in compliance, sending the first diagnosis data to the target lesion identification module, and if the anatomical label is in non-compliance, sending third prompt information, the type of the non-compliance and the measured value of the anatomical label to the structured report module.

Preferably, the target lesion recognition module further includes a determination unit for determining a relative position of the lesion with respect to the target organ or the target tissue based on a localization result of the lesion, and outputting related data, that is: the lesion is within the target organ or target tissue, the lesion invades the target organ or target tissue, and the lesion is outside the target organ or target tissue.

Preferably, the target lesion identification module further includes a key image generation unit for comparing sizes of all lesions in each target organ or each target tissue based on the lesion measurement values, generating a key image of a lesion conforming to a fourth preset rule, and transmitting the key image to the structured report module.

In another aspect, the present invention further provides a whole-body tumor MRI intelligent assessment method, including: when the patient finishes the whole body magnetic resonance examination item, the image information management module transmits the DICOM image of the patient to the image identification module through a DICOM protocol; the image identification module identifies DICOM images matched with a whole-body magnetic resonance examination item, extracts all part information based on DICOM image header file information, extracts a DICOM image sequence required by each part based on a first preset rule, defines the extracted DICOM image sequence as a first image, sends the first image to the image quality judgment module, and sends the part information, the sequence type of the first image corresponding to the part and the first image to the structural report module; when the DICOM image is not matched with the whole-body magnetic resonance examination item, the diagnosis process is stopped and first prompt information is sent to the structured report module; the image quality judging module analyzes the quality of the first image based on a preset condition, the first image meeting the preset condition is respectively sent to the variation judging module and the structural reporting module, and the first image meeting the preset condition is defined as a second image; if the quality of the first image does not meet the preset condition, stopping the diagnosis process; sending the judgment result and the second prompt message to a structured report module; the variation judging module judges whether the second image has postoperative change and/or congenital development variation, if no postoperative change and/or congenital development variation exists, the second image is respectively sent to the dissection module and the structural reporting module, if the postoperative change and/or congenital development variation exists, the type of the postoperative change and/or congenital development variation is identified, each parameter of a postoperative residual structure is measured, and the type of the postoperative change and/or congenital development variation and all the parameters are sent to the structural reporting module; the dissection segmentation module segments all target organs and all target tissues on the second image based on a second preset rule, sets dissection coordinates of each target organ and each target tissue, outputs first diagnosis data, sets dissection labels for each dissection coordinate, outputs a region of each dissection label, namely a third image, and respectively sends the first diagnosis data and the third image to the target focus identification module and the structured report module; the target focus identification module divides all focuses on a third image based on a third preset rule and first diagnosis data, sets focus coordinates, namely second diagnosis data, for each focus coordinate, sets a focus label for each focus coordinate, outputs a region of each focus label, namely a fourth image, compares the second diagnosis data with the first diagnosis data, positions and measures each focus, and sends a positioning result, a focus measurement value, a key image, second diagnosis data and the fourth image of each focus to the structured report module; the structured report module automatically generates a diagnosis impression according to a positioning result and a focus measurement value of a focus based on a built-in rule for a doctor to check; and stores all the received data and all the images.

Preferably, the method further comprises: a label judgment unit in the dissection module judges whether the dissection label is in compliance or not based on the first diagnosis data, and the judgment rule is as follows: judging whether the radial line volume of the maximum connected domain of each anatomical label is within a preset threshold value, judging whether the spatial relationship of adjacent anatomical labels is correct, judging the shapes of different anatomical labels, and sending the judgment result to a structured report module; and if the anatomical label is in compliance, sending the first diagnosis data to the target lesion identification module, and if the anatomical label is in non-compliance, sending third prompt information, the type of the non-compliance and the measured value of the anatomical label to the structured report module.

Preferably, the method further comprises: the judging unit in the target focus identification module judges the relative position of the focus and a target organ or target tissue based on the positioning result of the focus and outputs related data, namely: the lesion is within the target organ or target tissue, the lesion invades the target organ or target tissue, and the lesion is outside the target organ or target tissue.

Preferably, the method further comprises: and a key image generation unit in the target lesion identification module compares the sizes of all lesions in each target organ or each target tissue based on the lesion measurement value, generates a key image of the lesion conforming to a fourth preset rule, and sends the key image to the structured report module.

The invention has the technical effects that:

1. because the invention is provided with the image identification module, the image quality judgment module, the variation judgment module, the anatomical segmentation module, the target focus identification module and the structuralized report module, when a patient finishes shooting a whole body magnetic resonance (WB-MRI) examination item, the image identification module identifies DICOM image sequences related to DIOCOM images, the image quality judgment module analyzes and judges the quality of the required DICOM image sequences, identifies the DICOM images with poor quality caused by artifacts and the like to prevent influencing subsequent diagnosis, the variation judgment module analyzes the postoperative change of the DICOM image sequences with qualified quality, if the DICOM images with postoperative anatomical change and congenital malformation are not removed in advance, a great amount of misjudgment of a subsequent diagnosis flow focus analysis model can be caused, the diagnosis precision is influenced, the anatomical segmentation module carries out the segmentation of target organs and/or target tissues on the DICOM images without postoperative change, the target focus recognition module is used for segmenting and positioning the focus, the structured report module is used for integrating all processed data and images, storing the data and the images and outputting a diagnosis impression for reference of a doctor; the system associates the evaluation parts and the part evaluation methods with a plurality of AI diagnosis models, embeds the related evaluation methods in a structured report module, greatly reduces the recording strength, improves the diagnosis efficiency of doctors, and reduces the examination cost of patients; the original evaluation scheme which cannot be used in practice can be changed to be used on the ground; with the use of an accurate evaluation system of the whole-body heterogeneity, the selection of treatment schemes will be changed, high-level research evidence is accumulated, the image information and the clinical information are integrated together by receiving an information tool and an intelligent technology in the future, better clinical decision can be obtained, and the clinical value of the image service is improved;

