WO2021012637A1 - 放射影像报告的智能随访方法、系统、设备及存储介质 - Google Patents

放射影像报告的智能随访方法、系统、设备及存储介质 Download PDF

Info

Publication number
WO2021012637A1
WO2021012637A1 PCT/CN2020/070445 CN2020070445W WO2021012637A1 WO 2021012637 A1 WO2021012637 A1 WO 2021012637A1 CN 2020070445 W CN2020070445 W CN 2020070445W WO 2021012637 A1 WO2021012637 A1 WO 2021012637A1
Authority
WO
WIPO (PCT)
Prior art keywords
image
report
diagnosis
pathological
reports
Prior art date
Application number
PCT/CN2020/070445
Other languages
English (en)
French (fr)
Inventor
王耀伟
赵璐偲
王佳皓
陆佳洋
周绮伟
潘志君
岁波
钱起俊
周勇
Original Assignee
卫宁健康科技集团股份有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 卫宁健康科技集团股份有限公司 filed Critical 卫宁健康科技集团股份有限公司
Publication of WO2021012637A1 publication Critical patent/WO2021012637A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/22—Matching criteria, e.g. proximity measures
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H15/00—ICT specially adapted for medical reports, e.g. generation or transmission thereof
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H30/00—ICT specially adapted for the handling or processing of medical images
    • G16H30/20—ICT specially adapted for the handling or processing of medical images for handling medical images, e.g. DICOM, HL7 or PACS
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems

Definitions

  • the invention belongs to the field of consistency judgment between radiological examination and diagnosis and pathological diagnosis, and particularly relates to an intelligent follow-up method, system, equipment and storage medium for radiological image reports.
  • radiological examination As medical X-ray machines, CT and MR and other medical radiological examination equipment technology become more and more mature, radiological examination has become a common clinical examination, which plays an important role in the preliminary screening, positioning and characterization of diseases.
  • radiologists In radiology diagnosis, radiologists need to combine pathological diagnosis to confirm whether the image diagnosis is correct, so as to achieve radiology follow-up and achieve the goal of improving the level of business, but radiologists, especially radiologists in large and medium medical institutions, exist in diagnosis
  • the technical problem to be solved by the present invention is to overcome the defect that the consistency judgment between radiological diagnosis and pathological diagnosis in the prior art requires a lot of manpower, and to provide an intelligent follow-up method, system, equipment and storage medium for radiological image reports.
  • An intelligent follow-up method for radiographic image reports includes:
  • S20 Acquire all image reports of the patient within a preset time period before the time of the pathology report according to the identity information;
  • the intelligent follow-up method further includes:
  • a related dictionary is preset, and the related dictionary stores the corresponding relationship between pathological parts and radiological examination items;
  • step S30 the image diagnosis attribute is extracted on the filtered image report.
  • step S30 specifically includes:
  • step S40 specifically includes:
  • S402 Extract the historical pathological diagnosis attributes in each group of historical pathology reports and the historical image diagnostic attributes in the historical image reports respectively;
  • S404 Input the image diagnosis attribute in each image report and the pathology diagnosis attribute in the pathology report into the diagnostic attribute matching model, and output a result of whether each image report matches the pathology report.
  • the pathological diagnosis attributes include pathological location, pathological qualitativeness, and pathological relevance between pathological location and pathological qualitativeness
  • the image diagnostic attributes include image location, image qualitative, and image relevance of image location and image qualitative.
  • An electronic device includes a memory, a processor, and a computer program stored in the memory and capable of running on the processor.
  • the processor implements the above-mentioned intelligent follow-up method for radiographic image reports when the processor executes the computer program.
  • a computer-readable storage medium has a computer program stored thereon, and when the program is executed by a processor, the steps of the above-mentioned intelligent follow-up method for radiographic image reports are realized.
  • An intelligent follow-up system for radiological image reports includes a pathology report acquisition module, an image report acquisition module, a first extraction module, a first matching module, a labeling module, and a follow-up module;
  • the pathology report acquisition module is used to acquire a pathology report of a patient, and the pathology report includes the pathology report time and the patient's identity information;
  • the image report obtaining module is used to obtain all image reports of the patient within a preset time period before the pathology report time according to the identity information;
  • the first extraction module is used to extract the pathological diagnosis attributes in the pathology report and the image diagnostic attributes in each image report respectively;
  • the first matching module is used to match each group of pathological diagnosis attributes with the image diagnosis attributes in each image report. If the matching is unsuccessful, it is determined that the image diagnosis in the unsuccessful matching image report is incorrect Yes, if the matching is successful, it is determined that the image diagnosis in the image report of the successful matching is correct;
  • the labeling module is used to label the label that characterizes the matching result on the image report
  • the follow-up module is used to follow up the corresponding patient according to the label.
  • the intelligent follow-up system further includes a preset module, a second extraction module, a second matching module, and a filtering module;
  • the preset module is used to preset an associated dictionary, and the associated dictionary stores the corresponding relationship between pathological parts and radiological examination items;
  • the second extraction module is used to extract the current pathological location in the pathology report and the current radiological examination items in each image report;
  • the second matching module is configured to match the current pathological location with the current radiological examination item based on the associated dictionary
  • the filtering module is used to filter out image reports that do not match the current radiological examination item with the current pathological site;
  • the first extraction module is used for extracting the image diagnosis attribute of the filtered image report.
  • the first extraction module includes a first report acquisition unit and a first training unit;
  • the first report obtaining unit is configured to obtain a plurality of historical reports, and the diagnosis attributes in the historical reports have been marked;
  • the first training unit is configured to use the historical report as training data and obtain a diagnostic attribute recognition model based on conditional random field algorithm training;
  • the first extraction module is configured to input the pathology report and the image report into the diagnosis attribute recognition model, and output the pathology diagnosis attribute and the image diagnosis attribute.
  • the first matching module includes a second report acquisition unit, an attribute extraction unit and a second training unit;
  • the second report obtaining unit is used to obtain multiple sets of historical pathology reports and historical image reports with known consistency of diagnosis results;
  • the attribute extraction unit is used to extract the historical pathological diagnosis attributes in each group of historical pathology reports and the historical image diagnostic attributes in the historical image reports respectively;
  • the second training unit is used to use each group of historical pathological diagnosis attributes and historical image diagnosis attributes as training data, and obtain a diagnostic attribute matching model based on the Bert pre-training model and Word2Vec algorithm training;
  • the first matching module is used to input the image diagnosis attribute in each image report and the pathology diagnosis attribute in the pathology report into the diagnosis attribute matching model, and output whether each image report and the pathology report are The result of the match.
  • the positive progress effect of the present invention is that the present invention extracts the diagnostic attributes of the acquired pathology report of the patient and the image report within a preset time, and further realizes whether the diagnosis of the pathology report and the image report are consistent through the matching of the diagnostic attributes. In order to realize the automatic and intelligent follow-up of radiological image report.
  • Fig. 1 is a flow chart of the intelligent follow-up method for radiographic image reports according to Embodiment 1 of the present invention.
  • FIG. 2 is a flowchart of step S30 in the intelligent follow-up method for radiographic image reports according to Embodiment 2 of the present invention.
  • FIG. 3 is a flowchart of step S40 in the intelligent follow-up method for radiographic image reports according to Embodiment 3 of the present invention.
  • Fig. 4 is a schematic structural diagram of an electronic device according to Embodiment 4 of the present invention.
  • FIG. 5 is a schematic diagram of modules of an intelligent follow-up system for radiological image reports according to Embodiment 6 of the present invention.
  • FIG. 6 is a schematic diagram of the first extraction module in the intelligent follow-up system for radiological image reports according to Embodiment 7 of the present invention.
  • FIG. 7 is a schematic diagram of the first matching module in the intelligent follow-up system for radiological image reports according to Embodiment 8 of the present invention.
  • An intelligent follow-up method for radiographic image reports as shown in Fig. 1, the intelligent follow-up method includes:
  • the pathology report includes the pathology report time and the patient's identity information
  • S20 Acquire all image reports of the patient within a preset time period before the time of the pathology report according to the identity information;
  • the pathological diagnosis attributes include pathological location, pathological qualitative, and pathological relevance between pathological location and pathological qualitative
  • the image diagnostic attributes include image location, image qualitative, and image relevance of image location and image qualitative.
  • each corresponding diagnostic attribute needs to be matched one by one.
  • the pathological location in the pathology report is matched with the image location in the image report; in addition, more information may be extracted from the pathology report.
  • Multiple groups of image diagnostic attributes may also be extracted from the image report. It is necessary to match each group of pathological diagnostic attributes in the pathology report with the multiple groups of image diagnostic attributes in the image report one by one to match the pathological location with The matching of image positioning is explained as an example. When one of the multiple image positioning matches the pathological positioning, it is confirmed that the image diagnosis in the image report is correct. Only when the pathological positioning does not match all the image positionings , It is determined that the image positioning in the image diagnosis is wrong.
  • different diagnostic attributes may also be marked separately, such as correct positioning and incorrect qualitativeness.
  • the image report after the image report is obtained according to time, the image report can be further filtered based on the diagnosis location.
  • the intelligent follow-up method further includes:
  • a related dictionary is preset; the related dictionary stores the corresponding relationship between pathological parts and radiological examination items;
  • step S30 the image diagnosis attribute is extracted on the filtered image report.
  • the diagnosis attribute is extracted from the acquired pathology report of the patient and the image report within a preset time, and the diagnosis of the pathology report and the image report is determined by matching the diagnosis attribute to determine whether the diagnosis is consistent. Automated and intelligent follow-up of radiology report.
  • step S30 specifically includes:
  • diagnostic attributes positioning, qualitative, positioning and qualitative attribute relations
  • the obtained diagnostic attribute recognition model can identify the diagnostic attributes in the new report.
  • CRF conditional random field
  • the report text "Left hip joint osteoarthritis, osteoporosis, femoral head avascular necrosis waiting to be discharged, please combine with further clinical examination.
  • the right sacroiliac joint is dense, please follow up.”
