JP2018063707A - Image analysis system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an image analysis system that makes it possible to compare, as a means of identity confirmation using dental information, a dentistry panorama X-ray image of, for example, a discovered corpse with dentistry-related information about a subject whose identity is to be confirmed and thereby easily obtain information for identity confirmation, the image analysis system further having high reliability pertaining to tooth-number information for the improved accuracy of an image analysis process.SOLUTION: An image analysis system makes it possible to acquire tooth-number information by deep learning of a dentistry panorama X-ray image and easily obtain information for identity confirmation of a subject whose identity is to be confirmed, and further cut out a single tooth image from the dentistry panorama X-ray image by deep learning and, after obtaining tooth-number information by deep learning, manually correct an error in the tooth-number information.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a system for analyzing and classifying images, and in particular, an image analysis system that automatically collects dental information from a dental image using deep learning, a control method for the image analysis system, a program for the image analysis system, and a record It relates to the medium.

  Conventionally, the necessity of information standardization in dentistry is not only indispensable for digitizing medical charts related to medical examinations, but also from social demands other than dental practice, for example, effectively utilizing information in medical clinics, Realization of this is also strongly demanded for national convenience. In addition, there have been many disasters in Japan in recent years, and this has resulted in a large number of deaths, making identification work extremely difficult. In such a situation, the importance of identity verification by dentistry information has been reconfirmed because it is easy to preserve and obtain information, but the identity is confirmed after the death of unidentified persons, such as dental information before birth. It is possible to confirm the identity by comparing with dental information. However, creating the dental information of the dead imposes great difficulty on the person in charge such as a dentist. Furthermore, pre-existing dental information for confirmation of identity is conventionally created by relying on human hands from dental images acquired by general dental examinations, etc., so there is a problem of cost in addition to accuracy problems. To do. Furthermore, when creating dental information from previously stored dental images, if the above dental information is not standardized, effective tooth numbers and tooth status codes are unified to reduce the time required for identification. Further, the generation processing may be hindered, and the accumulated amount of dental images to be processed may be enormous, resulting in further difficulty.

  On the other hand, not only in the medical / dental field, but also in recent academic societies, the progress of research in the deep learning field called so-called deep learning is remarkable, and its application range is expanding.

  In such a situation, for example, as part of confirmation work for unidentified persons, analyzing dental images by deep learning using the latest technology to obtain dental information related to the dentist's dentistry can be used to identify unidentified persons. The inventor has come to be convinced that it not only plays an important role but also makes a great social contribution.

  In addition, as an image for grasping the dental situation of each individual, for example, various image collection techniques have been developed for conventional X-ray photographs and computed tomography (hereinafter abbreviated as CT) images. Therefore, it is easy to obtain dental information necessary for deep learning.

  However, even though deep learning technology is making great progress, there should be no mistakes in work that bears significant social responsibilities, such as confirmation work for unidentified persons. Needless to say. After all, parts that cannot be achieved by deep learning need to be perfect even if they rely on human hands.

  In relation to such an image analysis system, a technical idea for analyzing a medical / dental image by using various techniques and identifying it as an individual is disclosed.

JP2015-154918A JP 2014-036880 A

  For example, in Japanese Patent Application Laid-Open No. 2015-154918, for example, a deep learning (also referred to as deep learning) technique is applied, for example, lesion candidates are detected from medical information, anatomical objects are detected, and the positions and anatomy of lesion candidates are detected. The technical idea of verifying lesion candidates based on anatomical context information including information on the position of the target object and removing false positive lesion candidates has already been published, but for example, regarding dentistry, In particular, regarding the classification of tooth types and conditions, a detailed technical idea is not disclosed.

  Furthermore, it does not disclose a detailed technical idea of providing information for searching for an unidentified person. Further, for example, there is no disclosure regarding improvement of a learning function in a deep learning (also referred to as deep learning) technique.

  Further, for example, in Japanese Patent Application Laid-Open No. 2014-036880, for example, it is possible to display a periodontal treatment state superimposed on a digital image, and only by referring to the digital image without referring to a medical record or the like, the periodontal examination can be performed. The technical idea about medical image diagnosis support that not only makes it possible to grasp the guidelines but also visually confirm the periodontal treatment history has already been published. In the unlikely event that an error occurs in the information in which the tooth position and the tooth outline are automatically created from the above, a detailed technical idea for correction is not disclosed.

  In addition, it is indispensable as information for efficient search of unidentified people, for example, regarding the detailed technical idea of tooth state classification processing that classifies teeth as healthy teeth, partial restoration, full restoration, loss, etc. Not disclosed. Further, for example, a technical idea for recognizing that an error has occurred in a deep learning (also referred to as deep learning) technique and contributing to further improvement of the learning function is not disclosed.

  The present invention is intended to solve the above-described problems. In particular, as an aid to identification work utilizing dental information, for example, the analysis result and identification of a found body by a dental panoramic X-ray image are determined. An object of the present invention is to provide an image analysis system that makes it possible to easily obtain identification confirmation information by collating with information on dentistry before the subject's life. Another object of the present invention is to provide an image analysis system having high reliability with respect to tooth number information in order to improve the accuracy of image analysis processing.

Thus, in order to solve the above-described problems, the present application focuses on the fact that the image analysis result is important information for unidentified person search. That is, in the image analysis system according to the present invention,
In an image analysis system comprising a management system having at least a management apparatus server and a management apparatus database, and an analysis system having an analysis apparatus processor and an analysis apparatus database,
Means for inputting first information relating to an image type to be subjected to image analysis processing by the management device server and / or the analysis device processor;
Means for storing said first information as a first storage on said management device server and / or analysis device database by said management device server and management device database and / or by said analysis device processor and analysis device database;
Means for inputting second information relating to a tooth number assigned to the image according to the image type by the management device server and / or the analysis device processor;
Means for storing the second information as a second storage on the management device server and / or the analysis device database by the management device server and the management device database and / or by the analysis device processor and the analysis device database;
Means for inputting third information relating to a tooth state for each tooth number by the management device server and / or the analysis device processor;
Means for storing the third information as a third storage on the management device server and / or the analysis device database by the management device server and the management device database and / or by the analysis device processor and the analysis device database; It is characterized by having.
further,
The management system and / or the analysis system further includes means for displaying at least one of the first, second, and third information.
Also,
In an image analysis system comprising a management system having at least a management apparatus server and a management apparatus database, and an analysis system having an analysis apparatus processor and an analysis apparatus database,
Means for inputting first information relating to an image type to be subjected to image analysis processing by the management device server and / or the analysis device processor and cutting out an image of a single tooth from the image according to the image type;
Means for storing said first information as a first storage on said management device server and / or analysis device database by said management device server and management device database and / or by said analysis device processor and analysis device database;
Means for cutting out a single tooth image from an image according to the image type by the management device server and / or the analysis device processor;
Means for inputting second information relating to a tooth number attached to the cut tooth single image by the management device server and / or the analysis device processor;
Means for storing the second information as a second storage on the management device server and / or the analysis device database by the management device server and the management device database and / or by the analysis device processor and the analysis device database;
Means for inputting third information relating to a tooth state for each tooth number by the management device server and / or the analysis device processor;
Means for storing the third information as a third storage on the management device server and / or the analysis device database by the management device server and the management device database and / or by the analysis device processor and the analysis device database; It is characterized by having.
further,
The management system and / or the analysis system further includes means for displaying at least one of the first, second, and third information.
Also,
Means for inputting fourth information relating to manual change of information on the tooth number by the management device server and / or the analysis device processor;
Means for storing said fourth information as a fourth storage on said management device server and / or analysis device database by said management device server and management device database and / or by said analysis device processor and analysis device database;
Means for manually changing the tooth number by the fourth storage by the management device server and / or the analysis device processor;
Means for inputting fifth information related to a manual tooth number change by the management device server and / or the analysis device processor;
Means for storing the fifth information as a fifth storage on the management device server and / or the analysis device database by the management device server and the management device database and / or by the analysis device processor and the analysis device database; It is further provided with a feature.
further,
The management system and / or the analysis system further comprises means for displaying at least one of the fourth and fifth information, and means for displaying information relating to the fifth storage.
Also,
Means for inputting sixth information relating to incorporating information relating to the tooth number changed manually by the management device server and / or the analysis device processor as a resource of image analysis processing;
Means for storing the sixth information as a sixth storage on the management device server and / or the analysis device database by the management device server and the management device database and / or by the analysis device processor and the analysis device database;
Means for incorporating the fifth information as a resource for image analysis processing by the sixth information by the management device server and / or the analysis device processor.
Furthermore,
The management system and / or the analysis system further includes means for displaying the sixth information.