2. because the label judging unit is arranged, whether the anatomical label is in compliance or not can be judged based on the first diagnosis data, if the anatomical label is in compliance, the first diagnosis data is sent to the target focus identification module, and if the anatomical label is in non-compliance, the third prompt information is sent to the structured report module, so that the diagnosis inaccuracy caused by unqualified anatomical coordinates is avoided, meanwhile, the prompt information is sent, so that manual intervention and processing are facilitated in time, an AI diagnosis model is perfected, and the whole diagnosis process is more perfect and more systematic;

3. because the invention is provided with the judging unit, the relative position of the focus and the target organ or the target tissue can be judged based on the positioning result of the focus, and the related data is output, namely: the focus is in the target organ or target tissue, the focus invades the target organ or target tissue, the focus is outside the target organ or target tissue, feedback to the corresponding interface of the structural report module synchronously, help the diagnosis of the clinician;

4. because the invention is provided with the key image generating unit, the sizes of all focuses in each target organ or each target tissue can be compared based on the focus measured value, the key image of the focus which accords with the fourth preset rule (such as the maximum 3 focuses of each part) is generated, and the key image is sent to the structured report module, thereby improving the diagnosis efficiency of a clinician and leading the structured report interface to be more visual.

Drawings

The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the invention and together with the description serve to explain the invention without limiting the invention. In the drawings:

FIG. 1 is a schematic structural diagram of a whole-body tumor MRI intelligent evaluation system according to a first embodiment of the present invention;

fig. 2 is a schematic diagram of a structured report interface of a whole-body MRI for prostate cancer metastasis in the system for MRI intelligent assessment of a whole-body tumor according to a first embodiment of the present invention;

FIG. 3 is a schematic structural diagram of a whole-body tumor MRI intelligent evaluation system according to a second embodiment of the invention;

FIG. 4 is a schematic structural diagram of a whole-body tumor MRI intelligent evaluation system according to a third embodiment of the invention;

FIG. 5 is a schematic structural diagram of a whole-body tumor MRI intelligent evaluation system according to a fourth embodiment of the invention;

FIG. 6 is a flow chart of a whole-body tumor MRI intelligent evaluation method according to the fifth embodiment of the invention;

fig. 7 is a schematic diagram of a structured report interface of the whole-body MRI for prostate cancer metastasis in the MRI intelligent assessment method of the whole-body tumor according to the fifth embodiment of the present invention;

fig. 8 shows a specific processing flow chart of the MRI intelligent assessment method for the whole-body tumor according to the sixth embodiment of the present invention.

Detailed Description

The present invention will be described in detail below with reference to the accompanying drawings in conjunction with embodiments.

Example one

FIG. 1 is a schematic structural diagram of a whole-body tumor MRI intelligent evaluation system according to a first embodiment of the present invention; as shown in fig. 1, the system includes: an image information management module 10, an image recognition module 20, an image quality judgment module 30, a mutation judgment module 40, an anatomical segmentation module 50, a target lesion identification module 60, and a structured report module 70, wherein,

the image information management module 10 is connected with the image identification module 20 and is used for transmitting the DICOM image of the patient to the image identification module 20 through a DICOM protocol when the patient finishes shooting a whole body magnetic resonance examination item;

the image Information management module is an ris (radio Information system) system.

The image identification module 20 is connected with the image information management module 10, the image quality judgment module 30 and the structural report module 70, and is used for identifying a DICOM image matched with a whole-body magnetic resonance examination item, extracting all part information based on DICOM image header file information, extracting a DICOM image sequence required by each part based on a first preset rule, defining the extracted DICOM image sequence as a first image, sending the first image to the image quality judgment module 30, and sending the part information, the sequence type of the first image corresponding to the part and the first image to the structural report module 70; when the DICOM image is not matched with the whole-body magnetic resonance examination item, the diagnosis process is stopped and first prompt information is sent to the structured report module 70;

the region refers to a target organ or a target tissue, such as a lymph node, an internal organ, a prostate, a bone, and the like, and the region information and the sequence type corresponding to the region are returned to the corresponding control of the "technical assessment (scanning sequence)" of the structured report interface as required.

The DICOM image sequence types can be DWI _ High, DWI _ Low, ADC, T1WI _ In, T1WI _ Opp, T2WI _ Fs, T2WI and the like, and all DICOM image sequences required by each part are identified for subsequent AI model diagnosis. When the DICOM image is not matched with the whole-body magnetic resonance examination item, the diagnosis process is stopped, first prompt information is sent to the structural report module to prompt relevant responsible personnel to process in time, and the first image and the first prompt information are stored in a database of the structural report module.

