WO2020253886A1 - 一种病理辅助诊断方法 - Google Patents

一种病理辅助诊断方法 Download PDF

Info

Publication number
WO2020253886A1
WO2020253886A1 PCT/CN2020/100950 CN2020100950W WO2020253886A1 WO 2020253886 A1 WO2020253886 A1 WO 2020253886A1 CN 2020100950 W CN2020100950 W CN 2020100950W WO 2020253886 A1 WO2020253886 A1 WO 2020253886A1
Authority
WO
WIPO (PCT)
Prior art keywords
pathological
diagnosis
image
diagnosis result
identified
Prior art date
Application number
PCT/CN2020/100950
Other languages
English (en)
French (fr)
Inventor
王珣章
谢伟东
邓日强
杨艳
Original Assignee
广州智睿医疗科技有限公司
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 广州智睿医疗科技有限公司 filed Critical 广州智睿医疗科技有限公司
Priority to US17/619,616 priority Critical patent/US20220351370A1/en
Publication of WO2020253886A1 publication Critical patent/WO2020253886A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0012—Biomedical image inspection
    • G—PHYSICS
    • 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
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10056—Microscopic image
    • G06T2207/10061—Microscopic image from scanning electron microscope
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20081—Training; Learning
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20084—Artificial neural networks [ANN]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30004—Biomedical image processing
    • G06T2207/30096—Tumor; Lesion

