AUTOMATIC TECHNIQUE OF ACQUIRING DIGITAL ELECTRONIC MEDICAL IMAGE

Introduction and Classification

Medical data in DICOM image format, graphic and text format (alphanumeric) are usually generated from medical devices such as CT scanners. MRI, DR, functional diagnostic equipment and laboratory machines. The patient’s data, after being measured from the patient, is usually temporarily stored in the memory of the device in digital form. In addition, most of these devices’ manufacturers often have built-in digital communication standards to connect with peripheral devices. For current DICOM image devices, most of the output of the device is designed w'ith DICOM standard for digital image data communication with RJ45 connection. For devices creating data in graphical or alphanumeric format, the output of the device has standard RS232 digital communication. This section presents methods for automatically collecting digital electronic medical data from digital communication outputs on medical devices [10].

Automated Technique of Digital Image Acquisition According to DICOM Standard

DICOM is a world standard that sets the rules for formatting and exchanging digital imaging data and patient-related information in health. Digital image data generated from different devices following DICOM standards will create a common ‘language’, allowing the process of communicating images and related health information between applications to be done easily. DICOM is an open standard, and it exists through the procedures of the DICOM standards development committee. DICOM is based on a realistic model with patient information, imaging data and reporting data. These models are called entity relationship models (E-R). The approach to developing data structures based on an entity relational model is called object-oriented design (OOD). Objects are entities that are defined by the model. The characteristics of each entity are called attributes. DICOM calls model-based objects information objects (IO). Models and attribute tables define them as information object definitions (IODs). The entities in the model are abstract. If an actual value replaces a property, that entity is called an instance [13].

Process flow diagram automatically captures digital image data according to DICOM standards

FIGURE 10.8 Process flow diagram automatically captures digital image data according to DICOM standards.

Figure 10.8 shows a process diagram of automatic acquisition of digital image data according to DICOM standard. Digital data at the output of the medical imaging device will be sent to the computer via the RJ45 connector. Through the services in DICOM standard, the computer processing software will receive and store image data into the database [9, 10].

  • Building software to process image data according to DICOM standard on the computer: The function of the data processing software on the computer is to automatically receive digital data from imaging devices sent through the RJ45 connection port according to DICOM standard. Through services in the DICOM standard, this digital image data is re-created and stored into a database. It also provides display and image processing features according to DICOM standards in electronic medical model applications. With the DICOM Objects tool, image data processing according to the DICOM standard is performed by the following algorithms and processing functions:
  • Building algorithm to set parameters for connecting with imaging equipment. The library in DICOM Objects supports the DICOM standard C-STORE service, Therefore, in order to send images from an imaging device to an external device via the DICOM standard, a setup window must first be built to enter the IP address and port of the service class provider (SCP) image, set up application entity title (AET) of SCP and service class user (SCU). The algorithm of setting parameters connecting with imaging devices is implemented in the dot.NET environment as follows:

Dim ini As Dicomlmage;

Dim res As Integer: im = Viewer.Images:

res = im.SendilP address and image port of SCP. AET of SCU. AET of SCP):

Building algorithm to receive DICOM images on computers. All DICOM Viewer objects in DICOM Objects have the ability to ‘listen’ to the signal to send images from imaging devices. When an image signal is sent, DICOM Viewer will receive the data transmitted from SCP. The image receiving algorithm is implemented in the dot.NET environment as follows:

Dim server As New DicomServer: server = New DicomServer: server.Lislenfthe listening port):

DicomServer.

Building algorithm to save DICOM images into the database. Once DICOM images have been sent from the device to the computer, the DICOM image storage needs to be performed to write the data received to the storage drive, simultaneously reading file information and writing to the database for later retrieval and retrieval of data. The process of processing and recording files is performed as follows. Create a path that is expected to write the image file to the database. Here the image file name is taken using the labelled component (0008,0018). Extract the information in each image file and then push the information and the path of image recording into the DICOM database for searching and retrieving images. Use the M-WRITE service in DICOM standard to create new files in a file set and assign them a file index or update files if the existing file set already exists. The database for storing DICOM file information includes the components described in Table 10.1.

TABLE 10.1

Components Used to Store DICOM Image Information

No

Data Component Label

Data Component Description

1

(0008.0020)

Study Date

2

(0008.0021)

Series Date

3

(0008.0030)

Study Time

4

(0008.0031)

Series Time

5

(0008.0060)

Modality

6

(0008.1030)

Study Description

7

(0008.103E)

Series Description

8

(0008.1048)

Physician of Record

9

(0010.0010)

Patient’s Name

10

(0010.0020)

Patient’s ID

11

(0010.0030)

Patient’s Birth Date

12

(0010.0040)

Patient’s Sex

13

(0020.000D)

Study Instance UID

14

0020.000E)

Series Instance UID

15

(0020.0010)

Study ID

16

(0020.0011)

Series Number

17

(0020.0013)

Instance Number

The M-WRITE service in DICOMObjects will support the implementation of writing DICOM image data to the drive according to the previously created path. The algorithm for recording image data on dot. NET environment is as follows:

ImageReceived.Write(filename, “1.2.840.10008.1.2.1 "):

Where: «filename » is the index created, “1.2.840.10008.1.2.1” is the conversion syntax for existing VR and little-endian formats. Can be replaced by other conversion syntax. DICOMObjects supports all conversion syntax according to the DICOM standard defined.