The image quality judging module 30 is respectively connected to the image identifying module 20, the variation judging module 40 and the structural reporting module 70, and is configured to analyze the quality of the first image based on a preset condition, respectively send the first image meeting the preset condition to the variation judging module 40 and the structural reporting module 70, and define the first image meeting the preset condition as a second image; if the quality of the first image does not meet the preset condition, stopping the diagnosis process; and sends the judgment result and the second prompt message to the structured report module 70;

and identifying DICOM images with unqualified quality, and avoiding the conditions of image diagnosis, such as low image signal-to-noise ratio, magnetic sensitivity artifacts, breathing artifacts, motion artifacts, scan range complement, incomplete image sequences and the like of DWI. If the result is an unqualified DICOM image, the diagnosis process is stopped, second prompt information is sent to the structured report module, the judgment result is sent to a corresponding control of 'technical evaluation (image quality)' of the structured report interface, relevant responsible personnel are prompted to process the image in time, and the second image and the second prompt information are stored in a database of the structured report module.

A variation judging module 40 connected to the image quality judging module 30, the anatomical segmentation module 50 and the structural reporting module 70, for judging whether there is a postoperative change and/or an congenital development variation in the second image, if there is no postoperative change and/or congenital development variation, sending the second image to the anatomical segmentation module 50 and the structural reporting module 70, respectively, if there is a postoperative change and/or congenital development variation, identifying the type of the postoperative change and/or congenital development variation, measuring each parameter of the postoperative residual structure, and sending the type of the postoperative change and/or congenital development variation and all parameters to the structural reporting module 70;

determining whether there is a post-treatment change that alters the anatomy, such as: TURP, radical prostatectomy, other pelvic surgeries, etc., fed back into the overall assessment of the structured report interface.

The type of the postoperative change and/or the congenital development variation is TURP, radical prostatectomy, other pelvic surgeries and the like, and when the measured parameters such as the volume, radial lines and the like of the postoperative residual structure are judged to have the postoperative change, the postoperative change or the congenital development variation is sent to a corresponding control of 'overall evaluation (postoperative change, congenital development variation)' of a structured report interface.

An anatomical segmentation module 50, connected to the variation judgment module 40, the target lesion identification module 60, and the structural report module 70, respectively, for segmenting all target organs and all target tissues on the second image based on a second preset rule, setting an anatomical coordinate of each target organ and each target tissue, outputting first diagnostic data, setting an anatomical label for each anatomical coordinate, outputting a region of each anatomical label, i.e., a third image, and sending the first diagnostic data and the third image to the target lesion identification module 60 and the structural report module 70, respectively;

for example, for a whole body magnetic resonance examination of prostate cancer metastasis, the segmentation of the target organ and target tissue is as follows:

dividing structures of pelvic cavity soft tissues such as prostate, seminal vesicle, rectum, bladder, obturator internus muscle, levator ani and the like;

segmenting the soft tissue structures of the abdomen, the chest and the head;

dividing the lymph node region of the whole body, including pelvic lymph node, retroperitoneal lymph node and other lymph nodes;

the bone structures of the pelvic cavity, such as the lower lumbar vertebra, the ilium, the sacrum, the ischium, the pubis, the acetabulum, the femoral neck, the femoral head, the femoral neck and the like, and the bone structures of the skull, the cervical vertebra, the thoracic vertebra, the lumbar vertebra, the rib, the thorax and the like are segmented.

RECIST 1.1 therapeutic effect evaluation standard is used for three tissues/organs of lymph node, organ and prostate, and MET-RADS-P custom standard (progression, stability and response) is used for bone treatment response.

Detailed rules for comprehensive RAC evaluation: the RAC score definition standard table of PCWG3 is referred to in the Prostate Cancer Working Group, and will not be described in detail. The present systematic approach embeds these explicit rules in a structured report that automates the generation of RAC scores.

A target lesion recognizing module 60, respectively connected to the dissection segmentation module 50 and the structured reporting module 70, for segmenting all lesions on the third image based on a third preset rule and the first diagnostic data, setting a lesion coordinate, i.e. second diagnostic data, for each lesion coordinate, setting a lesion tag for each lesion coordinate, outputting a region of each lesion tag, i.e. a fourth image, comparing the second diagnostic data with the first diagnostic data, positioning and measuring each lesion, and sending a positioning result, a lesion measurement value, a key image, second diagnostic data, and a fourth image of each lesion to the structured reporting module 70;

for bone lesions, lymph node lesions, soft tissue lesions and prostate lesions, different image sequences and AI models are used for segmentation; this technique, although highly complex, is relatively mature and will not be described in detail. The result of the segmentation is coordinates of an anatomical label of the lesion and measurement information of the lesion.

The structural report module 70 is respectively connected with the image recognition module 20, the image quality judgment module 30, the variation judgment module 40, the anatomical segmentation module 50 and the target lesion recognition module 60, and is used for automatically generating a diagnosis impression for a doctor to check a lesion positioning result and a lesion measurement value based on a built-in rule; and stores all the received data and all the images.

Wherein the localization results and lesion measurements are returned to the "diagnostic impression" of the structured report.

Fig. 2 is a structural schematic view of a structural report interface of a whole-body magnetic resonance examination for prostate cancer metastasis in a structural schematic view of a whole-body tumor MRI intelligent evaluation system according to a first embodiment of the present invention; as shown in fig. 2, the technical assessment includes image sequence and image quality, the structured report is automatically associated with each AI module, and the overall assessment includes a list of corresponding parameters of the target organ and the target tissue lesion, including primary organ (prostate), liver, lung, other soft tissue region, pelvic lymph node, retroperitoneal lymph node, other lymph node, bone, skull, cervical vertebra, thoracic vertebra, sacral vertebra, pelvis, thoracic cage, limbs.