  • osteoarthritis and osteoporosis are all "qualitative”
  • the left hip joint is related to these two "qualitative” attributes.
  • step S40 specifically includes:
  • S402 Extract the historical pathological diagnosis attributes in each group of historical pathology reports and the historical image diagnostic attributes in the historical image reports respectively;
  • S404 Input the image diagnosis attribute in each image report and the pathology diagnosis attribute in the pathology report into the diagnostic attribute matching model, and output a result of whether each image report matches the pathology report.
  • historical follow-up data including historical pathology reports and historical image reports that are known to be consistent with the diagnosis results
  • the Bert pre-training model and Word2Vec are used for text feature extraction, and then the diagnostic attributes are obtained by training
  • the matching model is used to obtain the matching results of the part (location) and the disease (qualitative) in the pathology report and the part and the disease in the radiology report.
  • An electronic device including a memory, a processor, and a computer program stored on the memory and capable of running on the processor.
  • the processor executes the computer program to implement the description of any one of the embodiments 1-3 Intelligent follow-up method for radiographic report.
  • FIG. 4 is a schematic structural diagram of an electronic device provided by this embodiment.
  • Figure 4 shows a block diagram of an exemplary electronic device 90 suitable for implementing embodiments of the present invention.
  • the electronic device 90 shown in FIG. 4 is only an example, and should not bring any limitation to the function and application scope of the embodiment of the present invention.
  • the electronic device 90 may be in the form of a general-purpose computing device, for example, it may be a server device.
  • the components of the electronic device 90 may include but are not limited to: at least one processor 91, at least one memory 92, and a bus 93 connecting different system components (including the memory 92 and the processor 91).
  • the bus 93 includes a data bus, an address bus, and a control bus.
  • the memory 92 may include a volatile memory, such as a random access memory (RAM) 921 and/or a cache memory 922, and may further include a read-only memory (ROM) 923.
  • RAM random access memory
  • ROM read-only memory
  • the memory 92 may also include a program tool 925 having a set (at least one) program module 924.
  • program module 924 includes but is not limited to: an operating system, one or more application programs, other program modules, and program data. In these examples Each or some combination of may include the realization of the network environment.
  • the processor 91 executes various functional applications and data processing by running a computer program stored in the memory 92.
  • the electronic device 90 may also communicate with one or more external devices 94 (such as keyboards, pointing devices, etc.). This communication can be performed through an input/output (I/O) interface 95.
  • the electronic device 90 may also communicate with one or more networks (for example, a local area network (LAN), a wide area network (WAN), and/or a public network, such as the Internet) through the network adapter 96.
  • the network adapter 96 communicates with other modules of the electronic device 90 through the bus 93.
  • a computer-readable storage medium has a computer program stored thereon, and when the program is executed by a processor, the steps of the intelligent follow-up method for radiographic image reports described in any one of the embodiments 1-3 are realized.
  • the readable storage medium may more specifically include but not limited to: portable disk, hard disk, random access memory, read only memory, erasable programmable read only memory, optical storage device, magnetic storage device or any of the above The right combination.
  • the present invention can also be implemented in the form of a program product, which includes program code.
  • program product runs on a terminal device
  • the program code is used to make the terminal device execute the implementation. Steps of the intelligent follow-up method for radiographic image reports described in any of the embodiments 1-3.
  • program code used to execute the present invention can be written in any combination of one or more programming languages, and the program code can be completely executed on the user equipment, partially executed on the user equipment, as an independent
  • the software package is executed, partly on the user’s device, partly on the remote device, or entirely on the remote device.
  • the intelligent follow-up system includes a pathology report acquisition module 1, an image report acquisition module 2, a first extraction module 3, a first matching module 4, an annotation module 5, and Follow-up module 6;
  • the pathology report obtaining module 1 is used to obtain a pathology report of a patient, and the pathology report includes the time of the pathology report and the identity information of the patient;
  • the image report obtaining module 2 is configured to obtain all image reports of the patient within a preset time period before the pathology report time according to the identity information;
  • the first extraction module 3 is used to extract the pathological diagnosis attributes in the pathology report and the image diagnostic attributes in each image report respectively;
  • the first matching module 4 is used to match each group of pathological diagnosis attributes with the image diagnosis attributes in each image report, and if the matching is unsuccessful, it is determined that the image diagnosis in the image report with unsuccessful matching is correct If it is wrong, if the matching is successful, it is determined that the image diagnosis in the image report of the successful matching is correct;
  • the labeling module 5 is used to label a label that characterizes the matching result on the image report;
  • the follow-up module 6 is used to follow up the corresponding patient according to the label.
  • the pathological diagnosis attributes include pathological location, pathological qualitative, and pathological relevance between pathological location and pathological qualitative
  • the image diagnostic attributes include image location, image qualitative, and image relevance of image location and image qualitative.
  • each corresponding diagnostic attribute needs to be matched one by one.
  • the pathological location in the pathology report is matched with the image location in the image report; in addition, more information may be extracted from the pathology report.
  • Groups of pathological diagnosis attributes Multiple groups of image diagnostic attributes may also be extracted from the image report. It is necessary to match each group of pathological diagnostic attributes in the pathology report with the multiple groups of image diagnostic attributes in the image report one by one to match
  • the matching of image positioning is explained as an example. When one of the multiple image positioning matches the pathological positioning, it is confirmed that the image diagnosis in the image report is correct. Only when the pathological positioning does not match all the image positionings , It is determined that the image positioning in the image diagnosis is wrong.
  • different diagnostic attributes may also be marked separately, such as correct positioning and incorrect qualitativeness.
  • the intelligent follow-up system further includes a preset module 7, a second extraction module 8, and a second matching Module 9 and filtering module 10;
  • the preset module 7 is used to preset a related dictionary, and the related dictionary stores the corresponding relationship between pathological parts and radiological examination items;
  • the second extraction module 8 is used to extract the current pathological location in the pathology report and the current radiological examination items in each image report;
  • the second matching module 9 is configured to match the current pathological location with the current radiological examination item based on the associated dictionary
  • the filtering module 10 is used to filter out image reports that do not match the current radiological examination item with the current pathological site;
  • the first extraction module 3 is used for extracting the image diagnosis attribute of the filtered image report.
  • the diagnosis attribute is extracted from the acquired pathology report of the patient and the image report within a preset time, and the diagnosis of the pathology report and the image report is determined by matching the diagnosis attribute to determine whether the diagnosis is consistent. Automated and intelligent follow-up of radiology report.
  • the intelligent follow-up system for radiological image reports in this embodiment is a further improvement on the basis of Embodiment 6.
  • the first extraction module 3 includes a first report acquisition unit 31 and a first training unit 32;
  • the first report obtaining unit 31 is configured to obtain multiple historical reports, and the diagnosis attributes in the historical reports have been marked;
  • the first training unit 32 is configured to use the historical report as training data and obtain a diagnostic attribute recognition model based on conditional random field algorithm training;
  • the first extraction module 3 is configured to input the pathology report and the image report into the diagnosis attribute recognition model, and output the pathology diagnosis attribute and the image diagnosis attribute.
  • diagnostic attributes positioning, qualitative, positioning and qualitative attribute relations
  • the report text are marked, and then based on conditional random field (CRF) (which can also be combined with neural Network algorithm) method to train the model, the obtained diagnostic attribute recognition model can identify the diagnostic attributes in the new report.
  • CRF conditional random field
  • the intelligent follow-up system for radiographic image reports in this embodiment is further improved on the basis of Embodiment 6.
  • the first matching module 4 includes a second report acquisition unit 41, an attribute extraction unit 42 and a second Training unit 43;
  • the second report obtaining unit 41 is configured to obtain multiple sets of historical pathology reports and historical image reports with known consistency of diagnosis results;
  • the attribute extraction unit 42 is configured to extract the historical pathological diagnosis attributes in each group of historical pathology reports and the historical image diagnostic attributes in the historical image reports respectively;
  • the second training unit 43 is configured to use each group of historical pathological diagnosis attributes and historical image diagnosis attributes as training data, and obtain a diagnostic attribute matching model based on the Bert pre-training model and Word2Vec algorithm training;
  • the first matching module 4 is configured to input the image diagnosis attribute in each image report and the pathology diagnosis attribute in the pathology report into the diagnosis attribute matching model, and output each image report and the pathology report Whether the result matches.
  • historical follow-up data including historical pathology reports and historical image reports that are known to be consistent with the diagnosis results
  • the Bert pre-training model and Word2Vec are used for text feature extraction, and then the diagnostic attributes are obtained by training
  • the matching model is used to obtain the matching results of the part (location) and the disease (qualitative) in the pathology report and the part and the disease in the radiology report.