Definitions

  • the present invention relates to the field of disease diagnosis, in particular to a pathological auxiliary diagnosis method.
  • Pathological diagnosis is the "gold standard" for judging cancer, that is, the pathological tissue sampled from the patient's lesion is made into a section (usually HE staining), and the microscopic observation is performed under a microscope to confirm the type of disease.
  • HE staining the pathological tissue sampled from the patient's lesion is made into a section
  • the microscopic observation is performed under a microscope to confirm the type of disease.
  • doctors need to observe pathological sections under a microscope, and then write pathological reports on the computer. This process should be repeated 200 to 300 times a day.
  • Cancer pathological diagnosis includes a wide variety of information, including both global organization structure information and local differences and changes. The model is complex and very professional, and it must be completed by a senior doctor in the pathology department.
  • the purpose of the present invention is to provide a pathological assistant diagnosis method, which can assist doctors in disease diagnosis and reduce the doctor's workload.
  • the present invention provides a pathological auxiliary diagnosis method, which includes:
  • a diagnosis report is generated based on the diagnosis result and patient information.
  • the process of acquiring the pathological image model includes:
  • the parameters of the neural network model are adjusted until the first diagnosis result is the same as the second diagnosis result.
  • the parameters of the neural network model include: convolution kernel bias weight, fully connected layer weight, and fully connected layer bias weight.
  • the second pathological image is used as the pathological image to be recognized.
  • obtaining the pathological image to be identified further includes:
  • the method further includes generating a two-dimensional identification code according to the patient information.
  • generating a diagnosis report based on the diagnosis result and patient information includes:
  • diagnosis result into the first blank position of the diagnosis report template, where the diagnosis result includes the pathological image to be recognized, the pathological name, and the degree of disease;
  • the patient information includes the patient's name, gender, and ID number;
  • the present invention also provides a pathological auxiliary diagnosis device, which includes:
  • the acquisition module is used to acquire pathological images to be identified and patient information
  • the recognition module is configured to input the pathological image to be recognized into a preset pathological image model, and the pathological image model outputs a diagnosis result of the pathological image to be recognized;
  • the report generation module is used to generate a diagnosis report based on the diagnosis result and patient information.
  • the present invention also provides a computer device, which includes:
  • Memory for storing computer readable instructions
  • the processor is configured to run the computer-readable instructions, so that the computer device executes the foregoing method.
  • the present invention also provides a computer-readable storage medium for storing computer-readable instructions.
  • the computer-readable instructions When executed by a computer, the computer is caused to execute the above method.
  • the present invention obtains the diagnosis result by inputting the pathological image to be recognized into the preset pathological image model. This process has high accuracy and fast speed, can greatly shorten the work flow of pathologists, save labor costs, and can also be based on the diagnosis results and patients The information generates a diagnostic report, and the pathologist only needs to review the report without writing it.
  • FIG. 1 is a flowchart of a pathological assistant diagnosis method provided by an embodiment of the present invention
  • Fig. 2 is a schematic diagram of a pathological auxiliary diagnosis device provided by an embodiment of the present invention.
  • Fig. 1 is a flowchart of a pathological assistant diagnosis method provided by an embodiment of the present invention, and the method includes:
  • pathological image is also called a pathological electronic slice image, which refers to an image scanned by an electron microscope for pathological analysis, such as a pathological image of the stomach, a pathological image of the liver, and so on.
  • Pathology images usually have the following characteristics: 1There are many private formats. For example, pathology images can be pictures in svs (a private format of pathology images), or pictures in ndpi (NanoZoomer Digital Pathology Image) format.
  • the pathological section After acquiring the pathological section of the patient, scan the pathological section with a relatively low first resolution, so as to obtain a complete, low-resolution first image of the pathological section in a relatively short time. For example, In routine diagnosis, it is necessary to scan the pathological section at a magnification of 40 times to 100 times.
  • the resolution of the obtained high-resolution scanned image is usually at the level of 10 5 ⁇ 10 5 , and the storage space occupied is also 100 megabytes.
  • the pathological slice may be selected to be scanned at a magnification of 15 times to obtain a scanned image with a lower resolution, for example, only a 10 4 ⁇ 10 4 level image is required. It can be understood that the above examples are only used as examples and should not be understood as specific limitations.
  • artificial intelligence algorithms such as machine learning or deep learning can be used to determine that at least one local area in the first image is a suspected lesion area, where the suspected lesion area may be an area that is indeed a lesion. It may be an area with a high probability of occurrence of a lesion, or an area with a high degree of attention of doctors, which is not specifically limited in the embodiment of the present invention.
  • the second resolution is higher than the first resolution.
  • the second resolution is then used to scan the corresponding position of the suspected lesion area on the pathological slice to obtain the suspected lesion tissue corresponding to the suspected lesion area.
  • the second image at the second resolution. It can be understood that in the second image, the non-suspected lesion area can be filled with a different color from the suspected lesion area, or it can be filled with the non-suspected lesion area in the first image, which is not specifically limited in the present invention.
  • the second pathological image is used as the pathological image to be recognized.
  • the method further includes:
  • the parameters of the neural network model are adjusted until the first diagnosis result is the same as the second diagnosis result.
  • the parameters of the neural network model include: convolution kernel bias weight, fully connected layer weight, and fully connected layer bias weight.
  • the neural network model may be a regional classification network model and/or a segmented convolutional neural network model, where the regional classification neural network model includes but is not limited to a residual network (Residual Networks, ResNet) model, VGG16 model, VGGNet model, Inception Models, etc.
  • segmented convolutional neural network models include but are not limited to Fully Convolutional Networks (FCN) models, multi-task network cascaded MNC models, Mask-RCNN models, etc.
  • diagnosis result into the first blank position of the diagnosis report template, where the diagnosis result includes the pathological image to be recognized, the pathological name, and the degree of disease;
  • the patient information includes the patient's name, gender, ID number, and may also include information such as age, home address, etc.;
  • the method further includes generating a two-dimensional identification code according to the patient information, and importing the two-dimensional identification code into a third blank position of the diagnosis report.
  • the diagnosis report can be stored, managed, transmitted, and reproduced by electronic equipment, and can be understood as a digital version of traditional paper medical records.
  • the diagnosis report may also include admission records, admission diagnosis, course records, examination reports, examination reports, discharge diagnosis and other documents.
  • admission records may include current medical history, past history, personal history, family history, auxiliary examinations and other parts.
  • the diagnosis report can also give corresponding rehabilitation suggestions based on the diagnosis results, such as colds: drink plenty of water, keep warm, and avoid cold wind; bronchitis: avoid spicy, quit smoking, drink more tea, and have a light diet; chronic pharyngitis: less Stay up late, drink plenty of water, avoid spicy; Chronic pneumonia: Do more chest exercises, supplement water and electrolytes, etc.
  • the diagnosis report can also be data signed and encrypted, and the identity of the user's diagnosis doctor can be authenticated. Only when the user's identity is a legitimate user, can the user digitally sign the diagnostic report. After the signature is completed, the access authority can also be set for the diagnosis report, and only those who have the access authority can decrypt the diagnosis includes and browse it.
  • the present invention obtains the diagnosis result by inputting the pathological image to be recognized into the preset pathological image model. This process has high accuracy and fast speed, can greatly shorten the work flow of pathologists, save labor costs, and can also be based on the diagnosis results and patients The information generates a diagnostic report, and the pathologist only needs to review the report without writing it.
  • Fig. 2 is a schematic diagram of a pathological assistant diagnosis device provided by an embodiment of the present invention, and the device includes:
  • the acquisition module is used to acquire pathological images to be identified and patient information
  • the recognition module is configured to input the pathological image to be recognized into a preset pathological image model, and the pathological image model outputs a diagnosis result of the pathological image to be recognized;
  • the report generation module is used to generate a diagnosis report based on the diagnosis result and patient information.
  • a computer device including a memory and a processor, and computer-readable instructions are stored in the memory.
  • the computer-readable instructions are executed by one or more processors, the one or more processors execute The computer-readable instructions implement the steps of the face verification method described in the foregoing embodiment.
  • the computer device may also include a user interface, a network interface, a camera, a radio frequency (RF) circuit, a sensor, an audio circuit, a WI-FI module, and so on.
  • the user interface may include a display screen (Display), an input unit such as a keyboard (Keyboard), etc., and the optional user interface may also include a USB interface, a card reader interface, and the like.
  • the network interface can optionally include a standard wired interface, a wireless interface (such as a Bluetooth interface, a WI-FI interface), etc.
  • the embodiment of the present invention also provides a computer storage medium on which a computer program is stored, and when the program is executed by a processor, the steps of the aforementioned face verification method are realized.
  • a "computer-readable medium” can be any device that can contain, store, communicate, propagate, or transmit a program for use by an instruction execution system, device, or device or in combination with these instruction execution systems, devices, or devices.
  • computer readable media include the following: electrical connections (electronic devices) with one or more wiring, portable computer disk cases (magnetic devices), random access memory (RAM), Read only memory (ROM), erasable and editable read only memory (EPROM or flash memory), fiber optic devices, and portable compact disk read only memory (CDROM).
  • the computer-readable medium may even be paper or other suitable media on which the program can be printed, because it can be used, for example, by optically scanning the paper or other media, and then editing, interpreting, or other suitable media if necessary. The program is processed in a manner to obtain the program electronically and then stored in the computer memory.
  • each part of the present invention can be implemented by hardware, software, firmware or a combination thereof.
  • multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system.
  • a logic gate circuit for implementing logic functions on data signals
  • PGA programmable gate array
  • FPGA field programmable gate array

