CN114496177B - Method and system for detecting clinical infection source of infectious department based on big data - Google Patents
Method and system for detecting clinical infection source of infectious department based on big data Download PDFInfo
- Publication number
- CN114496177B CN114496177B CN202210081573.6A CN202210081573A CN114496177B CN 114496177 B CN114496177 B CN 114496177B CN 202210081573 A CN202210081573 A CN 202210081573A CN 114496177 B CN114496177 B CN 114496177B
- Authority
- CN
- China
- Prior art keywords
- infection source
- graph
- infection
- image
- online
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- 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
- G16H30/40—ICT specially adapted for the handling or processing of medical images for processing medical images, e.g. editing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- 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
- 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
- G16H50/70—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for mining of medical data, e.g. analysing previous cases of other patients
-
- 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
- G16H50/80—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for detecting, monitoring or modelling epidemics or pandemics, e.g. flu
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Public Health (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Theoretical Computer Science (AREA)
- Primary Health Care (AREA)
- Epidemiology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Radiology & Medical Imaging (AREA)
- Pathology (AREA)
- Biomedical Technology (AREA)
- Databases & Information Systems (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a method and a system for detecting clinical infection sources of infectious departments based on big data, wherein the method comprises the steps of mixing an infection source specific antibody test object with an infection source, and placing the mixed infection source on a platform to be detected; observing and shooting a detection platform by a detection camera to obtain an initial detection image of the infection source; the application of the online big data upper computer firstly divides the image into a plurality of units with equal area, and each unit is defined as a graph cell; then classifying the graph cells into a detected target class and a non-detected target class by adopting a per unit value method according to different numerical values of each graph cell; and (3) carrying out image segmentation on the graph cells with the detection target class by using a seed filling Fill method to obtain a segmentation graph, then comparing the segmentation graph with a threshold value of the characteristic, expanding and adjusting the image to determine the edge of the infection source marker, and determining the infection source marker so as to determine the number of the infection sources.
Description
Technical Field
The invention relates to a method and a system for detecting clinical infection sources of an infectious department based on big data.
Background
There is no efficient technology for detecting and judging infection sources in the related art, for example, the related art discloses a method for detecting infection sources, which adopts a method for detecting infection sources by obtaining first infection source label information; determining first epidemic situation grade information corresponding to the first epidemic situation area according to the real-time epidemic situation information; obtaining second infection source label information according to the first epidemic situation grade information and the first infection source label information; obtaining clinical symptom information of the first patient, and grading the clinical symptom information of the first patient to obtain a first severity grade; obtaining a predetermined severity level threshold; determining whether the first severity level is within the predetermined severity level threshold; if the first severity level is within the predetermined severity level threshold, obtaining a solution for a first diagnostic protocol based on the second infection source signature information and the first severity level. The infection source information is obtained by gradually analyzing and judging the epidemic situation information, although the technology discloses that the infection source information is gradually analyzed and judged through the epidemic situation information, the technology does not disclose more detailed specific technical content and actually the technology also determines the infection source by simple biochemical detection in the traditional technology, so that the real significance, accuracy and high efficiency cannot be realized.
Disclosure of Invention
In order to overcome the defects of the prior art, the invention provides a method and a system for detecting clinical infection sources of an infectious department based on big data.
