JPH09251516A - Method for recognizing id information - Google Patents

Abstract

PROBLEM TO BE SOLVED: To adequately convert ID information as image information into the one as text data in an ID information recognizing method. SOLUTION: Tomographic image information as a video signal transmitted from a CT device 60 and magnetic resonance image information as the video signal transmitted from an MRI device 70 are made into digital information by a multi-formatter 30 and also ID information as image information is extracted from the information which is made into digital information. ID information being the same as ID information transmitted from the devices is previously stored in an ID terminal 20 as text data concerning a numeral which is easily compared and collated within the characters constituting the ID information and comparison and collation with the numerals of multiple pieces of ID information stored in the ID terminal 20 are executed.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method of recognizing ID information, and more particularly to a method of recognizing ID information as image information attached to a part of image information for identifying an image.

[0002]

2. Description of the Related Art In various fields, it has been known to photoelectrically read an image recorded on a photographic film, obtain image information, apply appropriate image processing to the image information, and reproduce and record the image information as a visible image. It is done in.

Further, radiation image information of a subject such as a human body is temporarily recorded on a sheet-like stimulable phosphor, and the stimulable phosphor sheet is scanned with excitation light such as laser light to generate stimulable luminescence. A radiation image recording / reproducing that photoelectrically reads the obtained stimulated emission light to obtain an image signal, and outputs a radiation image of the subject as a visible image to a recording material such as a photographic photosensitive material or a CRT based on the image signal. System (hereinafter, CR
(Also called a computed radiography device)
Has already been put to practical use. This system has the practical advantage of being able to record images over a very large radiation exposure area compared to conventional radiographic systems using silver halide photography.

[0004] In the system using the stimulable phosphor sheet as described above, regarding the image recorded on the stimulable phosphor sheet, the patient name, photographing (recording) date,
It is necessary to associate various kinds of information such as an imaging part and a date of birth with each stimulable phosphor sheet. Among these information, the patient name is a search key item when managing the patient's medical history information, medication information, etc. in a database, and can be used as alphabets, numbers, symbols, kanji, katakana, hiragana, etc. ID (identification) information consisting of the characters

In a CR system using a stimulable phosphor sheet, the ID information and the like are input from a predetermined ID information terminal.

[0006] The ID information input from the ID information terminal is added as search information for radiation image information obtained by the CR system and stored in a predetermined storage medium. In this case, the ID information is obtained by inputting by key operation of an ID information terminal or by reading information recorded on an ID card.
It is ID information as text data.

In recent years, CT (computed tomograph)
y: tomographic imaging, MRI (magnetic resonance)
image; magnetic resonance image capturing) With the spread of devices and the like, images of these other modalities are also stored using the storage medium of the CR system or a laser imager,
It is considered that a more effective diagnostic image can be obtained if processing such as reproduction and image processing can be performed.

However, since there is generally no unified standard for not only transmission / reception standards but also image information itself between these other systems such as the CT apparatus and the CR system, it is possible to use other systems. The image information needs to be transmitted as a general-purpose video signal.

Therefore, it is necessary for the receiving side to convert the video signal from the CT apparatus or the like into information that can be processed by the receiving side (CR system) by using, for example, a multi-formatter or the like.

[0010]

By the way, although a video signal as image information is transmitted from the other modality, the ID information or the like regarding the image information is not always transmitted separately. . Also, even if it is transmitted, it cannot be received unless the standards such as synchronization between the transmitting and receiving sides are unified as described above.

However, the image information of the other modality includes ID information and the like as a part of the image represented by this image information, that is, as image information.

Therefore, if the receiving side can appropriately convert the ID information as the image information included in the transmitted image information into the ID information as the text data, based on the obtained ID information. It is also possible to input / output image information by another modality to / from the database.

The present invention has been made in view of the above circumstances, and a method of recognizing ID information capable of accurately converting ID information recorded as a part of an image into ID information as text data. It is intended to provide.

[0014]

According to the ID information recognition method of the present invention, in order to identify and recognize one image information from other image information, ID information attached as image information to a part of each image information is identified. In the recognition method of reading and recognizing the read ID information as text data, the ID information as the image information and the ID information as a large number of different text data previously recorded in a predetermined database are provided as the above-mentioned ID. Among the character strings forming the information, only the predetermined type of characters that are relatively easy to recognize are compared and collated. As a result of this comparison and collation, the ID as the image information is the most ID among the ID information as the large number of text data. Recognizing ID information having a high coincidence rate with information as text data about ID information as its image information. It is an feature.

As described above, the ID information as the image information to be recognized as the text data is converted into the ID information.
Rather than comparing and collating all the characters or symbols that make up the information individually, only the characters that make up the ID information that are relatively easy to collate are targeted for collation, and serve as the basis for the collation. By limiting the ID information to a certain one that is previously recorded as text data in a predetermined database by another method, characters having a poor correct answer rate due to matching are excluded from the target of matching in advance and the matching rate through matching is improved. By reducing the number of comparisons and collations, the time required for this collation can be shortened.

Here, the predetermined type of character that is relatively easy to recognize means a number or an alphabet of a small number of types, but it is also possible to apply katakana or hiragana except for an extremely large number of kanji. .