Landscapes

  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Medical Informatics (AREA)
  • General Health & Medical Sciences (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Radiology & Medical Imaging (AREA)
  • Public Health (AREA)
  • Physics & Mathematics (AREA)
  • Quality & Reliability (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Primary Health Care (AREA)
  • Epidemiology (AREA)
  • Biomedical Technology (AREA)
  • Data Mining & Analysis (AREA)
  • Databases & Information Systems (AREA)
  • Pathology (AREA)
  • Image Processing (AREA)
  • Measuring And Recording Apparatus For Diagnosis (AREA)

Abstract

本发明公开了一种病理辅助诊断方法，所述方法包括：获取待识别病理图像和患者信息；将所述待识别病理图像输入至预设的病理图像模型，所述病理图像模型输出对所述待识别病理图像的诊断结果；根据所述诊断结果和患者信息生成诊断报告。本发明还提供了一种装置、计算机设备和存储介质。采用本发明，可以辅助医生进行疾病诊断，减少医生的工作量。

Description

一种病理辅助诊断方法 技术领域
本发明涉及疾病诊断领域，特别是涉及一种病理辅助诊断方法。
背景技术
病理诊断是判断癌症的“金标准”，即为将从患者病变部位取样的病理组织制成切片(通常HE染色)，在显微镜下进行微观层面观察，以确认疾病类型。当放射科已进入数字化时代，而病理科医生的工作状态数十年如一日，仍是切片+光学显微镜。在病理信息中心里，医生需要在显微镜下观察病理切片，然后在电脑上撰写病理报告。每天这个过程要重复200～300次。癌症病理诊断包含种类繁多，既包括全局组织结构信息，又包括局部差异变化，模式复杂，专业性非常强，须由病理科资深医生来完成。
现有技术无法满足大批量的阅片需求，而且手工操作给病理医生带来大量的重复劳动，对于海量数据信息的处理，并不具有优势，需要充足的人力物力成本。目前病理医生短缺严重，培养周期长是医院病理科普遍的困境。病理图像数字化、信息化程度低。诊断质量有待进一步提高。
因此，目前亟需一种病理辅助诊断方法，作为辅助工具帮助医生进行诊断，来提升医院的诊断质量，同时也减轻医生重复性操作量。
发明内容
为了解决上述问题，本发明的目的是提供一种病理辅助诊断方法，可以辅助医生进行疾病诊断，减少医生的工作量。
基于此，本发明提供了一种病理辅助诊断方法，所述方法包括：
获取待识别病理图像和患者信息；
将所述待识别病理图像输入至预设的病理图像模型，所述病理图像模型输出对所述待识别病理图像的诊断结果；
根据所述诊断结果和患者信息生成诊断报告。
其中，所述病理图像模型的获取过程包括：
获取若干个病理图片作为训练集，所述病理图片对应于第一诊断结果；
利用所述训练集对神经网络模型进行训练，获取第二诊断结果；
若所述第一诊断结果与所述第二诊断结果不相同，则对所述神经网络模型的参数进行调整，直至所述第一诊断结果与所述第二诊断结果相同。
其中，所述神经网络模型的参数包括：卷积核偏置权值、全连接层权值和全连接层偏置权值。
其中，获取所述待识别病理图像之前包括：
采用第一分辨率扫描病理切片，得到第一病理图像；
确定所述第一病理图像的至少一局部区域为疑似病灶区域；
采用第二分辨率扫描所述疑似病灶区域，得到第二病理图像，所述第二分辨率高于所述第一分辨率；
将所述第二病理图像作为待识别病理图像。
其中，获取所述待识别病理图像还包括：
对所述待识别病理图像进行灰度处理，得到待识别病理灰度图像；
将所述待识别病理灰度图像进行去除噪声滤波处理，得到无失真的待识别病理灰度图像；
将所述无失真的待识别病理灰度图像中识别出灰度均值小于预设灰度值的区域，将所述无失真的待识别病理灰度图像中灰度均值小于预设灰度值得区域赋值为黑色。
其中，所述方法还包括根据患者信息生成二维识别码。
其中，所述根据所述诊断结果和患者信息生成诊断报告包括：
预设诊断报告模板；
将所述诊断结果导入所述诊断报告模板的第一空白位置，所述诊断结果包括所述待识别病理图像、病理名称、患病程度；
将所述患者信息导入所述诊断报告模板的第二空白位置，所述患者信息包括患者姓名、性别、身份证号码；
本发明还提供了一种病理辅助诊断装置，所述装置包括：
获取模块，用于获取待识别病理图像和患者信息；
识别模块，用于将所述待识别病理图像输入至预设的病理图像模型，所述病理图像模型输出对所述待识别病理图像的诊断结果；
生成报告模块，用于根据所述诊断结果和患者信息生成诊断报告。