Landscapes

  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Public Health (AREA)
  • Medical Informatics (AREA)
  • Data Mining & Analysis (AREA)
  • Primary Health Care (AREA)
  • General Health & Medical Sciences (AREA)
  • Epidemiology (AREA)
  • Theoretical Computer Science (AREA)
  • Biomedical Technology (AREA)
  • Evolutionary Biology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Computation (AREA)
  • Physics & Mathematics (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Radiology & Medical Imaging (AREA)
  • General Physics & Mathematics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Artificial Intelligence (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • General Engineering & Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Pathology (AREA)
  • Medical Treatment And Welfare Office Work (AREA)
  • Apparatus For Radiation Diagnosis (AREA)

Abstract

一种放射影像报告的智能随访方法、系统、设备及存储介质，所述智能随访方法包括：获取一患者的一病理报告（S10），所述病理报告包括病理报告时间和所述患者的身份信息；根据身份信息获取患者在病理报告时间前的一预设时间段内的所有影像报告（S20）；分别提取病理报告中的病理诊断属性及每个影像报告中的影像诊断属性（S30）；将每组病理诊断属性分别与每个影像报告中的影像诊断属性进行匹配（S40）；若匹配不成功，则确定影像诊断是有误的；若匹配成功，则确定影像诊断是正确的（S50）；将表征匹配结果的标签标注在影像报告上（S60）；根据标签对对应的患者进行随访（S70）。所述智能随访方法实现影像诊断与病理诊断的一致性的自动判断，从而实现放射影像报告的自动化、智能化的随访。