The technical scheme adopted by the invention for solving the technical problems is as follows:
the method for detecting the clinical infectious source of the infectious department based on the big data comprises the following steps of extracting the infectious source, mixing an infectious source specific antibody test object with the infectious source, placing the mixed infectious source on a platform to be detected, and enabling the infectious source specific antibody test object and the infectious source to be mixed to generate an infectious source marker;
secondly, observing and shooting a detection platform by a detection camera to obtain an initial detection image of the infection source, wherein the initial detection image has the characteristics of the infection source marker;
third, uploading the initial detection image to an online big data upper computer through a local host;
the fourth step, the application of the online big data upper computer divides the image into a plurality of units with equal area, each unit is defined as a graph cell, and then the graph cells are classified into a detected target class and a non-detected target class by adopting a per unit value method according to the difference of the numerical value of each graph cell; the detected target class is specifically an infection source marker in a graph unit cell, and the non-detected target class is characterized in that no infection source marker is in the graph unit cell;
fifthly, carrying out image segmentation on the graph cells with the detection target class by using a seed filling seed Fill method to obtain a segmentation graph, then expanding and adjusting the image after comparing the segmentation graph with a characteristic threshold value to determine the edge of the infection source marker and determine the infection source marker so as to determine the number of the infection sources;
the sixth step, online big data feeds back "determined infectious agent marker and determined infectious agent number" to the local host.
Further, the infection source specific antibody test object adopts colloidal gold, the detection camera adopts a dark field microscopic observation camera, and the detection platform adopts a microscopic observation objective table.
Further, the "determination of the number of infectious agents" is specifically the determination of the number of infectious agents by scan counting after the "determination of the markers of infectious agents". The detection system of clinical infection sources of the infectious department based on big data comprises a platform to be detected, a detection camera, a local host and an online big data upper computer; the online big data upper computer is provided with an infection source image dividing module, an infection source image classifying module and an infection source identification module which are connected in an application configuration mode, wherein the infection source image dividing module is used for dividing an image into a plurality of units with equal areas, each unit is defined as a graph unit cell, the infection source image classifying module is used for classifying the graph units into a detected target class and a non-detected target class according to the difference of the numerical values of each graph unit cell by adopting a per unit value method, the infection source identification module is used for carrying out image segmentation on the graph unit cells with the detected target class by using a seed filling seed Fill method to obtain a segmentation image, then the segmentation image is expanded and adjusted after the segmentation image is compared with a threshold value of characteristics to determine the edge of an infection source marker, and the infection source marker is determined to further determine the number of the infection source.
Further, the online big data upper computer is applied to configure an infected source image online shared database and an infected source image online management module, the infected source image online management module is used for sending requests for obtaining infected source images to a plurality of distributed storage positions, the plurality of distributed storage positions all store the infected source image data under the line, the plurality of distributed storage positions also store indexes and copies of the infected source image data, and the mapping reduction is run in parallel to support the parallel sending of the data to the infected source image online shared database; the infection source image on-line management module is further used for awakening the shared database on the infection source image line, so that the shared database on the infection source image line can acquire the infection source image data from a plurality of distributed storage positions.
And further, an online big data upper computer is also applied and provided with an online read-only learning training module, the online shared database of the infection source image line is also used for supporting the online read-only learning training module to acquire the infection source image data and perform learning training, and the online read-only learning training module is used for acquiring the infection source image data from the online shared database of the infection source image line and performing learning training to optimize the threshold value of the characteristics of the infection source marker.
Further, the threshold value of each infection source marker feature is determined by cross multiplication of two arrays, wherein one array is a feature basis value set (m1, m 2.... mi), and mi is an ordering element of the feature basis value set; another array is a variable characteristic coefficient group (ni, n 2.... no), ni is an ordering coefficient item of the variable characteristic coefficient group;
for any one of the graphs or graph cells with the detection target class or the segmentation graphs of the graph cells, sorting the values of the pixel points in the graph into a group to be identified (c1, c 2.... ci), wherein ci is the ith element of the group to be identified, and if the number of the pixel points is greater than i, adopting the characteristic values of the pixel points as the elements of the group to be identified;
when judging whether a graph or a graph cell of a detection target class or a segmentation graph of the graph cell has a target, calculating for the first time:
P1＝(m1＊n1-c1) 2 +(m2＊n2-c2) 2 +......+(mi＊ni-ci) 2 ；
then, rotating the 'graph or graph unit lattice with the detection target class or the segmentation graph of the graph unit lattice' forward by 360/j degrees, correspondingly changing the ordering of the elements ci of the group to be identified, and continuing the calculation to obtain p2, and iterating j-1 times until pj is obtained;
judging whether max (p1, p 2......... pj) is larger than a threshold value, and judging that the graph or the graph cell of the detected target class or the segmentation graph of the graph cell has a target if max (p1, p 2...... pj) is larger than the threshold value. The further learning training to optimize the threshold of the "infection source marker" feature is to change the value of the variable feature coefficient in the variable feature coefficient set (ni, n 2...... ni) by the data learning training with the decision result label.