本发明还提供了一种计算机装置，所述计算机装置包括：
存储器，用于存储计算机可读指令；
处理器，用于运行所述计算机可读指令，使得所述计算机装置执行上述方法。
本发明还提供了一种计算机可读存储介质，用于存储计算机可读指令，当所述计算机可读指令由计算机执行时，使得所述计算机执行上述方法。
本发明通过将待识别病理图像输入预设病理图像模型，获取诊断结果，此过程准确率高、速度快，可以大大缩短病理科医生工作流程、节省人力成本，还可以根据所述诊断结果和患者信息生成诊断报告，病理科医生只需要审核报告即可，无需手写。
附图说明
为了更清楚地说明本发明实施例或现有技术中的技术方案，下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本发明的一些实施例，对于本领域 普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
图1是本发明实施例提供的病理辅助诊断方法的流程图；
图2是本发明实施例提供的病理辅助诊断装置的示意图。
具体实施方式
下面将结合本发明实施例中的附图，对本发明实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本发明一部分实施例，而不是全部的实施例。基于本发明中的实施例，本领域普通技术人员在没有作出创造性劳动前提下所获得的所有其他实施例，都属于本发明保护的范围。
图1是本发明实施例提供的病理辅助诊断方法的流程图，所述方法包括：
S101、获取待识别病理图像和患者信息；
所述病理图像又称为病理电子切片图，是指由电子显微镜扫描出来的用于病理分析的图像，例如：胃部的病理图像、肝部的病理图像等等。病理图像通常具备如下特点：①私有格式繁多，例如病理图像可以是svs(一种病理图像的私有格式)格式的图片，也可以是ndpi(NanoZoomer Digital Pathology Image，纳米变焦数字病理图像)格式的图片等等，这主要是由于各电子显微镜的厂家各自设置的不同导致的；②尺寸较大；一张病理图像的大小通常在500MB(MByte，兆比特，简称兆)-1.5GB(Gigabyte，十亿字节)左右；③分辨率较高，一张病理图像的分辨率通常在40000*40000像素以上。
获取所述待识别病理图像之前包括：
采用第一分辨率扫描病理切片，得到第一病理图像；
在获取病人的病理切片之后，采用相对较低的第一分辨率对病理切片进行扫描，以便在相对较短的时间内，获得该病理切片完整的较低分辨率的第一图像，举例来讲，常规诊断中，需要将病理切片在放 大40倍至100倍的情况下进行扫描，得到的高分辨率的扫描图像分辨率通常在10 5×10 5级别，占据的存储空间也在百兆字节以上，本发明实施例中，可以选择将病理切片在放大15倍的情况下进行扫描，以得到较低分辨率的扫描图像，比如只要求得到10 4×10 4级别的图像。可以理解，上述例子仅用作举例，不能理解为具体限定。
确定所述第一病理图像的至少一局部区域为疑似病灶区域；
在获得病理切片的第一图像之后，可以采用机器学习或者深度学习等人工智能算法，确定第一图像中至少一处局部区域为疑似病灶区域，其中，疑似病灶区域可以是确实为病灶的区域，可以是出现病灶概率较高的区域，还可以是医生关注度高的区域，本发明实施例不作具体限定。
采用第二分辨率扫描所述疑似病灶区域，得到第二病理图像，所述第二分辨率高于所述第一分辨率；
其中，所述第二分辨率高于所述第一分辨率。在采用人工智能算法确定第一分辨率的第一图像中的疑似病灶区域之后，再采用第二分辨率扫描上述疑似病灶区域在病理切片上的对应位置，得到疑似病灶区域对应的疑似病灶组织在第二分辨率下的第二图像。可以理解，在第二图像中，非疑似病灶区域可以用与疑似病灶区域不同的颜色填充，也可以用第一图像中的非疑似病灶区域填充，本发明不作具体限定。
将所述第二病理图像作为待识别病理图像。
获取所述待识别病理图像之后还包括：
对所述待识别病理图像进行灰度处理，得到待识别病理灰度图像；
将所述待识别病理灰度图像进行去除噪声滤波处理，得到无失真的待识别病理灰度图像；
将所述无失真的待识别病理灰度图像中识别出灰度均值小于预设灰度值的区域，将所述无失真的待识别病理灰度图像中灰度均值小于预设灰度值得区域赋值为黑色。
S102、将所述待识别病理图像输入至预设的病理图像模型，所述病理图像模型输出对所述待识别病理图像的诊断结果；
获取若干个病理图片作为训练集，所述病理图片对应于第一诊断结果；
利用所述训练集对神经网络模型进行训练，获取第二诊断结果；
若所述第一诊断结果与所述第二诊断结果不相同，则对所述神经网络模型的参数进行调整，直至所述第一诊断结果与所述第二诊断结果相同。
所述神经网络模型的参数包括：卷积核偏置权值、全连接层权值和全连接层偏置权值。
所述神经网络模型可以是区域分类网络模型和/或分割卷积神经网络模型，其中，区域分类神经网络模型包括但不限于残差网络(Residual Networks，ResNet)模型，VGG16模型，VGGNet模型、Inception模型等，分割卷积神经网络模型包括但不限于全卷积网络(Fully Convolutional Networks，FCN)模型、多任务网络级联MNC模型、Mask-RCNN模型等。