Furthermore, the present invention can be realized as a control method for an image analysis system. That is,
In an image analysis system comprising a management system having at least a management apparatus server and a management apparatus database, and an analysis system having an analysis apparatus processor and an analysis apparatus database,
First information relating to an image type to be subjected to image analysis processing by the management device server and / or the analysis device processor is input,
Storing the first information as a first storage on the management device server and / or analysis device database by the management device server and management device database and / or by the analysis device processor and analysis device database;
Input second information relating to the tooth number assigned to the image according to the image type by the management device server and / or the analysis device processor,
Storing the second information as a second storage on the management device server and / or the analysis device database by the management device server and the management device database and / or by the analysis device processor and the analysis device database;
Input third information related to the tooth condition for each tooth number by the management device server and / or the analysis device processor,
The third information is stored as a third storage on the management device server and / or the analysis device database by the management device server and the management device database and / or by the analysis device processor and the analysis device database. It is characterized by that.
Also,
The management system and / or the analysis system further includes displaying at least one of the first, second, and third information.
further,
In an image analysis system comprising a management system having at least a management apparatus server and a management apparatus database, and an analysis system having an analysis apparatus processor and an analysis apparatus database,
Input first information relating to image types to be subjected to image analysis processing by the management device server and / or the analysis device processor and to cutting out a single tooth image from the image according to the image type,
Storing the first information as a first storage on the management device server and / or analysis device database by the management device server and management device database and / or by the analysis device processor and analysis device database;
A single tooth image is cut out from the image related to the image type by the management device server and / or the analysis device processor,
Input second information related to the tooth number attached to the cut tooth single image by the management device server and / or the analysis device processor,
Storing the second information as a second storage on the management device server and / or the analysis device database by the management device server and the management device database and / or by the analysis device processor and the analysis device database;
Input third information related to the tooth condition for each tooth number by the management device server and / or the analysis device processor,
The third information is stored as a third storage on the management device server and / or the analysis device database by the management device server and the management device database and / or by the analysis device processor and the analysis device database. It is characterized by that.
Also,
The management system and / or the analysis system further includes displaying at least one of the first, second, and third information.
Furthermore,
Input fourth information related to manually changing information on the tooth number by the management device server and / or the analysis device processor,
Storing the fourth information as a fourth storage on the management device server and / or the analysis device database by the management device server and the management device database and / or by the analysis device processor and the analysis device database;
The tooth number is manually changed by the fourth storage by the management device server and / or the analysis device processor,
Input fifth information related to manual tooth number change by the management device server and / or the analysis device processor,
Storing the fifth information as a fifth storage on the management device server and / or the analysis device database by the management device server and the management device database and / or by the analysis device processor and the analysis device database. It is characterized by having.
further,
The management system and / or the analysis system further comprises displaying at least one of the fourth and fifth information.
Also,
Inputting sixth information relating to incorporating information relating to the tooth number changed manually by the management device server and / or the analysis device processor as a resource for image analysis processing;
Storing the sixth information as a sixth storage on the management device server and / or the analysis device database by the management device server and the management device database and / or by the analysis device processor and the analysis device database;
The management apparatus server and / or the analysis apparatus processor further includes incorporating the fifth information as a resource for image analysis processing according to the sixth information.
further,
The management system and / or the analysis system further includes displaying the sixth information.

Furthermore, the present invention can be realized as a program for an image analysis system. That is,
In an image analysis system comprising a management system having at least a management apparatus server and a management apparatus database, and an analysis system having an analysis apparatus processor and an analysis apparatus database,
Means for inputting first information relating to an image type to be subjected to image analysis processing by the management device server and / or the analysis device processor;
Means for storing said first information as a first storage on said management device server and / or analysis device database by said management device server and management device database and / or by said analysis device processor and analysis device database;
Means for inputting second information relating to a tooth number assigned to the image according to the image type by the management device server and / or the analysis device processor;
Means for storing the second information as a second storage on the management device server and / or the analysis device database by the management device server and the management device database and / or by the analysis device processor and the analysis device database;
Means for inputting third information relating to a tooth state for each tooth number by the management device server and / or the analysis device processor;
Means for storing the third information as a third storage on the management device server and / or the analysis device database by the management device server and the management device database and / or by the analysis device processor and the analysis device database. It is provided with operating.
Also,
The management system and / or the analysis system further comprises a means for displaying at least one of the first, second, and third information.
further,
In an image analysis system comprising a management system having at least a management apparatus server and a management apparatus database, and an analysis system having an analysis apparatus processor and an analysis apparatus database,
Means for inputting first information relating to an image type to be subjected to image analysis processing by the management device server and / or the analysis device processor and cutting out an image of a single tooth from the image according to the image type;
Means for storing said first information as a first storage on said management device server and / or analysis device database by said management device server and management device database and / or by said analysis device processor and analysis device database;
Means for cutting out a single tooth image from an image according to the image type by the management device server and / or the analysis device processor;
Means for inputting second information relating to a tooth number attached to the cut tooth single image by the management device server and / or the analysis device processor;
Means for storing the second information as a second storage on the management device server and / or the analysis device database by the management device server and the management device database and / or by the analysis device processor and the analysis device database;
Means for inputting third information relating to a tooth state for each tooth number by the management device server and / or the analysis device processor;
Means for storing the third information as a third storage on the management device server and / or the analysis device database by the management device server and the management device database and / or by the analysis device processor and the analysis device database. It is provided with operating.
Also,
The management system and / or the analysis system further comprises a means for displaying at least one of the first, second, and third information.
further,
Means for inputting fourth information relating to manual change of information on the tooth number by the management device server and / or the analysis device processor;
Means for storing said fourth information as a fourth storage on said management device server and / or analysis device database by said management device server and management device database and / or by said analysis device processor and analysis device database;
Means for manually changing the tooth number by the fourth storage by the management device server and / or the analysis device processor;
Means for inputting fifth information related to a manual tooth number change by the management device server and / or the analysis device processor;
Means for storing the fifth information as a fifth storage on the management device server and / or the analysis device database by the management device server and the management device database and / or by the analysis device processor and the analysis device database. It is further characterized by being operated.
Furthermore,
The management system and / or the analysis system further comprises a means for displaying at least one of the fourth and fifth information.
further,
Means for inputting sixth information relating to incorporating information relating to the tooth number changed manually by the management device server and / or the analysis device processor as a resource of image analysis processing;
Means for storing the sixth information as a sixth storage on the management device server and / or the analysis device database by the management device server and the management device database and / or by the analysis device processor and the analysis device database;
The management apparatus server and / or the analysis apparatus processor further comprises means for incorporating the fifth information as a resource for image analysis processing according to the sixth information.
Also,
The management system and / or the analysis system further includes a means for displaying the sixth information.

  ADVANTAGE OF THE INVENTION According to this invention, it is possible to acquire tooth number information by deep learning of a dental panoramic X-ray image, to easily obtain the identity confirmation information of the person whose identity is to be determined, and to improve the accuracy of automatic generation processing. By using deep learning, an image of a single tooth is cut out from a dental panoramic X-ray image and used as an analysis target image in image analysis, and an error in tooth number information obtained by deep learning from an image of a single tooth cut out from the image as described above is obtained. It is possible to construct an image analysis system having higher accuracy and reliability by correcting manually.