Advantageous effects
Firstly, mixing an infection source specific antibody test object with an infection source, placing the mixed infection source on a platform to be detected, and generating an infection source marker after the infection source specific antibody test object and the infection source are mixed; then, observing and shooting a detection platform by a detection camera to obtain an initial detection image of the infection source, wherein the initial detection image has the characteristics of the infection source marker; then, uploading the initial detection image to an online big data upper computer through a local host; then, the application of the online big data upper computer divides the image into a plurality of units with equal area, and each unit is defined as a graph cell; then classifying the graph cells into a detected target class and a non-detected target class by adopting a per unit value method according to different numerical values of each graph cell; the detected target class is specifically an infection source marker in the graph unit cell, and the non-detected target class is characterized in that no infection source marker is in the graph unit cell; then, carrying out image segmentation on the graph cells with the detection target class by using a seed filling seed Fill method to obtain a segmentation graph, then comparing the segmentation graph with a threshold value of the characteristic, expanding and adjusting the image to determine the edge of an infection source marker, and determining the infection source marker to further determine the number of the infection sources; the online big data feeds back "the determined infective agent marker and the determined infective agent number" to the local host. According to the method, the infection source markers and the number of determined infection sources can be determined by applying a big data technology and an image analysis technology, and the detection process is efficient and convenient. The method is more accurate and comprehensive in distinguishing the infection source marker target.
Drawings
Fig. 1 is a flow chart of a method of an embodiment of the present application.
Detailed Description
In the specific implementation of the application, the application discloses a big data-based detection system for clinical infection sources of infectious departments, which comprises a platform to be detected, a detection camera, a local host and an online big data upper computer; the online big data upper computer is provided with an infection source image dividing module, an infection source image classifying module and an infection source identifying module which are connected in an application configuration mode, wherein the infection source image dividing module is used for dividing an image into a plurality of units with equal areas, each unit is defined as a graph unit cell, the infection source image classifying module is used for classifying the graph units into a detected target class and a non-detected target class by adopting a per unit value method according to different numerical values of each graph unit cell, the infection source identifying module is used for carrying out image segmentation on the graph unit cells with the detected target class by using a seed filling seed Fill method to obtain a segmentation image, then expanding and adjusting the image after comparing the segmentation image with a threshold value of a characteristic to determine the edge of an infection source marker, and determining the infection source marker to further determine the number of the infection source. The local host in the implementation can adopt a PC, and the step of "determining the number of the infection sources" is to determine the number of the infection sources by adopting scanning counting after "determining the markers of the infection sources".
In the specific implementation, one or more pixels can be used for each image unit cell, and in the specific implementation of the present application, as shown in fig. 1, firstly, the infection source specific antibody test substance is mixed with the infection source, and the mixed infection source is placed on the platform to be detected, and the infection source marker appears after the infection source specific antibody test substance is mixed with the infection source; then, observing and shooting a detection platform by a detection camera to obtain an initial detection image of the infection source, wherein the initial detection image has the characteristics of the infection source marker; then, uploading the initial detection image to an online big data upper computer through a local host; then, the application of the online big data upper computer divides the image into a plurality of units with equal area, and each unit is defined as a graph cell; then classifying the graph cells into a detected target class and a non-detected target class by adopting a per unit value method according to different numerical values of each graph cell; the detected target class is specifically an infection source marker in the graph unit cell, and the non-detected target class is characterized in that no infection source marker is in the graph unit cell; then, carrying out image segmentation on the graph cells with the detection target class by using a seed filling seed Fill method to obtain a segmentation graph, then comparing the segmentation graph with a threshold value of the characteristic, expanding and adjusting the image to determine the edge of the infection source marker, and determining the infection source marker to further determine the number of the infection sources; the online big data feeds back "the determined infective agent marker and the determined infective agent number" to the local host. According to the method, the infection source markers and the number of determined infection sources can be determined by applying a big data technology and an image analysis technology, and the detection process is efficient and convenient.