S103、根据所述诊断结果和患者信息生成诊断报告。
预设诊断报告模板；
将所述诊断结果导入所述诊断报告模板的第一空白位置，所述诊断结果包括所述待识别病理图像、病理名称、患病程度；
将所述患者信息导入所述诊断报告模板的第二空白位置，所述患者信息包括患者姓名、性别、身份证号码，还可以包括年龄、家庭住址等信息；
所述方法还包括根据患者信息生成二维识别码，将所述二维识别码导入所述诊断报告的第三空白位置。
诊断报告可以是以电子设备保存、管理、传输、重现病人的医疗记录，可以理解为传统纸质病历的数字化版本。
诊断报告中还可以包括入院记录、入院诊断、病程记录、检查报告、检验报告、出院诊断等文书，其中，入院记录可以包括现病史、既往史、个人史、家族史、辅助检查等部分。
诊断报告中还可以根据所述诊断结果给出相应的康复建议，如感冒：多喝水、注意保暖、忌吹冷风；支气管炎：忌辛辣、戒烟、多饮茶、饮食清淡；慢性咽炎：少熬夜、多喝水、忌辛辣；慢性肺炎：多做扩胸运动、补充水分和电解质等等。
对所述诊断报告还可以进行数据签名以及加密，并对用户的诊断医生的身份进行身份验证。只有在用户的身份为合法用户时，用户才能够对诊断报告进行数字签名。签名完成后，还可以对所述诊断报告设置访问权限，只有拥有访问权限的人员才能够解密诊断包括并进行浏览。
本发明通过将待识别病理图像输入预设病理图像模型，获取诊断结果，此过程准确率高、速度快，可以大大缩短病理科医生工作流程、节省人力成本，还可以根据所述诊断结果和患者信息生成诊断报告，病理科医生只需要审核报告即可，无需手写。
图2是本发明实施例提供的病理辅助诊断装置的示意图，所述装置包括：
获取模块，用于获取待识别病理图像和患者信息；
识别模块，用于将所述待识别病理图像输入至预设的病理图像模型，所述病理图像模型输出对所述待识别病理图像的诊断结果；
生成报告模块，用于根据所述诊断结果和患者信息生成诊断报告。
需要说明的是，本说明书中各个实施例采用递进的方式描述，每个实施例重点说明的都是与其他实施例的不同之处，各个实施例之间相同相似部分互相参见即可。对于实施例公开的装置而言，由于其与实施例公开的方法相对应，所以描述的比较简单，相关之处参见方法部分说明即可。
在一个实施例中，提出了一种计算机装置，包括存储器和处理器， 存储器中存储有计算机可读指令，计算机可读指令被一个或多个处理器执行时，使得一个或多个处理器执行计算机可读指令时实现上述实施例中所述人脸验证方法的步骤。
所述计算机装置还可以包括用户接口、网络接口、摄像头、射频(Radio Frequency，RF)电路，传感器、音频电路、WI-FI模块等等。用户接口可以包括显示屏(Display)、输入单元比如键盘(Keyboard)等，可选用户接口还可以包括USB接口、读卡器接口等。网络接口可选的可以包括标准的有线接口、无线接口(如蓝牙接口、WI-FI接口)等。
此外，本发明的实施例还提出一种计算机存储介质，其上存储有计算机程序，该程序被处理器执行时实现上述人脸验证方法的步骤。
在流程图中表示或在此以其他方式描述的逻辑和/或步骤，例如，可以被认为是用于实现逻辑功能的可执行指令的定序列表，可以具体实现在任何计算机可读介质中，以供指令执行系统、装置或设备(如基于计算机的系统、包括处理器的系统或其他可以从指令执行系统、装置或设备取指令并执行指令的系统)使用，或结合这些指令执行系统、装置或设备而使用。就本说明书而言，“计算机可读介质”可以是任何可以包含、存储、通信、传播或传输程序以供指令执行系统、装置或设备或结合这些指令执行系统、装置或设备而使用的装置。
计算机可读介质的更具体的示例(非穷尽性列表)包括以下：具有一个或多个布线的电连接部(电子装置)，便携式计算机盘盒(磁装置)，随机存取存储器(RAM)，只读存储器(ROM)，可擦除可编辑只读存储器(EPROM或闪速存储器)，光纤装置，以及便携式光盘只读存储器(CDROM)。另外，计算机可读介质甚至可以是可在其上打印所述程序的纸或其他合适的介质，因为可以例如通过对纸或其他介质进行光学扫描，接着进行编辑、解译或必要时以其他合适方式进行处理来以电子方式获得所述程序，然后将其存储在计算机存储器中。
应当理解，本发明的各部分可以用硬件、软件、固件或它们的组合来实现。在上述实施方式中，多个步骤或方法可以用存储在存储器 中且由合适的指令执行系统执行的软件或固件来实现。例如，如果用硬件来实现，和在另一实施方式中一样，可用本领域公知的下列技术中的任一项或他们的组合来实现：具有用于对数据信号实现逻辑功能的逻辑门电路的离散逻辑电路，具有合适的组合逻辑门电路的专用集成电路，可编程门阵列(PGA)，现场可编程门阵列(FPGA)等。
在本说明书的描述中，参考术语“一个实施例”、“一些实施例”、“示例”、“具体示例”、或“一些示例”等的描述意指结合该实施例或示例描述的具体特征、结构、材料或者特点包含于本发明的至少一个实施例或示例中。在本说明书中，对上述术语的示意性表述不一定指的是相同的实施例或示例。而且，描述的具体特征、结构、材料或者特点可以在任何的一个或多个实施例或示例中以合适的方式结合。
尽管已经示出和描述了本发明的实施例，本领域的普通技术人员可以理解：在不脱离本发明的原理和宗旨的情况下可以对这些实施例进行多种变化、修改、替换和变型，本发明的范围由权利要求及其等同物限定。