Building algorithm to display DICOM images on computers. The display of image data and related information in the DICOM file is the process of reading the data components in the DICOM file through decoding and displaying it on the screen. The diagram depicting the display of images and related information in the DICOM file is done according to the steps described in Figure 10.9.

In the DICOMObjects tool, there is a DICOM Viewer object that allows decoding data according to the image encoding standards in DICOM to display. Functions that control the display of image data include:

DicomViewer.lmages.Read (filename of images in local directory): DicomViewer.lmages.Add (e.lnstanceis the image received or retrieved from the server):

Display multiple images at the same time:

DicomViewer.Multicolums=n; (n is the number of columns); DicomViewer.Multirows=m; (m is the number of rows).

Building algorithm to display related information in DICOM file. To display related information, DICOMObjects supports DICOM Labels object. So

Diagram depicting the process of displaying image data and related information in a DICOM file on a computer

FIGURE 10.9 Diagram depicting the process of displaying image data and related information in a DICOM file on a computer.

TABLE 10.2

Information Related to DICOM Images

No

Related Information

Data Component

1

Model Name

(0008.1090)

2

Study ID

(0020.0010)

3

Series Description

(0008.103E)

4

Current Series

(0020.0011)

5

Current image/total number of images

(0020.0013)/(0020.1002)

6

Number of Rows

(0028.0010)

7

Number of Columns

(0028.0011)

8

Patient's Name

(0010.0010)

9

Name of Institute

(0008.0080)

10

Patient's ID

(0010.0020)

11

Study Date

(0008.0022)

12

Series Time

(0008.0032)

13

Image Size

(0018.1100)

14

Thickness

(0018.0050)

15

Brightness

(0028.1050)

16

Contrast

(0028.1051)

17

Specifications related to the study process

(0018.0091), (0018.0080). (0018.0081), (0018.1250)

read the information in the image and assign it to the DICOM Labels object.

In fact, information related to DICOM images is often displayed with image data including the information described in Table 10.2 as follows.

To access this component, DICOMViewer supports the following function:

DicomViewer.images(group, element);

InfoLabel=New DicomLabelf);

InfoLabel.Left= m;

InfoLabel. Height=n;

InfoLabel.Text=DicomViewer.images(group, element);

DicomViewer. Labels. Add(InfoLabel);

The process of automatically collecting digital image data according to DICOM standard has been researched and built by the algorithm presented above. Figure 10.10 depicts the results of the automated process of acquiring and displaying digital image data in accordance with the DICOM standard on the built-in processing software by connecting to Airis Mate-Hitachi MRI system.

Evaluation of the Quality of Image Data after Acquisition

Like the method of evaluating the quality of image data received from imaging devices with the analog output interface, the DICOM image data collected on the computer and

The results of the process of automatically acquiring and displaying digital image data in accordance with DICOM standard by connecting with Air is Mate-Hitachi MRI system,

FIGURE 10.10 The results of the process of automatically acquiring and displaying digital image data in accordance with DICOM standard by connecting with Air is Mate-Hitachi MRI system, (a) DICOM data retrieval result from data base and (b) display DICOM image data and related information on processing software.

the processing software is grayscale. The author used the results of the above software construction, proceeded to connect and receive DICOM medical imaging data on three imaging devices that follow the current popular DICOM standard, including (1) MRI device from Air is Mate-Hitachi; (2) DR device from RADspeed Safire-Shimadzu and (3) CT scanner device from Pronto XE-Hitachi. Each device will randomly receive 20 DICOM images of 20 different patients. Also, the original DICOM image data is from the temporary memory of each device. Parameters received for grayscale DICOM images from these devices include the following: The Air is Mate device has a 256 x 256 pixels image size and 16 bits encoding. RADspeed Safire device has image dimensions of 2544 x 3056 pixels, 12 bits encoding. The Pronto XE device has an image size of 512 x 512 pixels and 13 bits encoding. The trend of fluctuating PSNR values by aggregated results is shown in Graph 10.3 [11, 12].

From the calculation results, the chart shows the trend of changing PSNR values for data collection and evaluation of DICOM medical data quality as follows: For Air is Mate MRI equipment, the value of PSNR varies from 57.26 dB to 61.24 dB. the average is 59.36 dB. For Pronto XE CT scanner, the value of PSNR varies from 53.16 dB to 57.27 dB, the average is 55.48 dB. For RADspeed Safire DR equipment, the value of PSNR varies from 44.25 dB to 47.26 dB, the average is 45.83 dB. Thus, compared to the assessment criteria of image quality data processing in health with a good level. PSNR ranges from 35 to 40 dB for DR images and from 45 to 50 dB for CT and MRI images [11,12]; the automatic method of obtaining digital image data in DICOM format is to meet the requirements.

 
Source
< Prev   CONTENTS   Source   Next >