The whole body MRI examination (WB-MRI) uses basic T1, T2, DWI sequences (including ADC pictures; DWI sequences can not only show intraglandular tumors, but also lymph node metastasis and bone metastasis), and the quantitative detection of bone and lymph node metastasis can be completed within 30 minutes. If the detection of tissues and internal organs is added to the software, only 45-50 minutes are needed. The cost is low, no additional damage is caused, and the clinical effect is not inferior to that of PET-CT, so that a standard WB-MRI evaluation method has more clinical value.

The evaluation of WB-MRI is difficult in clinical use because there are many sites to be evaluated and the evaluation rule for each site is related to the type of disease. Without a standardized, intelligent tool, it is essentially impossible to achieve a correct assessment based on the physician's memory alone. Taking the WB-MRI prostate as an example, the range that needs to be scanned and evaluated includes: bones, lymph nodes, soft tissues and internal organs of the head and neck trunk, and tumors of the pelvic region around the prostate.

For tumors of bones, lymph, soft tissues, organs, methods of automatic feature extraction using specific AI, structured reporting methods for characterization of individual lesions, and clinical evaluation of lesion changes over time have all become mature. The present systemic method organizes the prostate cancer into their sequential order according to the logic of the assessment of systemic metastases: including the order of lesion evaluation, a method for personalized evaluation of different lesions, integrating post-processing/AI automatic extraction features into a report if they exist; finally, a score of 1-5 points is automatically given according to RAC (response assessment category) of WB-MRI-P, and clinical reference is provided.

And sending the target lesion segmentation information and the target lesion positioning information to a structured report system of the whole-body MRI prostate evaluation. The structured report system embeds RECIST evaluation method, bone reaction evaluation method and RAC scoring method in the system through conventional programs, and then automatically obtains evaluation conclusion according to data (or manually input numerical values) transmitted by the AI diagnosis modules.

The system can be used for evaluating not only WB-MRI-P prostate cancer systemic metastasis, but also tumors such as breast cancer, lymphoma, hematopathy and the like through configuration rules.

The embodiment of the invention is provided with an image identification module, an image quality judgment module, a variation judgment module, an anatomical segmentation module, a target lesion identification module and a structural report module, when a patient finishes shooting a whole body magnetic resonance (WB-MRI) examination item, the image identification module identifies DICOM image sequences related to DIOCOM images, the image quality judgment module analyzes and judges the quality of required DICOM image sequences, the DICOM images with poor quality caused by artifacts and the like are identified to prevent influencing subsequent diagnosis, the variation judgment module analyzes the postoperative change of the DICOM image sequences with qualified quality, if the DICOM images with postoperative anatomical change and congenital malformation are not removed in advance, a great amount of misjudgment of a lesion analysis model in a subsequent diagnosis process can be caused, the diagnosis precision is influenced, the anatomical segmentation module performs segmentation of target organs and/or target tissues on the DICOM images without postoperative change, the target focus recognition module is used for segmenting and positioning the focus, the structured report module is used for integrating all processed data and images, storing the data and the images and outputting a diagnosis impression for reference of a doctor; the system associates the evaluation parts and the part evaluation methods with a plurality of AI diagnosis models, embeds the related evaluation methods in a structured report module, greatly reduces the recording strength, improves the diagnosis efficiency of doctors, and reduces the examination cost of patients; the original evaluation scheme which cannot be used in practice can be changed to be used on the ground; with the use of an accurate evaluation system of the whole-body heterogeneity, the selection of treatment schemes will change, high-level research evidence is accumulated, and the image information and the clinical information are integrated by receiving an information tool and an intelligent technology in the future, so that a better clinical decision can be obtained, and the clinical value of the image service is improved.

Example two

Fig. 3 shows a schematic structural diagram of a whole-body tumor MRI intelligent evaluation system according to a second embodiment of the present invention, and as shown in fig. 3, the anatomical segmentation module 50 further includes a label determination unit 502, configured to determine whether the anatomical label is in compliance based on the first diagnostic data, where the rule of determination is: judging whether the radial line volume of the maximum connected domain of each anatomical tag is within a preset threshold value, judging whether the spatial relationship of adjacent anatomical tags is correct, judging the shapes of different anatomical tags, and sending the judgment result to the structured report module 70; if the anatomical tag is compliant, the first diagnostic data is sent to the target lesion identification module 60, and if the anatomical tag is non-compliant, a third prompt, the type of non-compliance, and the measured value of the anatomical tag are sent to the structured report module 70.

For example, it is determined whether the radial line volume of the largest connected component of each anatomical label is within a predetermined threshold, such as 10% -90%, if the first diagnostic data that is out of range is not compliant.

The embodiment of the invention is provided with the label judging unit which can judge whether the anatomical label is in compliance or not based on the first diagnosis data, if the anatomical label is in compliance, the first diagnosis data is sent to the target focus segmentation module, and if the anatomical label is in non-compliance, the third prompt message is sent to the structured report module, so that the diagnosis inaccuracy caused by unqualified anatomical coordinates is avoided, meanwhile, the prompt message is sent, so that the manual intervention and processing are facilitated in time, the AI diagnosis model is perfected, and the whole diagnosis process is more perfect and more systematic.

EXAMPLE III

Fig. 4 is a schematic structural diagram of a system for MRI intelligent evaluation of a whole-body tumor according to a third embodiment of the present invention, and as shown in fig. 4, the target lesion identification module 60 further includes a determination unit 602, configured to determine a relative position of a lesion with respect to a target organ or a target tissue based on a localization result of the lesion, and output related data, that is: the lesion is within the target organ or target tissue, the lesion invades the target organ or target tissue, and the lesion is outside the target organ or target tissue.