It is a functional block diagram of the whole image analysis system concerning one embodiment of the present invention. It is a block diagram of the management system which concerns on one Embodiment of this invention. It is a block diagram of the analysis system which concerns on one Embodiment of this invention. It is an image analysis management data table concerning one embodiment of the present invention. It is an image processing summary list concerning one embodiment of the present invention. It is an analysis process detail which concerns on one Embodiment of this invention. It is a tooth type and tooth number display data table concerning one embodiment of the present invention. It is a tooth type / tooth number display / partial display data table according to an embodiment of the present invention. 4 is a tooth type / tooth number display / synthesis display data table according to an embodiment of the present invention. It is a tooth number information list concerning one embodiment of the present invention.

  In the present invention, the tooth state information is finally acquired from the dental image information by deep learning, and the tooth number information error obtained by deep learning from the image of a single tooth cut out from the image is manually corrected. The aim is to realize an image analysis system with higher accuracy and reliability.

  Hereinafter, embodiments for carrying out the present invention will be described with reference to the drawings. FIG. 1 is a functional block diagram showing an outline of the configuration of an image analysis system according to an embodiment of the present invention. An image analysis system according to an embodiment of the present invention is configured by connecting a management system 20 and an analysis system 30 via a network 10. In addition, if necessary, for example, an image capturing system using dental panoramic X-rays, a related institution system 202 including an image database for storing captured images, and the like are connected via the network 10, and a system configuration is configured as necessary. You may comprise by what is called cloud computing. Further, for example, necessary software or the like may be installed from the management system 20 to the analysis system 30.

  Next, details of the management system 20 will be described with reference to FIG. As shown in FIG. 1, the management system 20 includes a management device database (hereinafter referred to as “management device DB”) 21, a management device server 22, a management output device 23, a management input device 24, and a management device interface 26. It is configured. The management apparatus server 22 processes data relating to image analysis management by a processor or the like, and the management apparatus DB 21 stores data such as management and image analysis information. The management output device 23 includes a display, a printer, and the like, and outputs and displays various types of information as necessary. The management input device 24 includes a keyboard, a barcode reader, a scanner, and the like, and inputs information as necessary. However, the management input device 24 includes all devices that can input information. The management system 20 may be composed of a single system or a plurality of systems as long as the management system 20 can finally perform the business as an image analysis system.

  Next, details of the analysis system 30 will be described with reference to FIG. As shown in the figure, the analysis system 30 includes an analysis device memory (also referred to as an analysis device database) 31, an analysis device processor 32, an analysis output device 33, an analysis input device 34, and an analysis device interface 36. . The analyzer processor 32 processes, for example, information sent from the management system 20 and displays the information on the analyzer output device 33, for example. The analysis output device 33 includes a display, a printer, and the like, and outputs and displays various types of information as necessary. The analysis input device 34 includes a keyboard, a barcode reader, a scanner, and the like, and inputs information as necessary, but includes all devices that enable input of information. In the analysis input device 34, input processing is performed from a keyboard or the like as necessary, and displayed on the analysis output device 33, for example, and information input from the analysis input device 34 is input to the analysis device interface 36 as necessary. To the management system 20. The management system 20 performs image analysis management processing according to the transmitted information, and when further information is required, further information input is performed from the analysis system 30. In this way, the analysis system 30 is operated to perform image analysis management processing.

  The outline and configuration of an embodiment of the present invention have been described above. Note that each system other than the management system 20 and the analysis system 30 also has a processor, a memory, and the like (not shown), and similarly to the management system 20 and the analysis system 30, performs various types of information processing as necessary. Assumed to be performed. Next, details and operations of the present invention will be described.

  First, the image analysis management data table 40 that stores information for managing a series of processes for analyzing and classifying images will be described with reference to FIG. The image analysis management data table 40 includes an analysis management code column 41 for storing an analysis management code and the like to be subjected to image analysis, a data table management column 42 for storing management contents, and an image type column 4301 indicating various types of image analysis information. , An image cutout column 4302, a deep learning column 4303, and so on. Further, the data table management column 42 is further provided with an item column 421, a content column 422,. The image analysis management data table 40 is, for example, an image analysis setting screen displayed on the display of the analysis output device 33 or information calculated from related information by the analysis device processor 32 according to a program stored in the analysis device memory 31. (Not shown), for example, information input by an operator from the keyboard of the analysis input device 34 of the analysis system 30 according to a program stored in the analysis device memory 31 is analyzed on the analysis device memory 31 by the analysis device processor 32. Shall be stored in The image analysis setting screen (not shown) has the same format as the image analysis management data table 40. Further, if necessary, an image analysis management data table display screen (not shown) according to the contents of the image analysis management data table 40 is displayed, for example, an analysis output device 33 of the analysis system 30 or a management output device 23 of the management system 20. Shall be displayed. The image analysis management data table 40 may be input on the management system 20 side and stored in the memory. The information input and stored on the analysis system 30 side may be transmitted to the management system 20 side via the network 10, for example. The information input and stored on the management system 20 side may be transmitted to the analysis system 30 side. You can send it.

  Next, a procedure for storing information in each column of the image analysis management data table 40 will be described with reference to FIG. First, when performing image analysis in dentistry, the image analysis management data table 40 is created for each analysis. The code for managing image analysis is 123456-001, and the created image analysis date is September 2016. If it is one day, for example, information “image analysis management table analysis management code: 123456-001 date: 2016.09.01” is stored. Next, for example, information “item” is stored in the item column 421 of the data table management column 42, and information “content” is stored in the content column 422, respectively. Further, the image type field 4301 includes, for example, an analysis target among images of panorama X-ray, CT, CT panorama specification conversion image, stereoscopic image (hereinafter abbreviated as 3D) that can be processed by the image analysis system. When the image is an image based on a dental panoramic X-ray and the image code based on the image based on the dental panoramic X-ray is SPX 16001, for example, information “dental panoramic X-ray: SPX 16001” is stored in the image type column 4302. The tooth is cut out from the image by the individual unit, and as a result, there is an image cut out. For example, SPX16001-RMU3, SPX16001-RMU2,..., SPX16001-RLU1,. If an image code such as “No” is added, “Yes: SPX16001-RMU” , SPX16001-RMU2,..., SPX16001-RLU1,..., ”In the deep learning column 4203, when processing the target image by deep learning, for example, using a network called“ subaru ” For example, when the network is numbered with a code such as ANM001-0001, the information “Subaru: ANM001-0001” is displayed in the framework column 4304. When the target image is processed by the framework, for example, the frame “tea” is displayed. When a work is used and a code such as MFW001-0012 is added to the framework, information “tea: MFW001-0012” is displayed in the output 1 column 4305, for example, as a result of image analysis. 7 dentistries for each tooth to be analyzed When the system is instructed to output the result classified into the position, and the output information output as a result of the instruction is given a code such as 123456-001-OP01-001, for example, “7 Dental classification: 123456-001” In the output 2 column 4306, for example, as an image analysis result, the system is instructed to output a tooth list as a result of analyzing the entire analysis target image, and the result of the instruction is output. For example, if a code of 123456-001-SS01-001 is added to the output information, information “dental formula: 123456-001-SS01-001” is obtained from the image of the dental panoramic X-ray in the image type column 4302, for example. The information “Yes: ...,” is stored so that each tooth is cut out individually. When the tooth number attached by the cutout process can be corrected manually, the output 3 field 4307 has information “Tooth number correction (manual) exists”, for example, the image cutout field 4302 has a dental panoramic X-ray. When the information “Yes:...” Is stored so as to perform the processing of cutting out the teeth from the image according to the above, and the tooth numbers numbered according to the above-described cutting processing can be manually corrected, the output 4 column In 4308, information indicating that “tooth number correction (manual) result feedback is present” is classified into, for example, information classified into four states of healthy teeth, partial repair, complete repair, and missing, for example, 123456-001-SJ01-001. Is stored in the output 5 column 4309 with the information “tooth state: 123456-001-SJ01-001”, respectively. And things. In addition, regarding various image types, when various images exist, the image analysis management data table 40 is created and managed for each individual image.