Description

放射影像报告的智能随访方法、系统、设备及存储介质
本申请要求申请日为2019/7/22的中国专利申请2019106604101的优先权。本申请引用上述中国专利申请的全文。
技术领域
本发明属于放射检查诊断与病理诊断的一致性判断领域，特别涉及一种放射影像报告的智能随访方法、系统、设备及存储介质。
背景技术
随着医用X光机、CT和MR等医学放射检查设备技术越来越成熟，放射检查已越来越成为一种临床常见检查，用于对疾病的初筛、定位和定性起着重要作用。
在放射科诊断中放射科医生需要结合病理诊断确认影像诊断是否正确，从而实现放射科随访，达到提升业务水平的目标，但放射科医生，特别是中大型医疗机构放射科医生在诊断中均存在以下痛点：其一、工作量巨大，往往无暇查看影像报告诊断是否符合临床和病理的诊断，无法确定影像诊断的正确性；其二、查看相关病理报告极为不便，需要去其他科室找到该患者对应的病理检查报告，人为进行判断，确定影像学诊断是否正确。。
发明内容
本发明要解决的技术问题是为了克服现有技术中放射诊断与病理诊断的一致性判断需要耗费很大人力的缺陷，提供一种放射影像报告的智能随访方法、系统、设备及存储介质。
本发明是通过下述技术方案来解决上述技术问题：
一种放射影像报告的智能随访方法，所述智能随访方法包括：
S10、获取一患者的一病理报告，所述病理报告包括病理报告时间和所述患者的身份信息；
S20、根据所述身份信息获取所述患者在所述病理报告时间前的一预设时间段内的所有影像报告；
S30、分别提取所述病理报告中的病理诊断属性及每个影像报告中的影像诊断属性；
S40、将每组病理诊断属性分别与所述每个影像报告中的影像诊断属性进行匹配；
S50、若匹配不成功，则确定匹配不成功的影像报告中的影像诊断是有误的；若匹配成功，则确定匹配成功的影像报告中的影像诊断是正确的；
S60、将表征匹配结果的标签标注在影像报告上；
S70、根据所述标签对对应的患者进行随访。
较佳地，步骤S30之前，所述智能随访方法还包括：
S21、预设一关联词典，所述关联词典存储有病理部位与放射检查项目的对应关系；
S22、提取所述病理报告中的当前病理部位及每个影像报告中的当前放射检查项目；
S23、基于所述关联词典将所述当前病理部位与所述当前放射检查项目进行匹配；
S24、滤除所述当前放射检查项目与所述当前病理部位不匹配的影像报告；
步骤S30中对滤除后的影像报告进行影像诊断属性的提取。
较佳地，步骤S30具体包括：
S301、获取多个历史报告，所述历史报告中的诊断属性已被标注；
S302、将所述历史报告作为训练数据，并基于条件随机场算法训练得到诊断属性识别模型；
S303、将所述病理报告和所述影像报告输入所述诊断属性识别模型，输出所述病理诊断属性和所述影像诊断属性。
较佳地，步骤S40具体包括：
S401、获取多组诊断结果一致性已知的历史病理报告和历史影像报告；
S402、分别提取每组历史病理报告中的历史病理诊断属性和历史影像报告中的历史影像诊断属性；
S403、将每组历史病理诊断属性和历史影像诊断属性作为一个训练数据，并基于Bert(一种词向量算法)预训练模型与Word2Vec(一种词向量算法)算法训练得到诊断属性匹配模型；
S404、分别将每个影像报告中的影像诊断属性与所述病理报告中的病理诊断属性输入所述诊断属性匹配模型，输出所述每个影像报告与所述病理报告是否匹配的结果。
较佳地，所述病理诊断属性包括病理定位、病理定性和病理定位与病理定性的病理关联性，所述影像诊断属性包括影像定位、影像定性和影像定位与影像定性的影像关联性。
一种电子设备，包括存储器、处理器及存储在存储器上并可在处理器上运行的计算机程序，所述处理器执行所述计算机程序时实现上述的放射影像报告的智能随访方法。
一种计算机可读存储介质，其上存储有计算机程序，所述程序被处理器执行时实现 上述的放射影像报告的智能随访方法的步骤。
一种放射影像报告的智能随访系统，所述智能随访系统包括病理报告获取模块、影像报告获取模块、第一提取模块、第一匹配模块、标注模块和随访模块；
所述病理报告获取模块用于获取一患者的一病理报告，所述病理报告包括病理报告时间和所述患者的身份信息；
所述影像报告获取模块用于根据所述身份信息获取所述患者在所述病理报告时间前的一预设时间段内的所有影像报告；
所述第一提取模块用于分别提取所述病理报告中的病理诊断属性及每个影像报告中的影像诊断属性；
所述第一匹配模块用于将每组病理诊断属性分别与所述每个影像报告中的影像诊断属性进行匹配，若匹配不成功，则确定匹配不成功的影像报告中的影像诊断是有误的，若匹配成功，则确定匹配成功的影像报告中的影像诊断是正确的；
所述标注模块用于将表征匹配结果的标签标注在影像报告上；
所述随访模块用于根据所述标签对对应的患者进行随访。
较佳地，所述智能随访系统还包括预设模块、第二提取模块、第二匹配模块和滤除模块；
所述预设模块用于预设一关联词典，所述关联词典存储有病理部位与放射检查项目的对应关系；
所述第二提取模块用于提取所述病理报告中的当前病理部位及每个影像报告中的当前放射检查项目；
所述第二匹配模块用于基于所述关联词典将所述当前病理部位与所述当前放射检查项目进行匹配；
所述滤除模块用于滤除所述当前放射检查项目与所述当前病理部位不匹配的影像报告；
所述第一提取模块用于对滤除后的影像报告进行影像诊断属性的提取。
较佳地，所述第一提取模块包括第一报告获取单元和第一训练单元；
所述第一报告获取单元用于获取多个历史报告，所述历史报告中的诊断属性已被标注；
所述第一训练单元用于将所述历史报告作为训练数据，并基于条件随机场算法训练得到诊断属性识别模型；
所述第一提取模块用于将所述病理报告和所述影像报告输入所述诊断属性识别模型， 输出所述病理诊断属性和所述影像诊断属性。
较佳地，所述第一匹配模块包括第二报告获取单元、属性提取单元和第二训练单元；
所述第二报告获取单元用于获取多组诊断结果一致性已知的历史病理报告和历史影像报告；
所述属性提取单元用于分别提取每组历史病理报告中的历史病理诊断属性和历史影像报告中的历史影像诊断属性；
所述第二训练单元用于将每组历史病理诊断属性和历史影像诊断属性作为一个训练数据，并基于Bert预训练模型与Word2Vec算法训练得到诊断属性匹配模型；
所述第一匹配模块用于分别将每个影像报告中的影像诊断属性与所述病理报告中的病理诊断属性输入所述诊断属性匹配模型，输出所述每个影像报告与所述病理报告是否匹配的结果。
本发明的积极进步效果在于：本发明通过对获取到的患者的病理报告及一预设时间内的影像报告进行诊断属性的提取，进一步通过诊断属性的匹配实现病理报告和影像报告的诊断是否一致的判定，从而实现放射影像报告的自动化、智能化的随访。
附图说明
图1为本发明实施例1的放射影像报告的智能随访方法的流程图。
图2为本发明实施例2的放射影像报告的智能随访方法中步骤S30的流程图。
图3为本发明实施例3的放射影像报告的智能随访方法中步骤S40的流程图。
图4为本发明实施例4的电子设备的结构示意图。
图5为本发明实施例6的放射影像报告的智能随访系统的模块示意图。
图6为本发明实施例7的放射影像报告的智能随访系统中第一提取模块的模块示意图。
图7为本发明实施例8的放射影像报告的智能随访系统中第一匹配模块的模块示意图。
具体实施方式
下面通过实施例的方式进一步说明本发明，但并不因此将本发明限制在所述的实施例范围之中。
实施例1
一种放射影像报告的智能随访方法，如图1所示，所述智能随访方法包括：
S10、获取一患者的一病理报告；所述病理报告包括病理报告时间和所述患者的身份信息；
S20、根据所述身份信息获取所述患者在所述病理报告时间前的一预设时间段内的所有影像报告；
S30、分别提取所述病理报告中的病理诊断属性及每个影像报告中的影像诊断属性；
S40、将每组病理诊断属性分别与所述每个影像报告中的影像诊断属性进行匹配；
S50、若匹配不成功，则确定匹配不成功的影像报告中的影像诊断是有误的；若匹配成功，则确定匹配成功的影像报告中的影像诊断是正确的；
S60、将表征匹配结果的标签标注在影像报告上；
S70、根据所述标签对对应的患者进行随访。
所述病理诊断属性包括病理定位、病理定性和病理定位与病理定性的病理关联性，所述影像诊断属性包括影像定位、影像定性和影像定位与影像定性的影像关联性。
需要说明的是，实际匹配时，需要将每个对应的诊断属性一一进行匹配，比如：病理报告中的病理定位与影像报告中的影像定位进行匹配；另外，病理报告中可能会提取出多组病理诊断属性，影像报告中也可能会提取出多组影像诊断属性，需要分别将病理报告中的每组病理诊断属性与影像报告中的多组影像诊断属性一一进行匹配，以病理定位与影像定位的匹配为例进行说明，当多个影像定位中有一个影像定位与病理定位匹配一致，则确认影像报告中的影像诊断是正确的，只有当病理定位与所有的影像定位都不匹配时，才确定影像诊断中的影像定位是有误的。另外，本实施例中，除了将匹配结果进行标注外，对于不同的诊断属性也可分别进行标注，比如定位正确、定性不正确等。
本实施例中，在根据时间获取影像报告后，可以基于诊断部位对影像报告进行进一步筛选过滤，参考图1，步骤S30之前，所述智能随访方法还包括：
S21、预设一关联词典；所述关联词典存储有病理部位与放射检查项目的对应关系；
S22、提取所述病理报告中的当前病理部位及每个影像报告中的当前放射检查项目；
S23、基于所述关联词典将所述当前病理部位与所述当前放射检查项目进行匹配；
S24、滤除所述当前放射检查项目与所述当前病理部位不匹配的影像报告；
步骤S30中对滤除后的影像报告进行影像诊断属性的提取。
本实施例中，通过对获取到的患者的病理报告及一预设时间内的影像报告进行诊断属性的提取，进一步通过诊断属性的匹配实现病理报告和影像报告的诊断是否一致的判定，从而实现放射影像报告的自动化、智能化的随访。
实施例2
本实施例的放射影像报告的智能随访方法是在实施例1的基础上进一步改进，如图2所示，步骤S30具体包括：