The embodiment of the invention is provided with a judging unit which can judge the relative position of the focus and the target organ or the target tissue based on the positioning result of the focus and output related data, namely: the focus is in the target organ or target tissue, the focus invades the target organ or target tissue, the focus is outside the target organ or target tissue, and the focus is synchronously fed back to the corresponding interface of the structured report module, thereby being beneficial to the diagnosis of a clinician.

Example four

Fig. 5 is a schematic structural diagram of a whole-body tumor MRI intelligent evaluation system according to a fourth embodiment of the present invention, and as shown in fig. 5, the target lesion identification module 60 further includes a key image generation unit 604 for comparing sizes of all lesions in each target organ or each target tissue based on lesion measurement values, generating a key image of a lesion meeting a fourth preset rule, and sending the key image to the structural reporting module 70.

For example, a key image is generated for the largest 3 lesions per site.

The embodiment of the invention is provided with the key image generation unit, and can compare the sizes of all focuses in each target organ or each target tissue based on focus measurement values, generate key images of the focuses which accord with a fourth preset rule (such as the largest 3 focuses of each part), and send the key images to the structured report module, so that the diagnosis efficiency of a clinician is improved, and a structured report interface is more visual.

EXAMPLE five

FIG. 6 is a flow chart of a whole-body tumor MRI intelligent evaluation method according to the fifth embodiment of the invention; as shown in fig. 6, the method comprises the steps of:

step S501, when the patient finishes the whole body magnetic resonance examination item, the image information management module transmits the DICOM image of the patient to the image identification module through the DICOM protocol;

wherein, the image Information management module is an RIS (radio Information System) system;

step S502, the image identification module identifies a DICOM image matched with a whole-body magnetic resonance examination item, extracts all part information based on DICOM image header file information, extracts a DICOM image sequence required by each part based on a first preset rule, defines the extracted DICOM image sequence as a first image, sends the first image to the image quality judgment module, and sends the part information, the sequence type of the first image corresponding to the part and the first image to the structural report module; when the DICOM image is not matched with the whole-body magnetic resonance examination item, the diagnosis process is stopped and first prompt information is sent to the structured report module;

Step S503, the image quality judging module analyzes the quality of the first image based on the preset condition, the first image meeting the preset condition is respectively sent to the variation judging module and the structured reporting module, and the first image meeting the preset condition is defined as a second image; if the quality of the first image does not meet the preset condition, stopping the diagnosis process; sending the judgment result and the second prompt message to a structured report module;

Step S504, a variation judging module judges whether the second image has postoperative change and/or congenital development variation, if no postoperative change and/or congenital development variation exists, the second image is respectively sent to an anatomical segmentation module and a structural report module, if the postoperative change and/or congenital development variation exists, the type of the postoperative change and/or congenital development variation is identified, each parameter of a postoperative residual structure is measured, and the type of the postoperative change and/or congenital development variation and all the parameters are sent to the structural report module;

Step S505, the dissection segmentation module segments all target organs and all target tissues on a second image based on a second preset rule, sets dissection coordinates of each target organ and each target tissue, outputs first diagnosis data, sets dissection labels for each dissection coordinate, outputs the region of each dissection label, namely a third image, and respectively sends the first diagnosis data and the third image to the target focus identification module and the structured report module;

segmenting the soft tissue structures of the abdomen, the chest and the head;

Step S506, the target focus identification module divides all focuses on a third image based on a third preset rule and first diagnosis data, sets focus coordinates for each focus, namely second diagnosis data, sets focus labels for each focus coordinate, outputs the region of each focus label, namely a fourth image, compares the second diagnosis data with the first diagnosis data, positions and measures each focus, and sends the positioning result, the focus measurement value, the key image, the second diagnosis data and the fourth image of each focus to a structured report module;

Step S507, the structured report module automatically generates a diagnosis impression for a doctor to check the positioning result and the focus measurement value of the focus based on a built-in rule; and stores all the received data and all the images.

Fig. 7 is a schematic diagram of a structured report interface of the whole-body MRI for prostate cancer metastasis in the MRI intelligent assessment method of the whole-body tumor according to the fifth embodiment of the present invention; as shown in fig. 7, the technical assessment includes image sequence and image quality, the structured report is automatically associated with each AI module, and the overall assessment includes a list of corresponding parameters of the target organ and target tissue lesion, including primary organ (prostate), liver, lung, other soft tissue region, pelvic lymph node, retroperitoneal lymph node, other lymph node, bone, skull, cervical vertebra, thoracic vertebra, sacral vertebra, pelvis, thoracic cage, limbs.

Wherein, the method also comprises: a label judgment unit in the dissection module judges whether the dissection label is in compliance or not based on the first diagnosis data, and the judgment rule is as follows: judging whether the radial line volume of the maximum connected domain of each anatomical label is within a preset threshold value, judging whether the spatial relationship of adjacent anatomical labels is correct, judging the shapes of different anatomical labels, and sending the judgment result to a structured report module; and if the anatomical label is in compliance, sending the first diagnosis data to the target lesion identification module, and if the anatomical label is in non-compliance, sending third prompt information, the type of the non-compliance and the measured value of the anatomical label to the structured report module.

Wherein, the method also comprises: the judging unit in the target focus identification module judges the relative position of the focus and a target organ or target tissue based on the positioning result of the focus and outputs related data, namely: the lesion is within the target organ or target tissue, the lesion invades the target organ or target tissue, and the lesion is outside the target organ or target tissue.