Claims (10)

  1. 一种病理辅助诊断方法，其特征在于，包括：
    获取待识别病理图像和患者信息；
    将所述待识别病理图像输入至预设的病理图像模型，所述病理图像模型输出对所述待识别病理图像的诊断结果；
    根据所述诊断结果和患者信息生成诊断报告。
  2. 如权利要求1所述的病理辅助诊断方法，其特征在于，所述病理图像模型的获取过程包括：
    获取若干个病理图片作为训练集，所述病理图片对应于第一诊断结果；
    利用所述训练集对神经网络模型进行训练，获取第二诊断结果；
    若所述第一诊断结果与所述第二诊断结果不相同，则对所述神经网络模型的参数进行调整，直至所述第一诊断结果与所述第二诊断结果相同。
  3. 如权利要求2所述的病理辅助诊断方法，其特征在于，所述神经网络模型的参数包括：卷积核偏置权值、全连接层权值和全连接层偏置权值。
  4. 如权利要求1所述的病理辅助诊断方法，其特征在于，获取所述待识别病理图像之前包括：
    采用第一分辨率扫描病理切片，得到第一病理图像；
    确定所述第一病理图像的至少一局部区域为疑似病灶区域；
    采用第二分辨率扫描所述疑似病灶区域，得到第二病理图像，所述第二分辨率高于所述第一分辨率；
    将所述第二病理图像作为待识别病理图像。
  5. 如权利要求1所述的病理辅助诊断方法，其特征在于，获取所述待识别病理图像还包括：
    对所述待识别病理图像进行灰度处理，得到待识别病理灰度图像；
    将所述待识别病理灰度图像进行去除噪声滤波处理，得到无失真的待识别病理灰度图像；
    将所述无失真的待识别病理灰度图像中识别出灰度均值小于预设灰度值的区域，将所述无失真的待识别病理灰度图像中灰度均值小于预设灰度值得区域赋值为黑色。
  6. 如权利要求1所述的病理辅助诊断方法，其特征在于，所述方法还包括根据患者信息生成二维识别码。
  7. 如权利要求1所述的病理辅助诊断方法，其特征在于，所述根据所述诊断结果和患者信息生成诊断报告包括：
    预设诊断报告模板；
    将所述诊断结果导入所述诊断报告模板的第一空白位置，所述诊断结果包括所述待识别病理图像、病理名称、患病程度；
    将所述患者信息导入所述诊断报告模板的第二空白位置，所述患者信息包括患者姓名、性别、身份证号码。
  8. 一种病理辅助诊断装置，其特征在于，包括：
    获取模块，用于获取待识别病理图像和患者信息；
    识别模块，用于将所述待识别病理图像输入至预设的病理图像模型，所述病理图像模型输出对所述待识别病理图像的诊断结果；
    生成报告模块，用于根据所述诊断结果和患者信息生成诊断报告。
  9. 一种计算机装置，其特征在于，包括：
    存储器，用于存储计算机可读指令；
    处理器，用于运行所述计算机可读指令，使得所述计算机装置执行如权利要求1到7的任一项所述方法。
  10. 一种计算机可读存储介质，用于存储计算机可读指令，当所述计算机可读指令由计算机执行时，使得所述计算机执行如权利要求1到7的任一项所述方法。
PCT/CN2020/100950 2019-06-18 2020-07-08 一种病理辅助诊断方法 WO2020253886A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US17/619,616 US20220351370A1 (en) 2019-06-18 2020-07-08 Auxiliary pathological diagnosis method