S301、获取多个历史报告，所述历史报告中的诊断属性已被标注；
S302、将所述历史报告作为训练数据，并基于条件随机场算法训练得到诊断属性识别模型；
S303、将所述病理报告和所述影像报告输入所述诊断属性识别模型，输出所述病理诊断属性和所述影像诊断属性。
本实施例中，通过医学顾问对历史报告进行数据标注，将报告文本中的诊断属性(定位、定性、定位与定性的属性关系)标注出来，然后基于条件随机场(CRF)(也可以结合神经网络算法)的方法训练模型，得到的诊断属性识别模型能够识别出新报告中的诊断属性。例如：报告文本“左髋关节骨性关节炎、骨质疏松，股骨头缺血坏死待排，请结合临床进一步检查。右侧骶髂关节处致密影，请随访。”中，左髋关节是“定位”，骨性关节炎与骨质疏松都属于“定性”，左髋关节与这两个“定性”属性都有关联。
实施例3
本实施例的放射影像报告的智能随访方法是在实施例1的基础上进一步改进，如图3所示，步骤S40具体包括：
S401、获取多组诊断结果一致性已知的历史病理报告和历史影像报告；
S402、分别提取每组历史病理报告中的历史病理诊断属性和历史影像报告中的历史影像诊断属性；
S403、将每组历史病理诊断属性和历史影像诊断属性作为一个训练数据，并基于Bert预训练模型与Word2Vec算法训练得到诊断属性匹配模型；
S404、分别将每个影像报告中的影像诊断属性与所述病理报告中的病理诊断属性输入所述诊断属性匹配模型，输出所述每个影像报告与所述病理报告是否匹配的结果。
本实施例中，将历史随访数据(包含已知诊断结果是否一致的历史病理报告和历史影像报告)作为模型的训练数据，使用Bert预训练模型与Word2Vec做文本的特征提取，进而训练得到诊断属性匹配模型，用于获取病理报告中的部位(定位)与疾病(定性)和放射报告中的部位与疾病的匹配结果。
实施例4
一种电子设备，包括存储器、处理器及存储在存储器上并可在处理器上运行的计算机程序，所述处理器执行所述计算机程序时实现实施例1-3中任意一个实施例所述的放射影像报告的智能随访方法。
图4为本实施例提供的一种电子设备的结构示意图。图4示出了适于用来实现本发明实施方式的示例性电子设备90的框图。图4显示的电子设备90仅仅是一个示例，不应对本发明实施例的功能和使用范围带来任何限制。
如图4所示，电子设备90可以以通用计算设备的形式表现，例如其可以为服务器设备。电子设备90的组件可以包括但不限于：至少一个处理器91、至少一个存储器92、连接不同系统组件(包括存储器92和处理器91)的总线93。
总线93包括数据总线、地址总线和控制总线。
存储器92可以包括易失性存储器，例如随机存取存储器(RAM)921和/或高速缓存存储器922，还可以进一步包括只读存储器(ROM)923。
存储器92还可以包括具有一组(至少一个)程序模块924的程序工具925，这样的程序模块924包括但不限于：操作系统、一个或者多个应用程序、其它程序模块以及程序数据，这些示例中的每一个或某种组合中可能包括网络环境的实现。
处理器91通过运行存储在存储器92中的计算机程序，从而执行各种功能应用以及数据处理。
电子设备90也可以与一个或多个外部设备94(例如键盘、指向设备等)通信。这种通信可以通过输入/输出(I/O)接口95进行。并且，电子设备90还可以通过网络适配器96与一个或者多个网络(例如局域网(LAN)，广域网(WAN)和/或公共网络，例如因特网)通信。网络适配器96通过总线93与电子设备90的其它模块通信。应当明白，尽管图中未示出，可以结合电子设备90使用其它硬件和/或软件模块，包括但不限于：微代码、设备驱动器、冗余处理器、外部磁盘驱动阵列、RAID(磁盘阵列)系统、磁带驱动器以及数据备份存储系统等。
应当注意，尽管在上文详细描述中提及了电子设备的若干单元/模块或子单元/模块，但是这种划分仅仅是示例性的并非强制性的。实际上，根据本申请的实施方式，上文描述的两个或更多单元/模块的特征和功能可以在一个单元/模块中具体化。反之，上文描述的一个单元/模块的特征和功能可以进一步划分为由多个单元/模块来具体化。
实施例5
一种计算机可读存储介质，其上存储有计算机程序，所述程序被处理器执行时实现实施例1-3中任意一个实施例所述的放射影像报告的智能随访方法的步骤。
其中，可读存储介质可以采用的更具体可以包括但不限于：便携式盘、硬盘、随机存取存储器、只读存储器、可擦拭可编程只读存储器、光存储器件、磁存储器件或上述的任意合适的组合。
在可能的实施方式中，本发明还可以实现为一种程序产品的形式，其包括程序代码，当所述程序产品在终端设备上运行时，所述程序代码用于使所述终端设备执行实现实施例1-3中任意一个实施例所述的放射影像报告的智能随访方法的步骤。
其中，可以以一种或多种程序设计语言的任意组合来编写用于执行本发明的程序代码，所述程序代码可以完全地在用户设备上执行、部分地在用户设备上执行、作为一个独立的软件包执行、部分在用户设备上部分在远程设备上执行或完全在远程设备上执行。
实施例6
一种放射影像报告的智能随访系统，如图5所示，所述智能随访系统包括病理报告获取模块1、影像报告获取模块2、第一提取模块3、第一匹配模块4、标注模块5和随访模块6；
所述病理报告获取模块1用于获取一患者的一病理报告，所述病理报告包括病理报告时间和所述患者的身份信息；
所述影像报告获取模块2用于根据所述身份信息获取所述患者在所述病理报告时间前的一预设时间段内的所有影像报告；
所述第一提取模块3用于分别提取所述病理报告中的病理诊断属性及每个影像报告中的影像诊断属性；
所述第一匹配模块4用于将每组病理诊断属性分别与所述每个影像报告中的影像诊断属性进行匹配，若匹配不成功，则确定匹配不成功的影像报告中的影像诊断是有误的，若匹配成功，则确定匹配成功的影像报告中的影像诊断是正确的；
所述标注模块5用于将表征匹配结果的标签标注在影像报告上；
所述随访模块6用于根据所述标签对对应的患者进行随访。
所述病理诊断属性包括病理定位、病理定性和病理定位与病理定性的病理关联性，所述影像诊断属性包括影像定位、影像定性和影像定位与影像定性的影像关联性。
需要说明的是，实际匹配时，需要将每个对应的诊断属性一一进行匹配，比如：病理报告中的病理定位与影像报告中的影像定位进行匹配；另外，病理报告中可能会提取出多组病理诊断属性，影像报告中也可能会提取出多组影像诊断属性，需要分别将病理报告中的每组病理诊断属性与影像报告中的多组影像诊断属性一一进行匹配，以病理定位与影像定位的匹配为例进行说明，当多个影像定位中有一个影像定位与病理定位匹配一致，则确认影像报告中的影像诊断是正确的，只有当病理定位与所有的影像定位都不匹配时，才确定影像诊断中的影像定位是有误的。另外，本实施例中，除了将匹配结果进行标注外，对于不同的诊断属性也可分别进行标注，比如定位正确、定性不正确等。
本实施例中，在根据时间获取影像报告后，可以基于诊断部位对影像报告进行进一步筛选过滤，参考图5，所述智能随访系统还包括预设模块7、第二提取模块8、第二匹配模块9和滤除模块10；
所述预设模块7用于预设一关联词典，所述关联词典存储有病理部位与放射检查项目的对应关系；
所述第二提取模块8用于提取所述病理报告中的当前病理部位及每个影像报告中的当前放射检查项目；
所述第二匹配模块9用于基于所述关联词典将所述当前病理部位与所述当前放射检查项目进行匹配；
所述滤除模块10用于滤除所述当前放射检查项目与所述当前病理部位不匹配的影像报告；
所述第一提取模块3用于对滤除后的影像报告进行影像诊断属性的提取。
本实施例中，通过对获取到的患者的病理报告及一预设时间内的影像报告进行诊断属性的提取，进一步通过诊断属性的匹配实现病理报告和影像报告的诊断是否一致的判定，从而实现放射影像报告的自动化、智能化的随访。
实施例7
本实施例的放射影像报告的智能随访系统是在实施例6的基础上进一步改进，如图6所示，所述第一提取模块3包括第一报告获取单元31和第一训练单元32；
所述第一报告获取单元31用于获取多个历史报告，所述历史报告中的诊断属性已被标注；
所述第一训练单元32用于将所述历史报告作为训练数据，并基于条件随机场算法训练得到诊断属性识别模型；
所述第一提取模块3用于将所述病理报告和所述影像报告输入所述诊断属性识别模型，输出所述病理诊断属性和所述影像诊断属性。
本实施例中，通过医学顾问对历史报告进行数据标注，将报告文本中的诊断属性(定位、定性、定位与定性的属性关系)标注出来，然后基于条件随机场(CRF)(也可以结合神经网络算法)的方法训练模型，得到的诊断属性识别模型能够识别出新报告中的诊断属性。
实施例8
本实施例的放射影像报告的智能随访系统是在实施例6的基础上进一步改进，如图7所示，所述第一匹配模块4包括第二报告获取单元41、属性提取单元42和第二训练单 元43；
所述第二报告获取单元41用于获取多组诊断结果一致性已知的历史病理报告和历史影像报告；
所述属性提取单元42用于分别提取每组历史病理报告中的历史病理诊断属性和历史影像报告中的历史影像诊断属性；
所述第二训练单元43用于将每组历史病理诊断属性和历史影像诊断属性作为一个训练数据，并基于Bert预训练模型与Word2Vec算法训练得到诊断属性匹配模型；
所述第一匹配模块4用于分别将每个影像报告中的影像诊断属性与所述病理报告中的病理诊断属性输入所述诊断属性匹配模型，输出所述每个影像报告与所述病理报告是否匹配的结果。
本实施例中，将历史随访数据(包含已知诊断结果是否一致的历史病理报告和历史影像报告)作为模型的训练数据，使用Bert预训练模型与Word2Vec做文本的特征提取，进而训练得到诊断属性匹配模型，用于获取病理报告中的部位(定位)与疾病(定性)和放射报告中的部位与疾病的匹配结果。
虽然以上描述了本发明的具体实施方式，但是本领域的技术人员应当理解，这仅是举例说明，本发明的保护范围是由所附权利要求书限定的。本领域的技术人员在不背离本发明的原理和实质的前提下，可以对这些实施方式做出多种变更或修改，但这些变更和修改均落入本发明的保护范围。