Wherein, the method also comprises: and a key image generation unit in the target lesion identification module compares the sizes of all lesions in each target organ or each target tissue based on the lesion measurement value, generates a key image of the lesion conforming to a fourth preset rule, and sends the key image to the structured report module.

For example, a key image is generated for the largest 3 lesions per site.

In the embodiment of the invention, the image identification module, the image quality judgment module, the variation judgment module, the anatomical segmentation module, the target lesion identification module and the structural report module are used, when a patient takes a whole body magnetic resonance imaging (WB-MRI) examination item, the image identification module identifies DICOM image sequences related to DIOCOM images, the image quality judgment module analyzes and judges the quality of the required DICOM image sequences, the DICOM images with poor quality caused by artifacts and the like are identified to prevent influencing subsequent diagnosis, the variation judgment module analyzes the postoperative change of the DICOM image sequences with qualified quality, if the DICOM images with postoperative anatomical change and congenital deformity are not removed in advance, a great amount of misjudgment of a subsequent diagnosis flow lesion analysis model can be caused, the diagnosis precision is influenced, the anatomical segmentation module performs segmentation of target organs and/or target tissues on the DICOM images without postoperative change, the target focus recognition module is used for segmenting and positioning the focus, the structured report module is used for integrating all processed data and images, storing the data and the images and outputting a diagnosis impression for reference of a doctor; the system associates the evaluation parts and the part evaluation methods with a plurality of AI diagnosis models, embeds the related evaluation methods in a structured report module, greatly reduces the recording strength, improves the diagnosis efficiency of doctors, and reduces the examination cost of patients; the original evaluation scheme which cannot be used in practice can be changed to be used on the ground; with the use of an accurate evaluation system of the whole-body heterogeneity, the selection of treatment schemes will be changed, high-level research evidence is accumulated, the image information and the clinical information are integrated together by receiving an information tool and an intelligent technology in the future, better clinical decision can be obtained, and the clinical value of the image service is improved; the label judging unit in the embodiment of the invention can judge whether the anatomical label is in compliance or not based on the first diagnosis data, if the anatomical label is in compliance, the first diagnosis data is sent to the target lesion segmentation module, and if the anatomical label is in non-compliance, the third prompt information is sent to the structured report module, so that the diagnosis inaccuracy caused by unqualified anatomical coordinates is avoided, meanwhile, the prompt information is sent, so that the manual intervention and processing are facilitated in time, an AI diagnosis model is perfected, and the whole diagnosis process is more perfect and more systematic; due to the determination unit in the embodiment of the present invention, based on the positioning result of the lesion, the relative position of the lesion and the target organ or target tissue may be determined, and relevant data may be output, that is: the focus is in the target organ or target tissue, the focus invades the target organ or target tissue, the focus is outside the target organ or target tissue, feedback to the corresponding interface of the structural report module synchronously, help the diagnosis of the clinician; because the key image generating unit in the invention can compare the sizes of all focuses in each target organ or each target tissue based on the focus measured value, generate the key image of the focus which accords with the fourth preset rule (such as the maximum 3 focuses of each part), and send the key image to the structured report module, the diagnosis efficiency of a clinician is improved, and the structured report interface is more visual.

EXAMPLE six

Fig. 8 shows a specific processing flow chart of a whole-body tumor MRI intelligent evaluation method according to a sixth embodiment of the present invention, as shown in fig. 8, the method includes the following steps:

in step S601, the image recognition module recognizes whether the DICOM image matches the examination item (whole body magnetic resonance imaging)? If not, sending a first prompt message to the structured report module, and if so, executing step S602;

step S602, the image recognition module extracts a required image sequence, sends a first image, a sequence type and part information to the structured report module, and sends the first image to the image quality judgment module;

in step S603, is the image quality meet a preset condition? If not, sending the judgment result and the second prompt message to a structured report module; if yes, go to step S604;

step S604, is there post-operative changes and/or congenital variations? If yes, sending the postoperative change and/or the congenital variation and the postoperative residual structure measurement parameters to a structured report module, and if not, executing a step S605;

step S605, the dissection segmentation module segments a target organ and a target tissue, sends the first diagnosis data and the third image to the result speech reporting module and executes the step S606;

step S606, whether the anatomical label is in compliance or not is judged, if not, third prompt information, the type of non-compliance and the measured value of the anatomical label are sent to the structural report module, and if yes, the step S607 is executed;

step S607, the target focus identification module segments the focus, positions and measures the focus, and sends the focus positioning result, the focus measuring value, the key image, the second diagnosis data and the fourth image to the structured report module;

step S608, the structured report module automatically generates a diagnosis impression for a doctor to check the positioning result and the measured value of the focus based on a built-in rule; and stores all the received data and all the images.