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910529275.7A CN110335256A (zh) 2019-06-18 2019-06-18 一种病理辅助诊断方法
CN201910529275.7 2019-06-18

Publications (1)

Publication Number Publication Date
WO2020253886A1 true WO2020253886A1 (zh) 2020-12-24

Family

ID=68142532

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/100950 WO2020253886A1 (zh) 2019-06-18 2020-07-08 一种病理辅助诊断方法

Country Status (3)

Country Link
US (1) US20220351370A1 (zh)
CN (1) CN110335256A (zh)
WO (1) WO2020253886A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112869744A (zh) * 2021-01-20 2021-06-01 上海市精神卫生中心（上海市心理咨询培训中心） 一种精神分裂症辅助诊断方法、系统和存储介质

Families Citing this family (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110335256A (zh) * 2019-06-18 2019-10-15 广州智睿医疗科技有限公司 一种病理辅助诊断方法
CN110766004B (zh) * 2019-10-23 2022-05-13 泰康保险集团股份有限公司 医疗鉴定数据处理方法、装置、电子设备及可读介质
CN110838103B (zh) * 2019-10-29 2023-05-16 重庆金山医疗技术研究院有限公司 一种图像处理方法、装置、诊断设备及计算机存储介质
CN110910361A (zh) * 2019-11-15 2020-03-24 首都医科大学附属北京安贞医院 一种基于血管支架的检出方法及装置
CN111899214B (zh) * 2020-06-12 2022-12-09 西安交通大学 一种病理切片扫描分析装置及病理切片扫描方法
CN111681749A (zh) * 2020-06-22 2020-09-18 韦志永 一种病理科规范化工作管理和诊断咨询系统及方法
CN112435745B (zh) * 2020-12-18 2024-04-05 深圳赛安特技术服务有限公司 就诊策略推荐方法、装置、电子设备及存储介质
CN113870968A (zh) * 2021-09-13 2021-12-31 上海联影医疗科技股份有限公司 医疗影像报告的生成方法、装置、计算机设备和存储介质
CN113808738B (zh) * 2021-09-18 2022-03-15 安徽爱朋科技有限公司 一种基于自识别影像的疾病识别系统
CN115954100B (zh) * 2022-12-15 2023-11-03 东北林业大学 胃癌病理图像智能辅助诊断系统
CN115631386B (zh) * 2022-12-19 2023-03-10 天津医之本医疗科技有限公司 一种基于机器学习的病理图像分类方法和系统

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20070093967A1 (en) * 2005-09-08 2007-04-26 Searete Llc, A Limited Liability Corporation Of The State Of Delaware Accessing data related to tissue coding
CN106372390A (zh) * 2016-08-25 2017-02-01 姹ゅ钩 一种基于深度卷积神经网络的预防肺癌自助健康云服务系统
CN106569673A (zh) * 2016-11-11 2017-04-19 北京昆仑医云科技有限公司 多媒体病历报告的显示方法及多媒体病历报告的显示设备
CN108510495A (zh) * 2018-04-09 2018-09-07 沈阳东软医疗系统有限公司 一种基于人工智能的肺部影像数据处理方法、装置及系统
CN109411086A (zh) * 2018-12-26 2019-03-01 深圳硅基智能科技有限公司 眼底图像病变的筛查系统
CN109685798A (zh) * 2018-12-26 2019-04-26 北京邮电大学 一种确定有效医学图像的方法及装置
CN110335256A (zh) * 2019-06-18 2019-10-15 广州智睿医疗科技有限公司 一种病理辅助诊断方法