Claims (11)

  1. 一种放射影像报告的智能随访方法，其特征在于，所述智能随访方法包括：
    S10、获取一患者的一病理报告，所述病理报告包括病理报告时间和所述患者的身份信息；
    S20、根据所述身份信息获取所述患者在所述病理报告时间前的一预设时间段内的所有影像报告；
    S30、分别提取所述病理报告中的病理诊断属性及每个影像报告中的影像诊断属性；
    S40、将每组病理诊断属性分别与所述每个影像报告中的影像诊断属性进行匹配；
    S50、若匹配不成功，则确定匹配不成功的影像报告中的影像诊断是有误的；若匹配成功，则确定匹配成功的影像报告中的影像诊断是正确的；
    S60、将表征匹配结果的标签标注在影像报告上；
    S70、根据所述标签对对应的患者进行随访。
  2. 如权利要求1所述的放射影像报告的智能随访方法，其特征在于，步骤S30之前，所述智能随访方法还包括：
    S21、预设一关联词典，所述关联词典存储有病理部位与放射检查项目的对应关系；
    S22、提取所述病理报告中的当前病理部位及每个影像报告中的当前放射检查项目；
    S23、基于所述关联词典将所述当前病理部位与所述当前放射检查项目进行匹配；
    S24、滤除所述当前放射检查项目与所述当前病理部位不匹配的影像报告；
    步骤S30中对滤除后的影像报告进行影像诊断属性的提取。
  3. 如权利要求1所述的放射影像报告的智能随访方法，其特征在于，步骤S30具体包括：
    S301、获取多个历史报告，所述历史报告中的诊断属性已被标注；
    S302、将所述历史报告作为训练数据，并基于条件随机场算法训练得到诊断属性识别模型；
    S303、将所述病理报告和所述影像报告输入所述诊断属性识别模型，输出所述病理诊断属性和所述影像诊断属性。
  4. 如权利要求1所述的放射影像报告的智能随访方法，其特征在于，步骤S40具体包括：
    S401、获取多组诊断结果一致性已知的历史病理报告和历史影像报告；
    S402、分别提取每组历史病理报告中的历史病理诊断属性和历史影像报告中的历史 影像诊断属性；
    S403、将每组历史病理诊断属性和历史影像诊断属性作为一个训练数据，并基于Bert预训练模型与Word2Vec算法训练得到诊断属性匹配模型；
    S404、分别将每个影像报告中的影像诊断属性与所述病理报告中的病理诊断属性输入所述诊断属性匹配模型，输出所述每个影像报告与所述病理报告是否匹配的结果。
  5. 如权利要求1所述的放射影像报告的智能随访方法，其特征在于，所述病理诊断属性包括病理定位、病理定性和病理定位与病理定性的病理关联性，所述影像诊断属性包括影像定位、影像定性和影像定位与影像定性的影像关联性。
  6. 一种电子设备，包括存储器、处理器及存储在存储器上并可在处理器上运行的计算机程序，其特征在于，所述处理器执行所述计算机程序时实现权利要求1至5任一项所述的放射影像报告的智能随访方法。
  7. 一种计算机可读存储介质，其上存储有计算机程序，其特征在于，所述程序被处理器执行时实现权利要求1至5任一项所述的放射影像报告的智能随访方法的步骤。
  8. 一种放射影像报告的智能随访系统，其特征在于，所述智能随访系统包括病理报告获取模块、影像报告获取模块、第一提取模块、第一匹配模块、标注模块和随访模块；
    所述病理报告获取模块用于获取一患者的一病理报告，所述病理报告包括病理报告时间和所述患者的身份信息；
    所述影像报告获取模块用于根据所述身份信息获取所述患者在所述病理报告时间前的一预设时间段内的所有影像报告；
    所述第一提取模块用于分别提取所述病理报告中的病理诊断属性及每个影像报告中的影像诊断属性；
    所述第一匹配模块用于将每组病理诊断属性分别与所述每个影像报告中的影像诊断属性进行匹配，若匹配不成功，则确定匹配不成功的影像报告中的影像诊断是有误的，若匹配成功，则确定匹配成功的影像报告中的影像诊断是正确的；
    所述标注模块用于将表征匹配结果的标签标注在影像报告上；
    所述随访模块用于根据所述标签对对应的患者进行随访。
  9. 如权利要求8所述的放射影像报告的智能随访系统，其特征在于，所述智能随访系统还包括预设模块、第二提取模块、第二匹配模块和滤除模块；
    所述预设模块用于预设一关联词典，所述关联词典存储有病理部位与放射检查项目的对应关系；
    所述第二提取模块用于提取所述病理报告中的当前病理部位及每个影像报告中的当 前放射检查项目；
    所述第二匹配模块用于基于所述关联词典将所述当前病理部位与所述当前放射检查项目进行匹配；
    所述滤除模块用于滤除所述当前放射检查项目与所述当前病理部位不匹配的影像报告；
    所述第一提取模块用于对滤除后的影像报告进行影像诊断属性的提取。
  10. 如权利要求8所述的放射影像报告的智能随访系统，其特征在于，所述第一提取模块包括第一报告获取单元和第一训练单元；
    所述第一报告获取单元用于获取多个历史报告，所述历史报告中的诊断属性已被标注；
    所述第一训练单元用于将所述历史报告作为训练数据，并基于条件随机场算法训练得到诊断属性识别模型；
    所述第一提取模块用于将所述病理报告和所述影像报告输入所述诊断属性识别模型，输出所述病理诊断属性和所述影像诊断属性。
  11. 如权利要求8所述的放射影像报告的智能随访系统，其特征在于，所述第一匹配模块包括第二报告获取单元、属性提取单元和第二训练单元；
    所述第二报告获取单元用于获取多组诊断结果一致性已知的历史病理报告和历史影像报告；
    所述属性提取单元用于分别提取每组历史病理报告中的历史病理诊断属性和历史影像报告中的历史影像诊断属性；
    所述第二训练单元用于将每组历史病理诊断属性和历史影像诊断属性作为一个训练数据，并基于Bert预训练模型与Word2Vec算法训练得到诊断属性匹配模型；
    所述第一匹配模块用于分别将每个影像报告中的影像诊断属性与所述病理报告中的病理诊断属性输入所述诊断属性匹配模型，输出所述每个影像报告与所述病理报告是否匹配的结果。
PCT/CN2020/070445 2019-07-22 2020-01-06 放射影像报告的智能随访方法、系统、设备及存储介质 WO2021012637A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910660410.1A CN110364236A (zh) 2019-07-22 2019-07-22 放射影像报告的智能随访方法、系统、设备及存储介质
CN201910660410.1 2019-07-22

Publications (1)

Publication Number Publication Date
WO2021012637A1 true WO2021012637A1 (zh) 2021-01-28

Family

ID=68221105

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/070445 WO2021012637A1 (zh) 2019-07-22 2020-01-06 放射影像报告的智能随访方法、系统、设备及存储介质

Country Status (2)