From the above description, it can be seen that the above-described embodiments of the present invention achieve the following technical effects: because the embodiment of the invention is provided with the image identification module, the image quality judgment module, the variation judgment module, the anatomical segmentation module, the target lesion identification module and the structural report module, when a patient finishes shooting a whole body magnetic resonance (WB-MRI) examination item, the image identification module identifies DICOM image sequences related to DIOCOM images, the image quality judgment module analyzes and judges the quality of the required DICOM image sequences, the DICOM images with poor quality caused by artifacts and the like are identified to prevent influencing subsequent diagnosis, the variation judgment module analyzes postoperative change of the DICOM image sequences with qualified quality, if the DICOM images with postoperative anatomical change and congenital malformation are not removed in advance, a great amount of misjudgment of a lesion analysis model in a subsequent diagnosis process can be caused, the diagnosis precision is influenced, the anatomical segmentation module performs segmentation of target organs and/or target tissues on the DICOM images without postoperative change, the target focus recognition module is used for segmenting and positioning the focus, the structured report module is used for integrating all processed data and images, storing the data and the images and outputting a diagnosis impression for reference of a doctor; the system associates the evaluation parts and the part evaluation methods with a plurality of AI diagnosis models, embeds the related evaluation methods in a structured report module, greatly reduces the recording strength, improves the diagnosis efficiency of doctors, and reduces the examination cost of patients; the original evaluation scheme which cannot be used in practice can be changed to be used on the ground; with the use of an accurate evaluation system of the whole-body heterogeneity, the selection of treatment schemes will be changed, high-level research evidence is accumulated, the image information and the clinical information are integrated together by receiving an information tool and an intelligent technology in the future, better clinical decision can be obtained, and the clinical value of the image service is improved; because the embodiment of the invention is provided with the label judging unit, whether the anatomical label is in compliance or not can be judged based on the first diagnosis data, if the anatomical label is in compliance, the first diagnosis data is sent to the target focus segmentation module, and if the anatomical label is in non-compliance, the third prompt information is sent to the structured report module, so that the diagnosis inaccuracy caused by unqualified anatomical coordinates is avoided, meanwhile, the prompt information is sent, so that the manual intervention and processing are facilitated in time, the AI diagnosis model is perfected, the whole diagnosis process is more perfect, and the system is more realized; because the embodiment of the invention is provided with the judging unit, the relative position of the focus and the target organ or the target tissue can be judged based on the positioning result of the focus, and relevant data is output, namely: the focus is in the target organ or target tissue, the focus invades the target organ or target tissue, the focus is outside the target organ or target tissue, feedback to the corresponding interface of the structural report module synchronously, help the diagnosis of the clinician; because the embodiment of the invention is provided with the key image generation unit, the sizes of all focuses in each target organ or each target tissue can be compared based on the focus measurement value, the focus which accords with the fourth preset rule (such as the maximum 3 focuses of each part) is generated into the key image, and the key image is sent to the structured report module, so that the diagnosis efficiency of a clinician is improved, and the structured report interface is more visual.

It will be apparent to those skilled in the art that the modules or steps of the present invention described above may be implemented by a general purpose computing device, they may be centralized on a single computing device or distributed across a network of multiple computing devices, and they may alternatively be implemented by program code executable by a computing device, such that they may be stored in a storage device and executed by a computing device, or fabricated separately as individual integrated circuit modules, or fabricated as a single integrated circuit module from multiple modules or steps. Thus, the present invention is not limited to any specific combination of hardware and software.

The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. An MRI intelligent evaluation system for whole body tumor is characterized by comprising an image information management module, an image identification module, an image quality judgment module, a variation judgment module, an anatomical segmentation module, a target focus identification module and a structural report module,

the image information management module is connected with the image identification module and is used for transmitting the DICOM image of the patient to the image identification module through a DICOM protocol when the patient finishes shooting a whole body magnetic resonance examination item;

the image identification module is connected with the image information management module, the image quality judgment module and the structural report module, and is used for identifying the DICOM image matched with the whole-body magnetic resonance examination item, extracting all part information based on DICOM image header file information, extracting a DICOM image sequence required by each part based on a first preset rule, defining the extracted DICOM image sequence as a first image, sending the first image to the image quality judgment module, and sending the part information, the sequence type of the first image corresponding to the part and the first image to the structural report module; when the DICOM image is not matched with the whole-body magnetic resonance examination item, stopping a diagnosis process and sending first prompt information to the structured report module;

the image quality judging module is respectively connected with the image identifying module, the variation judging module and the structural reporting module, and is used for analyzing the quality of the first image based on a preset condition, respectively sending the first image meeting the preset condition to the variation judging module and the structural reporting module, and defining the first image meeting the preset condition as a second image; if the quality of the first image does not meet the preset condition, stopping the diagnosis process; sending the judgment result and the second prompt message to the structured report module;

the variation judging module is connected with the image quality judging module, the anatomical segmentation module and the structural report module and is used for judging whether the second image has postoperative change and/or congenital development variation, if the second image does not have the postoperative change and/or congenital development variation, the second image is respectively sent to the anatomical segmentation module and the structural report module, if the second image has the postoperative change and/or congenital development variation, the type of the postoperative change and/or congenital development variation is identified, each parameter of a postoperative residual structure is measured, and the type of the postoperative change and/or congenital development variation and all the parameters are sent to the structural report module;

the anatomical segmentation module is connected to the variation judgment module, the target lesion identification module and the structured report module, and configured to segment all target organs and all target tissues on the second image based on a second preset rule, set anatomical coordinates of each target organ and each target tissue, output first diagnostic data, set an anatomical label for each anatomical coordinate, output a region of each anatomical label, that is, a third image, and send the first diagnostic data and the third image to the target lesion identification module and the structured report module, respectively;

the target lesion identification module is respectively connected with the anatomical segmentation module and the structured report module, and is configured to segment all lesions on the third image based on a third preset rule and the first diagnostic data, set a lesion coordinate, namely second diagnostic data, for each lesion coordinate, set a lesion label for each lesion coordinate, output a region of each lesion label, namely a fourth image, compare the second diagnostic data with the first diagnostic data, locate and measure each lesion, and send a location result, a lesion measurement value, a key image, the second diagnostic data, and the fourth image of each lesion to the structured report module;

the structural report module is respectively connected with the image identification module, the image quality judgment module, the variation judgment module, the anatomical segmentation module and the target focus identification module, and is used for automatically generating a diagnosis impression for a doctor to check the positioning result of the focus and the focus measurement value based on a built-in rule; and stores all the received data and all the images.