  Further, in the analysis image accumulation column 4310, information “Yes” is accumulated when the image subjected to image analysis is accumulated as learning data in deep learning, for example, and the image is accumulated in the accumulation image processing 1 column 4311. For example, when gamma conversion is performed and a code “GH-001” is added to the gamma conversion 001, information “gamma conversion 001: GH-001” is stored in the stored image processing 2 column 4312. When the image is stored, for example, contrast adjustment is performed. For example, when the code “CC-002” is added to the contrast adjustment 002, information “contrast adjustment 002: CC-002” is stored. Furthermore, in the accumulated image processing 3 column 4313, for example, noise is added when the image is accumulated, and the noise is added. For example, when a code NF-001 is added to 01, information “noise addition 001: NF-001” is stored respectively, and when further image processing is performed, the image analysis management data table 40 is stored. It shall be added to. Of the item information of the image analysis management data table 40, the image types that can be extracted from other item information, for example, the image types stored in the image type column 4301, for example, names related to panoramic X-rays, CT, 3D, etc. A list (not shown) based on codes may be created, selected from the list, extracted, and stored. Further, for example, with respect to deep learning stored in the deep learning column 4303, a list (not shown) with names and codes related to usable, for example, Subaru network, Subaru kunai network, to demo network, etc. is created, and the list You may select and extract from the table and store it. Further, regarding the accumulated image processing 1 column 4311, the accumulated image processing 2 column 4312, the accumulated image processing 3 column 4313,..., Etc., for example, gamma conversion 001, gamma conversion 002,. A list (not shown) regarding the set image processing such as adjustment 002,..., Noise addition 001, noise addition 002,... May be created and selected from the list. In these cases, it is assumed that the list creation process is performed by the analyzer processor 32 or the like in accordance with a program developed on the analysis system 30, for example. The same processing may be performed by the management system 20. Further, the result may be transmitted to the analysis system 30 via the network 10, for example.

  In the above description, as the storage information of each column of the image analysis management data table 40, for example, a character code by numbers is stored. However, for example, information such as a uniquely set code or the character code may be stored. The column may be further added with information such as a uniquely set code.

  Next, an image processing summary list 50 that summarizes the summary of image analysis processing according to the information stored in the image analysis management data table 40 will be described with reference to FIG. The image processing summary list 50 includes an image processing summary management column 51 indicating each item of the image processing summary, an image processing content column 52 indicating the image analysis status, and the like. The image processing summary management column 51 includes an image type column 511, an image cutout column 512, an analysis method (deep learning) column 513, a framework column 514, an analysis processing column 515, a result output column 516, and a learning data processing column 517. , A result analysis column 518, a feedback column 519, and so on.

  Next, specific contents shown in the image processing summary list 50 will be described as follows. First, in the image type column 511, a dental image type selection process 5111 is performed. Here, for example, when the information “dental panorama X-ray” is stored in the image type column 4301, the dental panorama X-ray is selected. Next, in the image cutout column 512, it is determined whether or not the information “Yes:...” Is stored in the image cutout column 4302 by the image cutout non-processing 5121 and the image cutout processing 5122. For example, if there is a dental panoramic X-ray image according to the image code stored in the image type column 4301 to the deep learning selection processing 5131 in the analysis method (deep learning) column 513 via the image cut-out non-processing 5121 Analyzing method (deep learning) of the dental panoramic X-ray image according to the image code stored in, for example, the image type column 4301, for example, the image extraction processing 5123 of a single tooth via the image cutout processing 5122 (FIG. (Not shown) to cut out an image of a single tooth and, for example, an image code stored in the image cutout column 4302, SPX 6001-RMU3, methods analyzed Fuban a ... etc. (deep learning) to the deep learning selection process 5131 column 513. Next, in the deep learning selection processing 5131 in the analysis method (deep learning) column 513, for example, deep learning is selected according to the deep learning code stored in the deep learning column 4303, for example, ANM001-0001, and each process in the image cutout column 512 is selected. Information related to image processing is sent to the framework selection processing 5141 in the framework column 514 together with the dental images sent from 5121 and 5123. In the framework selection processing 5141, for example, a framework is selected according to a framework code stored in the framework column 4304, for example, MFW001-0012, and information related to image processing sent from the deep learning selection processing 5131 together with the framework code. Is sent to the analysis processing 5151 in the analysis processing column 515. In the analysis processing 5151, deep learning is performed based on a learning database (hereinafter abbreviated as learning DB) 5152 and a teacher signal (not shown), and for example, the results for each of the seven dental classifications analyzed by the deep learning are output, for example. 7 dental classification-specific codes stored in the first column 4305 are numbered 16001-001-OP01-001. Further, for example, the result of the dental formula is, for example, the dental formula code stored in the output 2 column 4306, 16001-001. -SS01-001 is attached, for example, the tooth state result is attached with, for example, the tooth state code 16001-001-HJ01-001 stored in the output 5 column 4309, and the above information is stored in the result output column 516. The result is sent to the result output process 5161.

  In the result output processing 5162, for example, according to each output information stored in the output 1 column 4305, the output 2 column 4306, the output 3 column 4307,... While outputting and displaying with the apparatus 5162, the said dental analysis result information is sent to the result analysis process 5181 of the result analysis column 518. FIG. Furthermore, in the comparison process 5182, the above-mentioned dental analysis result information and separate deep analysis result data 5183 storing the result information of image analysis by deep learning by a method different from the processing method of the image type column 511 to the analysis processing column 515, etc. Alternatively, when there is information analyzed by different processing methods such as non-depth analysis results 5195 or manual analysis results 5196 other than deep analysis, for example, a comparison process with the analysis results by separately storing analysis result data 5192 stored therein is performed. For example, according to the setting of the image analysis management data table 40, for example, the analysis accuracy of dental image processing by deep learning application and the analysis accuracy by other dental image analysis processing are compared. The comparison result information such as the analysis accuracy is subjected to analysis processing to determine which information is fed back by the comparison result feedback processing 5191 in the feedback column 519, and is stored in, for example, the image type 5193, the image cutout 5194, or the like as necessary. Information shall be fed back to the relevant department. Furthermore, in the image type selection processing in the image type selection processing 5111, for example, a processed image that has been subjected to the image processing specified in the stored image processing 1 column 4311 may be selected. Further, the image information used for the image type selection process may be stored in, for example, the image type 5193, for example, together with the applied image processing content.

  Next, for example, when the information “present” is stored in the analysis image storage column 4310 of the image analysis management data table 40, the dental image used for the image analysis is stored in the learning DB for deep learning. However, when analysis processing is performed by deep learning, generally, the more information in the learning DB, the better the performance. Therefore, in order to increase image information, for example, stored in the accumulated image processing 1 column 4311 According to the information “gamma conversion 001: GH-001”, for example, conversion is performed by gamma conversion processing 5171, and the result is stored in the learning DB 5152. Similarly, for example, in accordance with the processing code “contrast adjustment 002: CC-002” stored in the accumulated image processing 2 column 4312, for example, image information adjusted by contrast adjustment 5172, and further in the accumulated image processing 3 column 4313 In accordance with the stored processing code “noise addition 001: NF-001”, for example, image information that has undergone noise addition by noise addition 5173,..., Etc. is incorporated into resources for improving the accuracy of deep learning, for example. Further, it is assumed that the data is further stored in the learning DB 5152. In addition, the image that has undergone the accumulated image processing is not only accumulated as learning information in the learning DB 5152, but for example, an image processed image, for example, an image type selection process 5111, a single tooth image extraction process 5123, an analysis process 5151, etc. May be used as a base image for each processing. Alternatively, for example, image processing by the accumulated image processing 1 column 4311, the accumulated image processing 2 column 4312, the accumulated image processing 3 column 4313,..., Etc. is sequentially performed at each processing stage of the image processing summary list 50 (not shown). ) For example, the accuracy of image analysis processing may be improved. Each process described above is performed by the analyzer processor 32 according to a program stored in the analyzer memory 31, for example. Further, if necessary, processing is performed by the management device server 22 and / or the processor (not shown) of the related engine system according to the program stored in the management device DB 21 and / or the memory (not shown) of the related engine system. For example, it may be transmitted to the analysis system 30 or the like via the network 10.