The application configuration of the online big data upper computer is used for configuring an infection source image online shared database and an infection source image online management module, the infection source image online management module is used for sending a request for acquiring an infection source image to a plurality of distributed storage positions, the plurality of distributed storage positions store the infection source image data under the line, the plurality of distributed storage positions also store indexes and copies of the infection source image data, and the mapping reduction is run in parallel to support the parallel sending of data to the infection source image online shared database; the on-line management module of the infected source image is also used for awakening the shared database on the infected source image line so that the shared database on the infected source image line can acquire the data of the infected source image from a plurality of distributed storage positions. The application of the online big data upper computer is also provided with an online read-only learning training module, the online shared database of the infection source image is also used for supporting the online read-only learning training module to acquire the infection source image data and carry out learning training, and the online read-only learning training module is used for acquiring the infection source image data from the online shared database of the infection source image and carrying out learning training to optimize the threshold value of the characteristics of the infection source marker.
Preferably, the threshold value of each infection source marker feature is determined by cross-multiplying two arrays, one of which is a feature basis set (m1, m2,..... mi), and mi is an ordering element of the feature basis set; another array is a variable characteristic coefficient group (ni, n 2.... no), ni is an ordering coefficient item of the variable characteristic coefficient group;
for any one of the graphs or graph cells with the detection target class or the segmentation graphs of the graph cells, sorting the values of the pixel points in the graph into a group to be identified (c1, c 2.. eta.. ci), wherein ci is the ith element of the group to be identified, and if the number of the pixel points is greater than i, adopting the characteristic values of the pixel points as the elements of the group to be identified;
when judging whether a graph or a graph cell of a detection target class or a segmentation graph of the graph cell has a target, calculating for the first time:
P1＝(m1＊n1-c1) 2 +(m2＊n2-c2) 2 +......+(mi＊ni-ci) 2 ；
then, rotating the 'graph or graph cell with the detection target class or the segmentation graph of the graph cell' forward by 360/j degrees, correspondingly changing the ordering of the elements ci of the group to be identified, continuing the calculation to obtain p2, and iterating j-1 times in this way until pj is obtained;
judging whether max (p1, p 2......... pj) is larger than a threshold value, and judging that the graph or the graph cell of the detected target class or the segmentation graph of the graph cell has a target if max (p1, p 2...... pj) is larger than the threshold value.
The learning training to optimize the threshold of the "infection source marker" feature is to change the value of the variable feature coefficient in the variable feature coefficient set (ni, n2,...... ni) by the data learning training with the judgment result label.
In the above implementation, P1 calculated for the first time is an initial discrimination parameter, in the implementation of the present application, through each iterative calculation, the graph is rotated forward by 360/j degrees, "the ordering of the elements ci of the group to be recognized is correspondingly changed" which means that after each iterative rotation "the graph or graph cell with the detection target class or the segmentation graph of the graph cell" is rotated, the positions of the pixels or pixel features in the graph are changed, the ordering of the elements in the group to be recognized (c1, c2,...... ci) corresponding to the pixels or the features of the pixels in the representation graph is also changed, and a new group to be recognized (c1, c2,..... ci) is obtained to participate in the calculation again, so that in the implementation of the present application, the discrimination can be performed sequentially for each angle of the graph or graph cell with the detection target class or the segmentation graph cell, the accuracy of the judgment result is prevented from being influenced by the distribution position of the target. In addition, whether max (p1, p 2...... cndot.) is larger than the threshold or not is judged, and whether max (p1, p 2.. cndot.) is larger than the threshold or not indicates that the judgment parameter is satisfactory under at least one angle, so that the existence of the target in the map can be judged, wherein the target specifically refers to the infection source marker or the characteristics of the infection source marker in specific implementation. The method and the device are more accurate and comprehensive when the target is judged in the implementation.