2. The MRI intelligent evaluation system for whole-body tumors according to claim 1, wherein the anatomical segmentation module further comprises a label determination unit for determining whether the anatomical label is in compliance based on the first diagnostic data, and the determination rule is: judging whether the radial line volume of the maximum connected domain of each anatomical label is within a preset threshold value, judging whether the spatial relationship of the adjacent anatomical labels is correct, judging the shapes of the different anatomical labels, and sending the judgment result to the structured report module; and if the anatomical label is in compliance, sending the first diagnosis data to the target lesion identification module, and if the anatomical label is in non-compliance, sending third prompt information, the type of non-compliance and the measured value of the anatomical label to the structured report module.

3. The MRI intelligent evaluation system for whole-body tumor according to claim 1, wherein the target lesion identification module further comprises a determination unit for determining a relative position of the lesion with respect to the target organ or the target tissue based on the localization result of the lesion, and outputting related data such that: the lesion is within the target organ or the target tissue, the lesion invades the target organ or the target tissue, the lesion is outside the target organ or the target tissue.

4. The MRI intelligent evaluation system for whole-body tumor according to claim 1, wherein the target lesion identification module further comprises a key image generation unit for comparing sizes of all lesions in each of the target organs or each of the target tissues based on the lesion measurement values, generating a key image of a lesion complying with a fourth preset rule, and transmitting the key image to the structured report module.

5. An MRI intelligent assessment method for whole-body tumors is characterized by comprising the following steps:

when the patient finishes the whole body magnetic resonance examination item, the image information management module transmits the DICOM image of the patient to the image identification module through a DICOM protocol;

the image identification module identifies the DICOM image matched with the whole-body magnetic resonance examination item, extracts all part information based on DICOM image header file information, extracts a DICOM image sequence required by each part based on a first preset rule, defines the extracted DICOM image sequence as a first image, sends the first image to an image quality judgment module, and sends the part information, the sequence type of the first image corresponding to the part and the first image to a structured report module; when the DICOM image is not matched with the whole-body magnetic resonance examination item, stopping a diagnosis process and sending first prompt information to the structured report module;

the image quality judging module analyzes the quality of the first image based on a preset condition, the first image meeting the preset condition is respectively sent to a variation judging module and the structural report module, and the first image meeting the preset condition is defined as a second image; if the quality of the first image does not meet the preset condition, stopping the diagnosis process; sending the judgment result and the second prompt message to the structured report module;

the variation judging module judges whether the second image has postoperative change and/or congenital development variation, if the second image does not have the postoperative change and/or congenital development variation, the second image is respectively sent to an anatomical segmentation module and the structural report module, if the second image has the postoperative change and/or congenital development variation, the type of the postoperative change and/or congenital development variation is identified, each parameter of a postoperative residual structure is measured, and the type of the postoperative change and/or congenital development variation and all the parameters are sent to the structural report module;

the anatomical segmentation module segments all target organs and all target tissues on the second image based on a second preset rule, sets anatomical coordinates of each target organ and each target tissue, outputs first diagnostic data, sets an anatomical label for each anatomical coordinate, outputs a region of each anatomical label, namely a third image, and respectively sends the first diagnostic data and the third image to a target focus identification module and the structured report module;

the target lesion identification module divides all lesions on the third image based on a third preset rule and the first diagnosis data, sets lesion coordinates, namely second diagnosis data, for each lesion coordinate, sets a lesion label for each lesion coordinate, outputs a region of each lesion label, namely a fourth image, compares the second diagnosis data with the first diagnosis data, positions and measures each lesion, and sends a positioning result, a lesion measurement value, a key image, the second diagnosis data and the fourth image of each lesion to the structured report module;

the structured report module automatically generates a diagnosis impression according to the positioning result of the focus and the focus measurement value based on a built-in rule for a doctor to check; and stores all the received data and all the images.

6. The MRI intelligent assessment method for whole-body tumor according to claim 5, further comprising: a label judgment unit in the dissection module judges whether the dissection label is in compliance or not based on the first diagnosis data, and the judgment rule is as follows: judging whether the radial line volume of the maximum connected domain of each anatomical label is within a preset threshold value, judging whether the spatial relationship of the adjacent anatomical labels is correct, judging the shapes of the different anatomical labels, and sending the judgment result to the structured report module; and if the anatomical label is in compliance, sending the first diagnosis data to the target lesion identification module, and if the anatomical label is in non-compliance, sending third prompt information, the type of non-compliance and the measured value of the anatomical label to the structured report module.

7. The MRI intelligent assessment method for whole-body tumor according to claim 5, further comprising: the determination unit in the target lesion recognition module determines the relative position of the lesion with respect to the target organ or the target tissue based on the localization result of the lesion, and outputs related data, that is: the lesion is within the target organ or the target tissue, the lesion invades the target organ or the target tissue, the lesion is outside the target organ or the target tissue.

8. The MRI intelligent assessment method for whole-body tumor according to claim 5, further comprising: and the key image generating unit in the target lesion identification module compares the sizes of all lesions in each target organ or each target tissue based on the lesion measurement values, generates the key image of the lesion conforming to a fourth preset rule, and sends the key image to the structured report module.