Family Cites Families (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP6235886B2 (ja) * 2013-01-08 2017-11-22 キヤノン株式会社 生体組織画像の再構成方法及び装置並びに該生体組織画像を用いた画像表示装置
WO2015054666A1 (en) * 2013-10-10 2015-04-16 Board Of Regents, The University Of Texas System Systems and methods for quantitative analysis of histopathology images using multi-classifier ensemble schemes
WO2016120441A2 (en) * 2015-01-30 2016-08-04 Ventana Medical Systems, Inc. Quality metrics for automatic evaluation of dual ish images
EP3345122B1 (en) * 2015-09-02 2021-06-30 Ventana Medical Systems, Inc. Automated analysis of cellular samples having intermixing of analytically distinct patterns of analyte staining
US10846854B2 (en) * 2017-10-13 2020-11-24 Shenzhen Keya Medical Technology Corporation Systems and methods for detecting cancer metastasis using a neural network
EP3716879A4 (en) * 2017-12-28 2022-01-26 Changi General Hospital Pte Ltd MOTION COMPENSATION PLATFORM FOR IMAGE-GUIDED PERCUTANEOUS ACCESS TO BODY ORGANS AND STRUCTURES
CN108564578A (zh) * 2018-04-13 2018-09-21 上海商汤智能科技有限公司 病理诊断辅助方法、装置及系统
US10943346B2 (en) * 2018-05-21 2021-03-09 Corista, LLC Multi-sample whole slide image processing in digital pathology via multi-resolution registration and machine learning
CA3138679A1 (en) * 2019-04-30 2020-11-05 The Trustees Of Dartmouth College System and method for attention-based classification of high-resolution microscopy images

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20070093967A1 (en) * 2005-09-08 2007-04-26 Searete Llc, A Limited Liability Corporation Of The State Of Delaware Accessing data related to tissue coding
CN106372390A (zh) * 2016-08-25 2017-02-01 姹ゅ钩 一种基于深度卷积神经网络的预防肺癌自助健康云服务系统
CN106569673A (zh) * 2016-11-11 2017-04-19 北京昆仑医云科技有限公司 多媒体病历报告的显示方法及多媒体病历报告的显示设备
CN108510495A (zh) * 2018-04-09 2018-09-07 沈阳东软医疗系统有限公司 一种基于人工智能的肺部影像数据处理方法、装置及系统
CN109411086A (zh) * 2018-12-26 2019-03-01 深圳硅基智能科技有限公司 眼底图像病变的筛查系统
CN109685798A (zh) * 2018-12-26 2019-04-26 北京邮电大学 一种确定有效医学图像的方法及装置
CN110335256A (zh) * 2019-06-18 2019-10-15 广州智睿医疗科技有限公司 一种病理辅助诊断方法

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112869744A (zh) * 2021-01-20 2021-06-01 上海市精神卫生中心（上海市心理咨询培训中心） 一种精神分裂症辅助诊断方法、系统和存储介质
CN112869744B (zh) * 2021-01-20 2024-05-31 上海市精神卫生中心（上海市心理咨询培训中心） 一种精神分裂症辅助诊断方法、系统和存储介质

Also Published As

Publication number Publication date
US20220351370A1 (en) 2022-11-03
CN110335256A (zh) 2019-10-15

Similar Documents

Publication Publication Date Title
WO2020253886A1 (zh) 一种病理辅助诊断方法
Richardson et al. Noninterpretive uses of artificial intelligence in radiology
CN109741346B (zh) 感兴趣区域提取方法、装置、设备及存储介质
White et al. MeshMonk: Open-source large-scale intensive 3D phenotyping
Lewis et al. Artificial Intelligence in medical imaging practice: looking to the future
KR101857624B1 (ko) 임상 정보를 반영한 의료 진단 방법 및 이를 이용하는 장치
CN104093354B (zh) 用于评估医学图像的方法和设备
WO2021189901A1 (zh) 图像分割方法、装置、电子设备及计算机可读存储介质
CN111325714B (zh) 感兴趣区域的处理方法、计算机设备和可读存储介质
US20150173705A1 (en) Apparatus and method for adapting diagnostic model for computer-aided diagnosis
Langner et al. Kidney segmentation in neck-to-knee body MRI of 40,000 UK Biobank participants
CN110400617A (zh) 医学成像中的成像和报告的结合
CN111080583B (zh) 医学图像检测方法、计算机设备和可读存储介质
CN115830017B (zh) 基于图文多模态融合的肿瘤检测系统、方法、设备及介质
CN116386902B (zh) 基于深度学习的大肠癌人工智能辅助病理诊断系统
Ito et al. Automated segmentation of articular disc of the temporomandibular joint on magnetic resonance images using deep learning
CN111223158B (zh) 心脏冠脉图像的伪影校正方法和可读存储介质
Zhang et al. C2‐GAN: Content‐consistent generative adversarial networks for unsupervised domain adaptation in medical image segmentation
Sendra-Balcells et al. Generalisability of fetal ultrasound deep learning models to low-resource imaging settings in five African countries
Gheorghiță et al. Improving robustness of automatic cardiac function quantification from cine magnetic resonance imaging using synthetic image data
CN116433695B (zh) 一种乳腺钼靶图像的乳腺区域提取方法及系统
CN113610923A (zh) 扫描定位方法、装置、计算机设备及计算机可读存储介质
JP2022179433A (ja) 画像処理装置及び画像処理方法
JP7366111B2 (ja) 病理診断支援方法
CN114037830A (zh) 增强图像生成模型的训练方法、图像处理方法及装置

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: 20826337

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: 20826337

Country of ref document: EP

Kind code of ref document: A1