In the concrete implementation, the application also discloses a method for detecting clinical infection sources of an infectious department based on big data, which comprises the following steps:
the method comprises the following steps of firstly, extracting an infection source, mixing an infection source specific antibody test object with the infection source, placing the mixed infection source on a platform to be detected, and generating an infection source marker after the infection source specific antibody test object and the infection source are mixed;
secondly, observing and shooting a detection platform by a detection camera to obtain an initial detection image of the infection source, wherein the initial detection image has the characteristics of the infection source marker;
third, uploading the initial detection image to an online big data upper computer through a local host;
the fourth step, the application of the online big data upper computer divides the image into a plurality of units with equal area, each unit is defined as a graph cell, and then the graph cells are classified into a detected target class and a non-detected target class by adopting a per unit value method according to the difference of the numerical value of each graph cell; the detected target class is specifically an infection source marker in the graph unit cell, and the non-detected target class is characterized in that no infection source marker is in the graph unit cell;
fifthly, using a seed filled seed Fill method to carry out image segmentation on the graph cells with the detection target class to obtain a segmentation graph, then comparing the segmentation graph with a threshold value of the characteristic, expanding and adjusting the image to determine the edge of the infection source marker and determine the infection source marker so as to determine the number of the infection sources;
the sixth step, online big data feeds back "determined infectious agent marker and determined infectious agent number" to the local host.
In a preferred implementation, the infection source specific antibody test substance adopts colloidal gold, the detection camera adopts a dark-field microscopic observation camera, and the detection platform adopts a microscopic observation objective table. It will be appreciated by those skilled in the art that the invention may be embodied in other specific forms without departing from the spirit or essential characteristics thereof. The embodiments disclosed above are illustrative and not exclusive in all respects. All changes which come within the scope of or are equivalent to the scope of the invention are intended to be embraced therein.
Claims (1)
1. A method for detecting clinical infection sources of an infectious department based on big data is characterized by comprising the following steps:
the method comprises the following steps of firstly, extracting an infection source, mixing an infection source specific antibody test object with the infection source, placing the mixed infection source on a platform to be detected, and generating an infection source marker after the infection source specific antibody test object and the infection source are mixed;
secondly, observing and shooting a detection platform by a detection camera to obtain an initial detection image of the infection source, wherein the initial detection image has the characteristics of the infection source marker;
third, uploading the initial detection image to an online big data upper computer through a local host;
the fourth step, the application of the online big data upper computer divides the image into a plurality of units with equal area, and each unit is defined as a graph cell; then classifying the graph cells into a detected target class and a non-detected target class by adopting a per unit value method according to different numerical values of each graph cell; the detection target type is specifically an infection source marker in the graph unit cell, and the non-detection target type is specifically an infection source marker in the graph unit cell;
fifthly, carrying out image segmentation on the graph cells with the detection target class by using a seed filling seed Fill method to obtain a segmentation graph, then expanding and adjusting the image after comparing the segmentation graph with a characteristic threshold value to determine the edge of the infection source marker, and determining the infection source marker to further determine the number of the infection sources;
step six, feeding back the determined infection source marker and the determined number of the infection sources to the local host by the online big data; the infection source specific antibody test object adopts colloidal gold, the detection camera adopts a dark field microscopic observation camera, and the detection platform adopts a microscopic observation objective table; the determination of the number of the infection sources is specifically to determine the number of the infection sources by adopting scanning counting after determining the markers of the infection sources;