Country Link
CN (1) CN110364236A (zh)
WO (1) WO2021012637A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN118262220A (zh) * 2024-05-29 2024-06-28 青岛美迪康数字工程有限公司 放射影像报告的质量评估方法、装置和设备

Families Citing this family (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110364236A (zh) * 2019-07-22 2019-10-22 卫宁健康科技集团股份有限公司 放射影像报告的智能随访方法、系统、设备及存储介质
CN110991535A (zh) * 2019-12-04 2020-04-10 中山大学 一种基于多类型医学数据的pCR预测方法
CN112216381B (zh) * 2020-10-14 2022-01-25 南通市第一人民医院 一种影像科患者检查核对方法及系统
CN112562816A (zh) * 2020-11-13 2021-03-26 陈卫霞 肿瘤影像报告诊断结果与病理结果对应与评价系统及方法
CN112382360B (zh) * 2020-12-03 2024-05-31 卫宁健康科技集团股份有限公司 一种诊断报告自动生成系统、存储介质及电子设备
CN112712869B (zh) * 2021-01-15 2024-07-23 杨帆 从历史影像结构化报告中动态获取随访数据的系统及方法
CN113297836B (zh) * 2021-05-28 2024-08-27 善诊(上海)信息技术有限公司 影像报告标签评估方法、装置、计算机设备和存储介质
CN114464288B (zh) * 2022-04-13 2022-07-05 广州盛安医学检验有限公司 基于互联网的质谱标本数据管理系统
CN116052875A (zh) * 2022-06-23 2023-05-02 三峡大学 一种基于随访结果的诊断数据评价系统及方法
CN116343980B (zh) * 2023-05-30 2023-08-29 深圳市即达健康医疗科技有限公司 一种基于智慧医疗复诊随访数据处理方法及系统

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20080046286A1 (en) * 2005-09-16 2008-02-21 Halsted Mark J Computer implemented healthcare monitoring, notifying and/or scheduling system
CN106415555A (zh) * 2013-11-26 2017-02-15 皇家飞利浦有限公司 用于病理学报告与放射学报告的相关联的系统和方法
CN108319605A (zh) * 2017-01-16 2018-07-24 医渡云（北京）技术有限公司 医学检查数据的结构化处理方法及系统
CN109961828A (zh) * 2019-03-06 2019-07-02 卫宁健康科技集团股份有限公司 多模态医学影像及报告数据管理方法及系统
CN110364236A (zh) * 2019-07-22 2019-10-22 卫宁健康科技集团股份有限公司 放射影像报告的智能随访方法、系统、设备及存储介质

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE102007026520A1 (de) * 2007-06-08 2008-12-11 Siemens Ag Bildakquisitons-, Bildarchivierungs- und Bildrenderingsystem zur Reproduzierung von eingestellten Untersuchungsparametern einer CT-, PET-CT- bzw. MRT- gestützten Erstuntersuchung bei nachfolgenden radiologischen Kontrolluntersuchungen
CN105808712A (zh) * 2016-03-07 2016-07-27 陈宽 将文本类医疗报告转换为结构化数据的智能系统及方法
CN106570331A (zh) * 2016-11-07 2017-04-19 万里云医疗信息科技（北京）有限公司 一种动态诊断报告的生成方法及生成系统
CN107273657A (zh) * 2017-05-15 2017-10-20 慧影医疗科技（北京）有限公司 影像诊断图文报告的生成方法及存储设备
CN108198608A (zh) * 2018-01-31 2018-06-22 郑会敏 一种医学影像诊断系统

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20080046286A1 (en) * 2005-09-16 2008-02-21 Halsted Mark J Computer implemented healthcare monitoring, notifying and/or scheduling system
CN106415555A (zh) * 2013-11-26 2017-02-15 皇家飞利浦有限公司 用于病理学报告与放射学报告的相关联的系统和方法
CN108319605A (zh) * 2017-01-16 2018-07-24 医渡云（北京）技术有限公司 医学检查数据的结构化处理方法及系统
CN109961828A (zh) * 2019-03-06 2019-07-02 卫宁健康科技集团股份有限公司 多模态医学影像及报告数据管理方法及系统
CN110364236A (zh) * 2019-07-22 2019-10-22 卫宁健康科技集团股份有限公司 放射影像报告的智能随访方法、系统、设备及存储介质

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN118262220A (zh) * 2024-05-29 2024-06-28 青岛美迪康数字工程有限公司 放射影像报告的质量评估方法、装置和设备

Also Published As

Publication number Publication date
CN110364236A (zh) 2019-10-22

Similar Documents

Publication Publication Date Title
WO2021012637A1 (zh) 放射影像报告的智能随访方法、系统、设备及存储介质
US20220130554A1 (en) Method for an ai powered automated analyzer for displaying comprehensive medical reports, a supervised learning ml classifier for automated protocol change, a reinforcement learning ml classifier for predetermined protocol change, an iterative or recurrent neural network natural language processor and a successive or recursive neural network natural language processor for medical learning, understanding, prediction, and prescription
Annarumma et al. Automated triaging of adult chest radiographs with deep artificial neural networks
US20230215552A1 (en) Early detection of patients for coordinated application of healthcare resources based on bundled payment
CN109741806B (zh) 一种医学影像诊断报告辅助生成方法及其装置
CN107799160B (zh) 用药辅助决策方法及装置、存储介质、电子设备
US20220172809A9 (en) Report generating system and methods for use therewith
US9779505B2 (en) Medical data processing apparatus and method
CN111986182A (zh) 辅助诊断方法、系统、电子设备及存储介质
US9265472B2 (en) Methods, systems, services and circuits for generating rapid preliminary reports with bone mineral density measurements and/or osteoporosis risk assessments
CN103370629B (zh) 临床质量分析系统
CN112401915A (zh) 一种新冠肺炎ct复查的图像融合比对方法
WO2020110774A1 (ja) 画像処理装置、画像処理方法、及びプログラム
WO2020082804A1 (zh) 一种医疗数据分类存储方法和装置
CN117637092B (zh) 一种基于人工智能模型的病历预编码方法及装置
CN111279424B (zh) 用于优化图像采集工作流的设备、系统和方法
US20200043583A1 (en) System and method for workflow-sensitive structured finding object (sfo) recommendation for clinical care continuum
US20240006039A1 (en) Medical structured reporting workflow assisted by natural language processing techniques
CN112766314A (zh) 解剖结构的识别方法、电子设备及存储介质
US20200152303A1 (en) Methods for electronically processing insurance claims and devices thereof
WO2020082796A1 (zh) 基于数据分析的就诊信息处理方法、装置、设备及介质
CN115759040A (zh) 一种电子病历解析方法、装置、设备和存储介质
US20200027534A1 (en) Device, system, and method for updating problem lists
US20150339602A1 (en) System and method for modeling health care costs
CN115798667A (zh) 体检报告生成方法、装置、计算机设备和存储介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20844300

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20844300

Country of ref document: EP

Kind code of ref document: A1

122 Ep: pct application non-entry in european phase

Ref document number: 20844300

Country of ref document: EP

Kind code of ref document: A1