The Kanji is used as a limited type of characters in the ID information (for example, "inside" in the character string "internal medicine", "outside" in surgery, etc.), and these are limited. It is also possible to make comparison and collation only for those having a dedicated mask pattern for collating types of Kanji.

The image information with ID information is CT
It may be a video signal directly sent from a device such as a device, an MRI device, or another CR device, or by reading an image recorded by using these devices and recorded on a recording medium such as a film. It may be obtained.

The ID information as text data is recorded in the database. For example, when the image information obtained by a certain CR device is recorded in the database, the ID information corresponding to a predetermined ID information input device is input. In a hospital, the ID information of the patient at the time of reception of an outpatient is input to a predetermined terminal connected to a database or a card or the like emits light. Good.

As described above, if the ID information recorded in the database at the time of reception of an outpatient is retained in the database only during the period when the patient stays in the hospital (during examination, examination, hospitalization, etc.). Of course, during the other period, the database may be transferred to another storage medium. In this case, since only a limited number of ID information including the patient to be examined by the CT device or the like is stored in the database, the target of comparison and collation with the ID information as image information sent from the CT device or the like. The number of ID information as the text data becomes extremely small, the ID information is not erroneously recognized as the ID information of the person who is not staying in the hospital, and the accuracy rate of recognition of the ID information is significantly improved. it can.

[0021]

According to the ID information recognition method of the present invention,
ID information as image information to be recognized as text data is not individually compared and collated for all the characters or symbols constituting the ID information.
Of the characters that make up the ID information, only the characters that are relatively easy to collate are targeted for comparison and collation, and the ID information that is the basis of the collation is recorded as text data in a predetermined database by another method. By limiting the number to a fixed number, characters with an unstable matching correct rate are excluded from the target of matching in advance to improve the matching rate by matching and reduce the number of comparison matching, thereby shortening the time required for this matching. be able to.

When performing information communication only via a video signal as described above, it is not necessary to transmit and receive the image information and the ID information in synchronization with each other, so that the load of changing existing software is small. There are also advantages.

[0023]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of an ID information recognition method of the present invention will be described below with reference to the drawings.

FIG. 1 is a schematic block diagram showing the configuration of an image information system including an ID information recognition method of the present invention. The illustrated image information system scans a stimulable phosphor sheet with excitation light such as laser light to generate stimulated emission light, and photoelectrically reads the obtained stimulated emission light to obtain image information on X-ray absorption. The CR device 10 described above that obtains
For the ID terminal 20 that inputs the ID information as text data in order to attach the corresponding ID information to the image information obtained by the device 10, and for the specific patient who has already obtained the image information by the CR device 10. , CT image 60 tomographic image information transmitted as a video signal and MRI apparatus 70
The magnetic resonance image information as a video signal transmitted by
Similarly to the image information obtained by the R device 10, the information is converted into a digital information to be made into a database, and the ID information as a part of the digitalized image information is extracted from the image information, and this is converted into a text. A multi-formatter 30 that performs various signal processings such as conversion to data, and image information and other information digitalized as described above in association with ID information as text data, such as patient name, date of birth, etc. Filing device to store
40, and a laser imager 50 that reproduces these digitalized image information as a visible image.

Here, the ID information is information for individual identification given to each patient, for example, Chinese characters representing a name,
It is a character string consisting of alphabets, numbers and other symbols.

The tomographic image information sent by the CT device 60 and the magnetic resonance image information sent by the MRI device 70 are each image information of the patient recorded by the function of each device, and ID information of the patient is image information. Is recorded as.

In the CT device 60 and the MRI device 70, based on the image information taken by the CR device 10,
Tests are often performed on patients who require more detailed tests.

Next, the operation of the image information system of this embodiment will be described.

First, an X-ray transmission image is taken by the CR device 10 for a patient having certain specific ID information. At the time of this radiography, an operator such as an X-ray radiographer looks at the patient's medical chart, etc.
Input to D terminal 20. This ID information is the CR device
It is stored in the ID terminal 20 as a history showing that the photographing by 10 was performed. Further, when the X-ray transmission image information obtained by the CR device 10 is output as a visible image to the X-ray film using the imager 50, this ID information is exposed to the film together with the image. Further, the ID information is added as search information when the X-ray transmission image information obtained by the CR device 10 is stored in the filing device 40.

Here, when the doctor sees the film recorded with the X-ray transmission image information output by the imager 50 and determines that a more precise examination is necessary, the doctor needs the patient. To inspect by a different CT device 60 or MRI device 70.

The CT device 60 or the MRI device 70 obtains tomographic image information or magnetic resonance image information of the patient. The ID information of the patient is recorded as image information in a part of each image information as described above. To be done. This ID
The information was entered into the device by the operator during inspection by these devices.

Since the CT device 60 and the MRI device 70 are usually designed and manufactured so that these devices are used alone, information communication between these devices and the image information system is performed by a video signal. Not in

Therefore, the CT device 60 or the MRI device 70
The ID information input to is not transmitted as it is to the image information system as text data.