the detection method is realized on the basis of a big data clinical infection source detection system of the infectious department, and the system comprises a platform to be detected, a detection camera, a local host and an online big data upper computer; the online big data upper computer is provided with an infection source image dividing module, an infection source image classifying module and an infection source identifying module which are connected in an application configuration mode, wherein the infection source image dividing module is used for dividing an image into a plurality of units with equal areas, each unit is defined as a graph unit cell, the infection source image classifying module is used for classifying the graph units into a detected target class and a non-detected target class by adopting a per unit value method according to different numerical values of each graph unit cell, the infection source identifying module is used for carrying out image segmentation on the graph unit cells with the detected target class by using a seed filling seed Fill method to obtain a segmentation image, then the segmentation image is expanded and adjusted after the segmentation image is compared with a threshold value of a characteristic to determine the edge of an infection source marker, and the infection source marker is determined to further determine the number of the infection source; the application configuration of the online big data upper computer is used for configuring an infection source image online shared database and an infection source image online management module, the infection source image online management module is used for sending a request for acquiring an infection source image to a plurality of distributed storage positions, the plurality of distributed storage positions store the infection source image data under the line, the plurality of distributed storage positions also store indexes and copies of the infection source image data, and the mapping reduction is run in parallel to support the parallel sending of data to the infection source image online shared database; the infection source image on-line management module is also used for awakening the shared database on the infection source image line, so that the shared database on the infection source image line acquires the infection source image data from a plurality of distributed storage positions; the application of the online big data upper computer is also provided with an online read-only learning training module, the online shared database of the infection source image line is also used for supporting the online read-only learning training module to acquire the infection source image data and carry out learning training, and the online read-only learning training module is used for acquiring the infection source image data from the online shared database of the infection source image line and carrying out learning training to optimize the threshold value of the infection source marker feature; the threshold value of each infection source marker feature is determined by cross multiplication of two arrays, wherein one array is a characteristic basis value set (m1, m 2.... mi), and mi is an ordering element of the characteristic basis value set; another array is a variable characteristic coefficient group (ni, n 2.... no), ni is an ordering coefficient item of the variable characteristic coefficient group;
for any one of the graphs or graph cells with the detection target class or the segmentation graphs of the graph cells, sorting the values of the pixel points in the graph into a group to be identified (c1, c 2.. eta.. ci), wherein ci is the ith element of the group to be identified, and if the number of the pixel points is greater than i, adopting the characteristic values of the pixel points as the elements of the group to be identified;
when judging whether a graph or a graph cell of a detection target class or a segmentation graph of the graph cell has a target, calculating for the first time:
P1＝(m1＊n1-c1) 2 +(m2＊n2-c2) 2 +......+(mi＊ni-ci) 2 ；
then, rotating the graph or graph cell with the detection target class or the segmentation graph of the graph cell forward by 360/j degrees, correspondingly changing the ordering of the elements ci of the group to be identified, continuing the calculation to obtain p2, and iterating j-1 times in this way until pj is obtained;