  As described above, the processing outline of the image analysis management system according to the present invention is shown according to the image analysis management data table 40 and the image processing outline list 50. However, the image analysis management data table 40 is in principle new each time image analysis is performed. For example, an analysis management code “123456-001”, which is created and stored in the image analysis management code column 41, is assigned. On the other hand, the code of each output information generated as a result of image analysis stored in the output 1 column 4305, the output 2 column 4306, the output 3 column 4307,. Code information “123456-0001-OP01-001” is stored, but “123456-0001” included in the code for each of the seven dental classifications is an analysis management code “123456-” stored in the image analysis management code column 41. 001 "indicates that they are related to each other. Further, by making the information of the codes related to each analysis result into a data table (not shown), the output information of the image analysis management system according to the present invention is accumulated and, for example, the corresponding image analysis management data table 40 is stored. Manage relationships by associating. Further, in the case where the analyzed image information is stored in the learning DB 5152 in accordance with, for example, the information “present” stored in the analysis image accumulation column 4310, for example, the single image code for each tooth stored in the image cutout column 4302 Numbering shall be performed in accordance with “SPX16001-RMU3” etc. and managed separately. Note that, when a specific tooth is selected from various expression formats, for example, the output 2 column 4306 may store information on the specific tooth.

  Next, details of the single tooth image extraction processing 5123 in the image cutout column 512 of the image processing summary list 50 will be described below. The image cutout in the image extraction process 5123 is, for example, from a dental panoramic X-ray image stored in the image type column 4301 by deep learning, for example, according to a separately set learning DB (not shown) relating to a single tooth image. Are cut out and set as analysis target images. In the image cutout process in the image extraction process 5123 for a single tooth, information for the cutout process may be obtained from a processed image by image processing specified in the accumulated image processing 1 column 4311, for example. Further, the image information used for the cutout process may be stored in, for example, the image cutout 5194 together with the contents of the applied image processing. The details of the deep learning process related to image clipping are based on the deep learning in the analysis method (deep learning) column 513, the framework column 514, and the analysis processing column 515. In addition, the deep learning in the analysis processing column 515 and the deep learning (not shown) in the image cutout column 512 may be provided with different deep learning mechanisms, but the same deep learning function is used for each processing by the deep learning. They can be used in association.

  As described above, the outline of the image analysis processing has been described by, for example, integrating the analysis processing 5151 shown in FIG. 5, but actually, at least the tooth number recognition and, for example, the state of the teeth is healthy teeth, partial restoration, all The processing is divided into four categories, namely, restoration and defect classification. The tooth number recognition process result is not directly passed to the tooth state classification process, which is also executed by deep learning, but it is determined whether there is an error in the tooth number identification process, for example, manually, and the result is It is assumed that the tooth state classification processing is performed based on the original. Details of the above processing will be described in accordance with analysis processing details 60 shown in FIG. The analysis processing details 60 are the detailed drawings of the analysis processing 5151 shown in FIG. 5, the tooth number recognition processing 51511, the tooth number correction determination processing 51512, the tooth number correction processing (manual) 51513, the tooth state classification processing 51514, the tooth number correction. The result is composed of Fordback switching processing 515131 and the like. Note that the tooth number recognition processing 51511 and the tooth state classification processing 51514 are processed by, for example, deep learning, and the specific processing content is the same as the analysis processing 5151. Next, details of the tooth number correction determination process 51512 and the tooth number correction process (manual) 51513 will be described below. First, the tooth number correction determination processing 51512 is processed by, for example, deep learning, for example, according to the information “having tooth number correction (manpower) present” stored in the output 3 column 4307 of the image analysis management data table 40 shown in FIG. Information from the tooth number recognition process 51511 is sent to a tooth number correction process (manual) 51513 by a tooth number correction determination process 51512, and a manual tooth number correction process is performed. Also, if the output 3 column 4307 stores, for example, information “No tooth number correction (manual)” or no information is stored, the information from the tooth number recognition process 51511 is the tooth number correction. The decision processing 51512 sends the tooth state classification processing 51514 without correction. As described above, for example, the information “... Yes” is stored in the output 3 column 4307, so that the selection of the tooth number correction determination process 51512 becomes the correction side, and the correction is performed in the tooth number correction process (manual) 51513. Processing is performed. It should be noted that information storage in the output 3 column 4307 conforms to other items of the image analysis management data table 40.

  Next, specific processing in the tooth number correction processing (manual) 51513 will be described with reference to the tooth type / tooth number display data table 70 shown in FIG. The tooth type / tooth number display data table 70 is a tooth type / tooth number management code column 71 for storing an analysis management code related to a tooth type / tooth number to be corrected, and correction management 1 for storing management information about correction. A column 721, a correction management 2 column 722, a tooth selection column 7311 for storing the selection content of the tooth, a tooth information column 7312 for storing tooth information, a display switching column 7313 for storing display switching information of tooth number, and The tooth number storage 1 column 7321, the tooth number storage 2 column 7323, the tooth number storage 3 column 7325, the tooth number storage 4 column 7327,... A single image 1 column 7322, a single image 2 column 7324, a single image 3 column 7326, a single image 4 column 7328,... Further, the display switching field 7313 includes a switching 1 field 73131, a switching 2 field 73132, a switching 3 field 73133,..., And a tooth number storage 1 field 7321, a tooth number storage 2 field 7323, and a tooth number storage 3 field. 7325,..., Single image 1 column 7322, single image 2 column 7324, single image 3 column 7326,..., Etc. are tooth number 1 column 72201, tooth number 2 column 72202, tooth number 3 , No. 8 column 72208,..., Etc. are further provided. The tooth type / tooth number display data table 70 is, for example, information calculated from related information by the analyzer processor 32 according to a program stored in the analyzer memory 31 or teeth displayed on the display of the analysis output device 33. According to a formula / tooth number display data table setting screen (not shown), for example, information input by an operator from the keyboard of the analysis input device 34 of the analysis system 30 is analyzed according to a program stored in the analysis device memory 31. It is assumed that the data is stored on the analyzer memory 31 by the processor 32. Further, the tooth type / tooth number display data table setting screen (not shown) has the same format as the tooth type / tooth number display data table 70. Further, if necessary, a tooth type / tooth number display data table display screen (not shown) according to the contents of the tooth type / tooth number display data table 70 is displayed, for example, an analysis output device 33 of the analysis system 30, a management system. It is displayed on 20 management output devices 23. The tooth type / tooth number display data table 70 may be input and stored in the memory on the management system 20 side. The information input and stored on the analysis system 30 side may be transmitted to the management system 20 side via the network 10, for example. The information input and stored on the management system 20 side may be transmitted to the analysis system 30 side. You can send it.

  Next, a procedure for storing information in each column of the tooth type / tooth number display data table 70 will be described with reference to FIG. First, when tooth number correction is performed according to the information “tooth number correction (manual) present” stored in the output 3 column 4307 of the image analysis management data table 40, the tooth type and tooth number are displayed for each analysis based on the tooth image. The data table 70 is created, and in the tooth type / tooth number management code column 71, a code for managing image analysis is 123456-001. Further, for example, an image type used for image analysis is an image based on dental panoramic X-ray. Code: SPX16001, and the date when the created image analysis was performed is September 1, 2016, for example, “Tooth type / tooth number display analysis management code: 123456-001-SPX16001 Date: 2016.09.01 Is stored. Next, for example, “Tooth Formula” is displayed in the correction management 1 column 721, “Tooth List Display” is displayed in the Correction Management 2 column 722, and “Tooth Formula Selection: 1 2 3... ”In the tooth number storage 1 field 7321, for example,“ 1 ”in the tooth number 1 field 72201,“ 2 ”in the tooth number 2 field 72202, and“ 3 ”in the tooth number 3 field 72203, for example. ,..., Information “8”, for example, in the tooth number 8 column 72208, and similarly, for example, “9”, tooth number 2 in the tooth number 1 column 72201 of the tooth number storage 2 column 7323. For example, information “10” is stored in the column 72202, information “11” is stored in the tooth number 3 column 72203, and information “16” is stored in the tooth number 8 column 72208. For example, information “32” is stored in the tooth number 8 column 72208 of the field 7327, respectively. To.

  The storage of basic information in the tooth type / tooth number display data table 70 has been described above, but the storage of detailed information in the tooth type / tooth number display data table 70 will be described below. For example, in the analysis system 30, for example, when an operator instructs the display of the tooth type / tooth number display data table 70 using, for example, the keyboard of the analysis input device 34, the tooth number display data table is displayed on the display of the analysis output device 33, for example. A tooth type / tooth number display data table display screen (not shown) according to the content of 70 is displayed. At this stage, it is assumed that nothing is displayed in the tooth information column 7312, the single image 1 column 7322, the single image 2 column 7324, the single image 3 column 7326, or the single image 4 column 7328. Here, for example, according to the display of the tooth selection column 7311 by the operator, for example, “1” on the display is designated by the mouse, for example, “1” on the display is pressed, or the analysis input device 34, for example, when “1” is pressed by the keyboard, the analysis system 30 follows the above universal system, assuming that “1” in the dental formula selection field 7311 corresponds to, for example, the dental formula of “universal system”. Are displayed in the tooth information field 7312. At the same time, according to the tooth number of the universal system tooth, for example, the single tooth image cut out by deep learning is displayed in each column such as the single image 1 field 7322 according to the universal system tooth number. Note that the tooth type / tooth number display data table 70 shown in FIG. 7 shows a display state after the tooth type is designated.

  Next, a method for displaying only a specific tooth single image will be described. For example, when the operator designates “partial display” in the switching 1 field 73131 of the display switching field 7313 after the operator designates “9” in the tooth number list display, for example, as in the above-described tooth formula designation, as shown in FIG. It shall be displayed as shown in the tooth type / tooth number display / partial display data table 8A0. Here, “tooth number: 9 image cut out: LCU1” displayed in the tooth number portion display field 8A221 indicates information related to the image displayed in the tooth number single display 8A2211 (Central incisor). When the operator selects, for example, display addition 8A321, display enlargement 8A322, display rotation 8A323,... With the mouse, the display addition 8A321 performs partial display in the tooth number part display field 8A221. The display switching process to the tooth type / tooth number display data table 70 is performed. In the display enlargement 8A322, the partial display displayed in the tooth number part display field 8A221 is enlarged and reduced. In the display rotation 8A323, the tooth number part display is performed. It is assumed that processing such as rotating the partial display displayed in the column 8A221 is performed.

  Similarly, for example, after specifying each tooth number storage column of the tooth type / tooth number display data table 70, for example, specifying “composite display” in the display switching field, the tooth type / tooth number display / combination shown in FIG. The display data table 8B0 is displayed. In this case, if the tooth numbers “9” and “24” are specified and “composite display” is specified, the tooth number “9” is displayed as in the composite 1 display 8B2211 and the composite 2 display 8B2212 in the composite display field 8B221. It is assumed that the single tooth image of “24” is combined and displayed at the corresponding position of the tooth type.

  As described above, for example, after observing a single tooth image cut out by deep learning in various display forms, for example, an operator having a regular qualification has an error with respect to the number related to the tooth number by deep learning, for example. If it is determined, for example, from each tooth number in the tooth type / tooth number display data table 70, the single tooth image determined to be incorrect is designated by, for example, a mouse which is an input device of the analysis input device 34. Thereafter, the image of the single tooth is moved by moving to the tooth number column determined to be correct. At this time, for example, the display of “Central incisor” added to the tooth number and the single tooth image may be changed according to the changed tooth number. In addition, when a single tooth image has already been pasted in the tooth number field where the single tooth image has moved due to the change of the tooth number, the moved single tooth image is displayed, and the conventional single tooth image is displayed, for example, It may be displayed by moving to the correction management 2 column 722 or the like. Furthermore, for example, even when it is determined that the numbering cannot be performed by deep learning, for example, by cutting out a single tooth by deep learning, it may be displayed in the correction management 2 column 722 or the like as described above. Furthermore, for example, when the same tooth number is assigned to multiple tooth single images as a result of numbering by deep learning, for example, the most likely tooth single image is displayed in the corresponding tooth number column, otherwise For example, it may be displayed in the correction management 2 column 722 or the like, or a plurality of single tooth images may be displayed in the same tooth number column.

  The tooth type / tooth number display / partial display data table 8A0 shown in FIG. 8A and the tooth type / tooth number display / composite display data table 8B0 shown in FIG. 8B have been described. Information is stored in the data tables 8A0 and 8B0. The processing to be performed is based on the processing by the tooth type / tooth number display data table 70.

  Next, regarding the tooth number information list 90 that stores tooth number information including information obtained by correcting the tooth number by the operator, for example, with respect to the single tooth image information cut out by deep learning, for example, as described above, according to FIG. explain. The tooth number information list 90 includes a tooth number information list code column 91 for storing a cut image code and the like to be managed for tooth number, a tooth number information list management column 92 for storing management contents, and various information on the tooth number information list. , And the like, the tooth number information 1 column 9301, the tooth number information 2 column 9302, the tooth number information 3 column 9203, and so on. Further, the tooth number information list management column 92 includes a number column 921, an image code column 922, a change 1 column 923, a change 2 column 924, ..., and further, the tooth number information list management column 92 includes a change content column 9201, A change date column 9202,... The tooth number information list 90 is, for example, information calculated from related information by the analyzer processor 32 according to a program stored in the analyzer memory 31 or a tooth number information list setting displayed on the display of the analysis output device 33. According to a screen (not shown), for example, information input by an operator from the keyboard of the analysis input device 34 of the analysis system 30 is analyzed by the analysis device processor 32 according to a program stored in the analysis device memory 31. Shall be stored above. Further, the tooth number information list setting screen (not shown) has the same format as the tooth number information list 90. Furthermore, if necessary, a tooth number information list display screen (not shown) according to the contents of the tooth number information list 90 is displayed on, for example, the analysis output device 33 of the analysis system 30 and the management output device 23 of the management system 20. It shall be. The tooth number information list 90 may be input and stored in the memory on the management system 20 side. The information input and stored on the analysis system 30 side may be transmitted to the management system 20 side via the network 10, for example. The information input and stored on the management system 20 side may be transmitted to the analysis system 30 side. You can send it.

  Next, a procedure for storing information in each column of the tooth number information list 90 will be described with reference to FIG. First, for example, when the information “Yes:...” Is stored in the image cutout column 4302 of the image analysis management data table 40 and the image cutout process of a single tooth is performed, the tooth number information list 90 is created for each cutout process. As an example, if the image code to be cut out is SPX 16001 and the date when the tooth number information list is created is October 1, 2016, the tooth number information list code column 91 includes For example, information of “tooth number information list cut out image code: SPX 16001 date: 2016. 10.01” is stored. Next, in the tooth number information list management column 92, information such as “number” in the number column 921, “image code” in the image code column 922, and “change content” in the change content column 9201, for example. Are stored respectively. Further, the number field 921 of the tooth number information 1 column 9301 is “1”, for example, the tooth number information 2 field 9302 is “2”, for example, “2” is the number number of the tooth number information 3 column 9303, and so on. For example, information “3”,... Is sequentially stored in the column 921. Next, in the image code field 922 of the tooth number-specific information 1 field 9301, for example, an image relating to a single tooth is cut out by deep learning, and the tooth number recognition process is performed. When a tooth is located at the farthest upper right and is numbered with an image code of “upper right 8: RMU3”, for example, information “upper right 8: RMU3” is stored. Similarly, in the image code field 922 of the tooth number-specific information 2 field 9302, the tooth of the image subjected to the second tooth number recognition process is the tooth located in the second upper right part, and is called upper right 7: RMU2. When the image code is numbered, for example, the information “upper right 7: RMU2” is stored, and the image code is stored in the order of the tooth number recognition processing.

  As described above, the procedure for storing the information of the tooth number information list 90 regarding the result of the tooth number recognition performed by the tooth number recognition processing 51511 of the analysis processing details 60 shown in FIG. 6 has been described. An information storage procedure when the tooth number is corrected by processing (manual) 51513 will be described. First, for example, by the tooth numbering process by deep learning, the image of the ninth processed single tooth is identified as, for example, the upper left, the front tooth, and as a result, the tooth number-specific information 9 column of the tooth number information list 90 In the image code field 922 of 9309, information “upper left 1: LCU1” is stored as an image code, and further, for example, an image of the 24th processed single tooth is obtained by the tooth numbering processing by deep learning. For example, as a result of being identified as the lower left front tooth, the image code column 922 of the tooth number information 24 column 9324 of the tooth number information list 90 stores information “lower left 1: LCD1” as the image code. Sometimes, for example, on the tooth / tooth number display data table display screen (not shown) relating to the tooth / tooth number display data table 70 of FIG. If it is determined that there is an error in the number recognition, for example, the operator performs a process of exchanging both by operating the mouse or the like, and if the date is October 2, 2016, the teeth in the tooth number information list 90 In the change 1 column 923 of the number-specific information column 9309, information “lower left 1” is stored as the lower right front tooth, and in the change 1 column 923 of the tooth number-specific information column 9324, information “upper left 1” is stored. In the change date column 9202, the information “2016.10.02” is stored as the change date. However, when no information is stored as in the change 1 field 923 of the tooth number-specific information field 9301, the tooth number related to the single image of the designated tooth indicates that correction has not been performed. In addition, when there is a change by further modification of the tooth number, change information by the correction is sequentially stored in the corresponding change 2 column 924.

  In the above description, as the storage information of each column of the tooth number information list 90, for example, a character code by numbers is stored. However, for example, information such as a uniquely set code or a character code string may be stored. In addition, information such as a uniquely set code may be further added.

  Next, in the output 4 column 4308, for example, when “information on tooth number correction (manual) result feedback exists” is stored, information on manual correction stored in the tooth number information list 90 indicates the tooth number correction result. The feedback switching process 515131 feeds back to the learning DB 5152 as a resource for improving accuracy, and further learning of deep learning is performed. However, if the information “... Presence” is not stored in the output 4 column 4308, the manual correction result is not fed back to the learning DB 5152. This is because the results have already been obtained by deep learning, and if it is determined that no further learning is performed at this time will shorten the processing time, the information “... Yes” is stored. do not do. However, if it is determined that the information related to manual tooth number correction should be learned by deep learning, even after the fact, the information “... It is assumed that information regarding DB correction is sent at that time. In this case, whether to send all the changes or the latest changes may be selected by a separate selection process (not shown). Further, even when the image is not cut out in accordance with the information stored in the image cutout column 4302, as in the case where the information “... You may perform the process. In this case, for example, the image analysis management data table 40 may be provided with a column (not shown) for storing information instructing whether or not to perform feedback to the learning DB even when there is no cutout in the image cutout column 4302. Absent.

  The embodiment of the present invention has been described above. In the above description, the present invention is applied to a system including the management system 20 and the analysis system 30 connected via the network 10 shown in FIG. Although described as operating, it may be operated by a plurality of systems including a management system 20 and an analysis system 30 connected via the network 10, or the management system 20, analysis system 30, and related institution system. 202 may be operated as an integrated system. Alternatively, the functions of each system such as the management system 20, the analysis system 30, and the related institution system 202 in the present invention may be supplemented by another system, and as a result, the system may have the same function as the present invention.

  According to the present invention, it is possible to utilize image analysis results, particularly image analysis results in dentistry, for example for confirming the identity of a missing person when a disaster occurs. Furthermore, since the tooth number and the tooth state are effectively utilized, it contributes to the promotion and establishment of standardization of medical information, and can be applied to image analysis not only in dentistry but also in general medicine. In addition, for example, it is possible to cut out each part included in one image, and by combining this analysis and cut-out process, it can be used in a wide range of fields. Furthermore, in the process of assigning the tooth number to each tooth that has been subjected to image segmentation by deep learning, for example, if it is determined that it should be corrected manually, the correction can be performed manually. Whether or not to feed back the contents of correction by the learning DB of the corresponding deep learning, and even in the case of feeding back, it becomes possible to set the time, so that a more flexible image analysis system can be constructed.

DESCRIPTION OF SYMBOLS 10 ... Network 20 ... Management system 30 ... Analysis system 40 ... Image analysis management data table 50 ... Image processing summary list 60 ... Analysis processing details 70 ... Tooth type and tooth number Display data table 8A0 ・ ・ Tooth type ・ Tooth number display ・ Partial display data table 8B0 ・ ・ Tooth type ・ Tooth number display ・ Composite display data table 90 ・ ・ ・ Tooth number information

Claims (25)

In an image analysis system comprising a management system having at least a management apparatus server and a management apparatus database, and an analysis system having an analysis apparatus processor and an analysis apparatus database,
Means for inputting first information relating to an image type to be subjected to image analysis processing by the management device server and / or the analysis device processor;
Means for storing said first information as a first storage on said management device server and / or analysis device database by said management device server and management device database and / or by said analysis device processor and analysis device database;
Means for inputting second information relating to a tooth number assigned to the image according to the image type by the management device server and / or the analysis device processor;
Means for storing the second information as a second storage on the management device server and / or the analysis device database by the management device server and the management device database and / or by the analysis device processor and the analysis device database;
Means for inputting third information relating to a tooth state for each tooth number by the management device server and / or the analysis device processor;
Means for storing the third information as a third storage on the management device server and / or the analysis device database by the management device server and the management device database and / or by the analysis device processor and the analysis device database; An image analysis system characterized by comprising. The image analysis system according to claim 1, further comprising means for displaying at least one of the first, second, and third information in the management system and / or analysis system. In an image analysis system comprising a management system having at least a management apparatus server and a management apparatus database, and an analysis system having an analysis apparatus processor and an analysis apparatus database,
Means for inputting first information relating to an image type to be subjected to image analysis processing by the management device server and / or the analysis device processor and cutting out an image of a single tooth from the image according to the image type;
Means for storing said first information as a first storage on said management device server and / or analysis device database by said management device server and management device database and / or by said analysis device processor and analysis device database;
Means for cutting out a single tooth image from an image according to the image type by the management device server and / or the analysis device processor;
Means for inputting second information relating to a tooth number attached to the cut tooth single image by the management device server and / or the analysis device processor;
Means for storing the second information as a second storage on the management device server and / or the analysis device database by the management device server and the management device database and / or by the analysis device processor and the analysis device database;
Means for inputting third information relating to a tooth state for each tooth number by the management device server and / or the analysis device processor;
Means for storing the third information as a third storage on the management device server and / or the analysis device database by the management device server and the management device database and / or by the analysis device processor and the analysis device database; An image analysis system characterized by comprising. The image analysis system according to claim 3, further comprising means for displaying at least one of the first, second, and third information in the management system and / or analysis system. Means for inputting fourth information relating to manual change of information on the tooth number by the management device server and / or the analysis device processor;
Means for storing said fourth information as a fourth storage on said management device server and / or analysis device database by said management device server and management device database and / or by said analysis device processor and analysis device database;
Means for manually changing the tooth number by the fourth storage by the management device server and / or the analysis device processor;
Means for inputting fifth information related to a manual tooth number change by the management device server and / or the analysis device processor;
Means for storing the fifth information as a fifth storage on the management device server and / or the analysis device database by the management device server and the management device database and / or by the analysis device processor and the analysis device database; The image analysis system according to claim 1, further comprising: an image analysis system according to claim 1. 6. The image analysis system according to claim 5, further comprising means for displaying at least one of the fourth and fifth information in the management system and / or analysis system. Means for inputting sixth information relating to incorporating information relating to the tooth number changed manually by the management device server and / or the analysis device processor as a resource of image analysis processing;
Means for storing the sixth information as a sixth storage on the management device server and / or the analysis device database by the management device server and the management device database and / or by the analysis device processor and the analysis device database;
7. The image according to claim 5, further comprising means for incorporating the fifth information as a resource for image analysis processing by the sixth information by the management device server and / or the analysis device processor. Analysis system. 8. The image analysis system according to claim 7, further comprising means for displaying the sixth information in the management system and / or the analysis system. In an image analysis system comprising a management system having at least a management apparatus server and a management apparatus database, and an analysis system having an analysis apparatus processor and an analysis apparatus database,
First information relating to an image type to be subjected to image analysis processing by the management device server and / or the analysis device processor is input,
Storing the first information as a first storage on the management device server and / or analysis device database by the management device server and management device database and / or by the analysis device processor and analysis device database;
Input second information relating to the tooth number assigned to the image according to the image type by the management device server and / or the analysis device processor,
Storing the second information as a second storage on the management device server and / or the analysis device database by the management device server and the management device database and / or by the analysis device processor and the analysis device database;
Input third information related to the tooth condition for each tooth number by the management device server and / or the analysis device processor,
The third information is stored as a third storage on the management device server and / or the analysis device database by the management device server and the management device database and / or by the analysis device processor and the analysis device database. A control method for an image analysis system characterized by the above. The method according to claim 9, further comprising displaying at least one of the first, second, and third information in the management system and / or the analysis system. In an image analysis system comprising a management system having at least a management apparatus server and a management apparatus database, and an analysis system having an analysis apparatus processor and an analysis apparatus database,
Input first information relating to image types to be subjected to image analysis processing by the management device server and / or the analysis device processor and to cutting out a single tooth image from the image according to the image type,
Storing the first information as a first storage on the management device server and / or analysis device database by the management device server and management device database and / or by the analysis device processor and analysis device database;
A single tooth image is cut out from the image related to the image type by the management device server and / or the analysis device processor,
Input second information related to the tooth number attached to the cut tooth single image by the management device server and / or the analysis device processor,
Storing the second information as a second storage on the management device server and / or the analysis device database by the management device server and the management device database and / or by the analysis device processor and the analysis device database;
Input third information related to the tooth condition for each tooth number by the management device server and / or the analysis device processor,
The third information is stored as a third storage on the management device server and / or the analysis device database by the management device server and the management device database and / or by the analysis device processor and the analysis device database. A control method for an image analysis system characterized by the above. The method according to claim 11, further comprising displaying at least one of the first, second, and third information in the management system and / or the analysis system. Input fourth information related to manually changing information on the tooth number by the management device server and / or the analysis device processor,
Storing the fourth information as a fourth storage on the management device server and / or the analysis device database by the management device server and the management device database and / or by the analysis device processor and the analysis device database;
The tooth number is manually changed by the fourth storage by the management device server and / or the analysis device processor,
Input fifth information related to manual tooth number change by the management device server and / or the analysis device processor,
Storing the fifth information as a fifth storage on the management device server and / or the analysis device database by the management device server and the management device database and / or by the analysis device processor and the analysis device database. 13. The control method for an image analysis system according to claim 9, further comprising: The method according to claim 13, further comprising displaying at least one of the fourth and fifth information in the management system and / or the analysis system. Inputting sixth information relating to incorporating information relating to the tooth number changed manually by the management device server and / or the analysis device processor as a resource for image analysis processing;
Storing the sixth information as a sixth storage on the management device server and / or the analysis device database by the management device server and the management device database and / or by the analysis device processor and the analysis device database;
The image analysis according to claim 13 or 14, further comprising incorporating the fifth information as a resource for image analysis processing by the sixth information by the management device server and / or the analysis device processor. How to control the system. The method according to claim 15, further comprising displaying the sixth information in the management system and / or the analysis system. In an image analysis system comprising a management system having at least a management apparatus server and a management apparatus database, and an analysis system having an analysis apparatus processor and an analysis apparatus database,
Means for inputting first information relating to an image type to be subjected to image analysis processing by the management device server and / or the analysis device processor;
Means for storing said first information as a first storage on said management device server and / or analysis device database by said management device server and management device database and / or by said analysis device processor and analysis device database;
Means for inputting second information relating to a tooth number assigned to the image according to the image type by the management device server and / or the analysis device processor;
Means for storing the second information as a second storage on the management device server and / or the analysis device database by the management device server and the management device database and / or by the analysis device processor and the analysis device database;
Means for inputting third information relating to a tooth state for each tooth number by the management device server and / or the analysis device processor;
Means for storing the third information as a third storage on the management device server and / or the analysis device database by the management device server and the management device database and / or by the analysis device processor and the analysis device database. An image analysis system program characterized by being operated. 18. The image analysis system program according to claim 17, further causing the management system and / or the analysis system to operate as means for displaying at least one of the first, second, and third information. In an image analysis system comprising a management system having at least a management apparatus server and a management apparatus database, and an analysis system having an analysis apparatus processor and an analysis apparatus database,
Means for inputting first information relating to an image type to be subjected to image analysis processing by the management device server and / or the analysis device processor and cutting out an image of a single tooth from the image according to the image type;
Means for storing said first information as a first storage on said management device server and / or analysis device database by said management device server and management device database and / or by said analysis device processor and analysis device database;
Means for cutting out a single tooth image from an image according to the image type by the management device server and / or the analysis device processor;
Means for inputting second information relating to a tooth number attached to the cut tooth single image by the management device server and / or the analysis device processor;
Means for storing the second information as a second storage on the management device server and / or the analysis device database by the management device server and the management device database and / or by the analysis device processor and the analysis device database;
Means for inputting third information relating to a tooth state for each tooth number by the management device server and / or the analysis device processor;
Means for storing the third information as a third storage on the management device server and / or the analysis device database by the management device server and the management device database and / or by the analysis device processor and the analysis device database. An image analysis system program characterized by being operated. 20. The image analysis system program according to claim 19, further causing the management system and / or the analysis system to further operate as means for displaying at least one of the first, second, and third information. Means for inputting fourth information relating to manual change of information on the tooth number by the management device server and / or the analysis device processor;
Means for storing said fourth information as a fourth storage on said management device server and / or analysis device database by said management device server and management device database and / or by said analysis device processor and analysis device database;
Means for manually changing the tooth number by the fourth storage by the management device server and / or the analysis device processor;
Means for inputting fifth information related to a manual tooth number change by the management device server and / or the analysis device processor;
Means for storing the fifth information as a fifth storage on the management device server and / or the analysis device database by the management device server and the management device database and / or by the analysis device processor and the analysis device database. The program of the image analysis system according to any one of claims 17 to 20, wherein the program is further operated. The image analysis system program according to claim 21, further causing the management system and / or the analysis system to further operate as means for displaying at least one of the fourth and fifth information. Means for inputting sixth information relating to incorporating information relating to the tooth number changed manually by the management device server and / or the analysis device processor as a resource of image analysis processing;
Means for storing the sixth information as a sixth storage on the management device server and / or the analysis device database by the management device server and the management device database and / or by the analysis device processor and the analysis device database;
23. The image according to claim 21, wherein the management apparatus server and / or the analysis apparatus processor is further operated as means for incorporating the fifth information as a resource of image analysis processing by the sixth information. Analysis system program. 24. The image analysis system program according to claim 23, further causing the management system and / or the analysis system to operate as means for displaying the sixth information.   25. A recording medium on which the image analysis system program according to any one of claims 17 to 24 is mounted.

Images