Therefore, it is necessary for the image information system side to read the ID information from each image information as a video signal transmitted from these devices and convert it into text data.

Here, each image information is input to the multi-formatter 30 from the CT device 60 or the MRI device 70.

The multi-formatter 30 digitizes the tomographic image information and the magnetic resonance image information as the input video signals, and extracts the ID information as the image information from the digitized image information.

For the extraction of the ID information, it is examined beforehand in which position range in each image the ID information is recorded for each of the tomographic image represented by the tomographic image information and the magnetic resonance image represented by the magnetic resonance image information. Then, information that can be determined as a character within the position range may be extracted. Note that it is not necessary to specifically specify those characters at the extraction stage.

Next, the multi-formatter 30 compares and collates the extracted character string (ID information) with only the characters constituting the character string, which are judged to be numerical, with the numerical text data. And in this collation,
The ID information stored in the ID terminal 20 is used supplementarily.

That is, among the extracted character strings, only the numbers are compared and collated with the text data to determine which number of the text data the numbers represent. After making a determination for all the numbers of the characters that make up this character string, the number string thus determined is compared with each number string of a large number of ID information accumulated in the ID terminal 20.

That is, since the ID information as the image information sent from the CT device 60 or the MRI device 70 is often already stored in the ID terminal 20, any of the stored numeric strings of ID information can be used. Numerical strings that do not match mean that there is a high probability that there is an error in comparison and matching.

If any one of the ID information stored in the ID terminal 20 matches the numeral string of the ID information, the ID having the matched numeral string.
Information (text data) is attached to digitalized tomographic image information or the like as search information.

The tomographic image information and the like to which the ID information (text data) is attached is stored in the filing device 40, and can be subjected to subsequent image processing, reproduction on a film or monitor screen, and other various processing. .

On the other hand, if none of the ID information stored in the ID terminal 20 matches the number string of the ID information, one or more than one number having the number string closest to "match" is selected. The ID information is selected as a candidate, and an alphabet (or an alphabet string composed of a plurality of alphabets) is extracted from the characters forming the ID information for the selected ID information.

Alphabets (columns) are also extracted from the ID information (image information) that is the object of comparison and collation, and this alphabet (column) is compared and collated with the alphabet (column) of the selected ID information. One piece of ID information closest to "match" is selected. Then, the selected ID information (text data) is attached to digitalized tomographic image information or the like as search information,
The tomographic image information and the like to which the ID information is added is stored in the filing device 40 and can be subjected to subsequent image processing, reproduction on a film or monitor screen, and other various processing.

It should be noted that when one ID information close to "match" cannot be selected even if the object of comparison and collation is expanded to the alphabet. Only when it is unavoidable to make a warning sound, the operator can read the ID information recorded in the tomographic image information with the naked eye and manually input the ID information from the ID terminal 20. Good.

Even when the number string of two or more ID information among the large number of ID information accumulated in the ID terminal 20 is matched, the comparison and collation up to the alphabet (column) is performed as described above. The target is expanded to select one piece of ID information whose number string and alphabet (column) are “perfect match” or closest to “match”. Then, the selected ID information (text data) is attached to digitalized tomographic image information or the like as search information, and the tomographic image information or the like with the ID information is attached to a filing device.
It is stored in 40 and can be a target for later image processing, reproduction on a film or a monitor screen, and other various processing.

As described above, when there are two or more candidates for "match" in the collation using only the numbers, the collation is performed for the characters including the alphabet in addition to the numbers, and still one ID information cannot be specified. In that case, the collation is performed for characters including symbols.

Note that another CR device may be provided in parallel with the CT device 60 and the like together with the CT device 60 and the MRI device 70.

The ID terminal 20 may be connected to information received at the hospital (for example, information that has been accepted as an outpatient) both online and offline. As information received at the hospital,
For example, a medical accounting system or the like can be applied.

[Brief description of drawings]

FIG. 1 is a schematic block diagram showing the configuration of an image information system including an ID information recognition method of the present invention.

[Explanation of symbols]

 10 Computed radiography (CR) device 20 ID terminal 30 Multi-formatter 40 Filing device 50 Laser imager 60 CT device 70 MRI device

Claims (3)

[Claims] 1. In order to identify and recognize one image information from other image information, ID information added as image information to a part of each image information is read, and the read ID information is recognized as text data. In the recognition method, the ID information as the image information and the ID information as a large number of different text data recorded in a predetermined database in advance are relatively easy to recognize among the character strings forming the ID information. Only the characters of a predetermined type are compared and collated, and as a result of the comparison and collation, the ID information having the highest matching rate with the ID information as the image information among the ID information as the plurality of text data is determined as the image information. A method for recognizing ID information, characterized in that the ID information is recognized as text data. 2. The image information having ID information attached to the part is a video signal transmitted from at least one of a tomographic image capturing apparatus, a magnetic resonance image capturing apparatus, and a radiation image capturing apparatus. The method for recognizing ID information according to claim 1, characterized in that. 3. The ID information according to claim 1, wherein the ID information as the text data is recorded in the database by inputting the ID information into a predetermined ID information input device. Recognition method.