judging whether max (p1, p 2........ pj) is greater than a threshold value or not, and judging that a graph or graph cell of a detection target class or a segmentation graph of the graph cell has a target if max (p1, p 2........ pj) is greater than the threshold value; the learning training to optimize the threshold of the infection source marker feature is to change the value of the variable feature coefficient in the variable feature coefficient set (ni, n2,...... ni) through the data learning training with the judgment result label.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210081573.6A CN114496177B (en) | 2022-01-24 | 2022-01-24 | Method and system for detecting clinical infection source of infectious department based on big data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210081573.6A CN114496177B (en) | 2022-01-24 | 2022-01-24 | Method and system for detecting clinical infection source of infectious department based on big data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114496177A CN114496177A (en) | 2022-05-13 |
| CN114496177B true CN114496177B (en) | 2022-09-16 |
Family
ID=81474770
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210081573.6A Active CN114496177B (en) | 2022-01-24 | 2022-01-24 | Method and system for detecting clinical infection source of infectious department based on big data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114496177B (en) |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106803247B (en) * | 2016-12-13 | 2021-01-22 | 上海交通大学 | Microangioma image identification method based on multistage screening convolutional neural network |
| AU2018225170A1 (en) * | 2017-02-22 | 2019-10-03 | Healthtell Inc. | Methods for screening infections |
| EP3432198B1 (en) * | 2017-07-19 | 2024-04-17 | Tata Consultancy Services Limited | Crowdsourcing and deep learning based segmenting and karyotyping of chromosomes |
| CN109830303A (en) * | 2019-02-01 | 2019-05-31 | 上海众恒信息产业股份有限公司 | Clinical data mining analysis and aid decision-making method based on internet integration medical platform |
| AU2020261370A1 (en) * | 2019-04-24 | 2021-10-14 | Exini Diagnostics Ab | Systems and methods for automated and interactive analysis of bone scan images for detection of metastases |
| CN112884770B (en) * | 2021-04-28 | 2021-07-02 | 腾讯科技（深圳）有限公司 | Image segmentation processing method and device and computer equipment |
| CN113343819B (en) * | 2021-05-31 | 2023-01-03 | 中国电子科技集团公司第十四研究所 | Efficient unmanned airborne SAR image target segmentation method |
-
2022
- 2022-01-24 CN CN202210081573.6A patent/CN114496177B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN114496177A (en) | 2022-05-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9489562B2 (en) | Image processing method and apparatus | |
| CN110245657B (en) | Pathological image similarity detection method and detection device | |
| CN110021425B (en) | Comparison detector, construction method thereof and cervical cancer cell detection method | |
| CN108830332A (en) | A kind of vision vehicle checking method and system | |
| CN111524132B (en) | Method, device and storage medium for identifying abnormal cells in sample to be detected | |
| EP3843036B1 (en) | Sample labeling method and device, and damage category identification method and device | |
| JP5718781B2 (en) | Image classification apparatus and image classification method | |
| CN112102229A (en) | Intelligent industrial CT detection defect identification method based on deep learning | |
| CN114387201B (en) | Cytopathic image auxiliary diagnosis system based on deep learning and reinforcement learning | |
| CN110555836A (en) | Automatic identification method and system for standard fetal section in ultrasonic image | |
| CN108764312B (en) | Optimize multi objective dam defect image detecting method based on DS | |
| CN112132166A (en) | Intelligent analysis method, system and device for digital cytopathology image | |
| CN113658174B (en) | Microkernel histology image detection method based on deep learning and image processing algorithm | |
| CN111914902A (en) | Traditional Chinese medicine identification and surface defect detection method based on deep neural network | |
| CN114445356A (en) | Multi-resolution-based full-field pathological section image tumor rapid positioning method | |
| EP3837664A1 (en) | Generating annotation data of tissue images | |
| CN116468690B (en) | Subtype analysis system of invasive non-mucous lung adenocarcinoma based on deep learning | |
| CN114496177B (en) | Method and system for detecting clinical infection source of infectious department based on big data | |
| CN106682604B (en) | Blurred image detection method based on deep learning | |
| CN112614570A (en) | Sample set labeling method, pathological image classification method and classification model construction method and device | |
| CN111414930B (en) | Deep learning model training method and device, electronic equipment and storage medium | |
| CN115176289A (en) | Cell line development image characterization using convolutional neural networks | |
| CN116205918A (en) | Multi-mode fusion semiconductor detection method, device and medium based on graph convolution | |
| KR101782366B1 (en) | Vision inspection method based on learning data using normalization of sample scale | |
| CN114255388A (en) | Artificial intelligent automatic detection method for embedding quality of seismic acquisition receiving device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |