• Share
  • Email
  • Embed
  • Like
  • Save
  • Private Content
AIBE 68
 

AIBE 68

on

  • 689 views

 

Statistics

Views

Total Views
689
Views on SlideShare
688
Embed Views
1

Actions

Likes
1
Downloads
5
Comments
0

1 Embed 1

http://www.slideshare.net 1

Accessibility

Categories

Upload Details

Uploaded via as Adobe PDF

Usage Rights

© All Rights Reserved

Report content

Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
  • Full Name Full Name Comment goes here.
    Are you sure you want to
    Your message goes here
    Processing…
Post Comment
Edit your comment

    AIBE 68 AIBE 68 Document Transcript

    • NEW ASPECTS of APPLIED INFORMATICS, BIOMEDICAL ELECTRONICS & INFORMATICS and COMMUNICATIONS A novel Standardized Uptake Value (SUV) calculation of PET DICOM files using MATLAB. PAWITRA MASA-AH and SOMPHOB SOONGSATHITANON* Nuclear Medicine Division, Department of Radiology, Faculty of Medicine Siriraj Hospital, Mahidol University, 2 Prannok Rd., Bangkoknoi, Bangkok, 10700 THAILAND *Corresponding author: sissst@mahidol.ac.thAbstract: - This paper presents a novel SUV calculation scheme for PET DICOM files usingMATLAB and the results are compared with the SUV taken from the well-known applicationsoftware from GE healthcare. The performance of the scheme is evaluated by using the DICOM filestaken from the widely used standard PET phantom scan. The results show that the SUV calculationscheme for PET DICOM file using MATLAB is comparable with the SUV taken from the GEhealthcare application software. The strength of this scheme is that the interchangeability of theDICOM files can be done conveniently without the special application software from any vendors.Key-Words: - Positron Emission Tomography combined Computed tomography (PET/CT),Standardized Uptake Value (SUV), The Digital Imaging and Communications in Medicine(DICOM)1 Introduction A DICOM file is divided into two major sections. there are part of the DICOM imageIn Nuclear medicine, Positron Emission and the data header (DICOM elements orTomography (PET) scanner plays an important Metadata). DICOM image (Fig 1) shows arole in both diagnosis and staging of cancers. more detailed list of the DICOM header asThe hybrid system of a PET scanner combined displayed by the software. The data header orwith CT scanner (PET/CT) has gained DICOM elements (Fig 1) required depends onpopularity in the oncological community since the image type.its commercial introduction to the market in In this research we use MATLAB as aearly 2001. DICOM viewer and also for analysis. These For the F18-FDG uptake evaluation of are the reviewed DICOM file by MATLAB.the tumor metabolism quantitatively. SUV(Standardized Uptake Value) is the one thatquite often used. It is most commonly usedsemi-quantitative parameter utilized for (a)analyzing FDG-PET images in routine clinicalpractice. It came to be used as a tool tosupplement visual interpretation. DICOM file was a file format that (b)developed by the DICOM StandardsCommittee whose members are also partlymembers of NEMA. The Strength of DICOMfiles is the exchangeability between twoentities that are capable of receiving image andpatient data. DICOM also enables theintegration of scanners, servers, workstations, Fig. 1. The DICOM file is reviewed byprinters, and network hardware from multiple MATLAB (a.) Image part (b.) Metadatamanufacturers into a picture archiving and partcommunication system (PACS).ISSN: 1792-460X 413 ISBN: 978-960-474-216-5
    • NEW ASPECTS of APPLIED INFORMATICS, BIOMEDICAL ELECTRONICS & INFORMATICS and COMMUNICATIONS2 Study Framework varies for every image slice. So the tissueThis is a block diagram in our study that show activity is calculated using formula 1.the process, After a set of DICOM filesthroughout our created scheme, the scheme (1)yielded the SUVmaximum and SUVmean Where m is rescale Slope (which is different in each image slice), SV is the stored value (Pixel intensity Value), b is rescale intercept (for PET scan is always Zero) and U is units of value after conversion (for our study is Bq/ml, Shown in Unit tag) The conversion formula above is the fundamental defined by NEMA, however the Fig. 2. The Study Framework formula maybe slightly different for each vendor, In this study we used GE DICOM file. So the method of calculation is based on GE3 Calculation of Standardized conformance statement [8]Uptake Value (SUV). SUVbody weight (kg / ml ) = Activity Concentration in ROI ( Bq / ml ) (2)SUV (Standardized Uptake Value) is the one ⎛ Injected Dose( Bq ) ⎞ ⎜ ⎟that quite often used. It is most commonly ⎜ body weight ⎟ ⎝ ( kg ) ⎠used semi-quantitative parameter utilized foranalyzing FDG-PET images in routine clinicalpractice. It came to be used as a tool to ⎛ ⎜ Pixel Value ⎞ ⎜ Im age Scale Factor ⎟ ⎟supplement visual interpretation. The SUV is ⎝ ⎠ SUV body weight ( kg / ml ) = ⎛ actual activity ⎞ (3)also used to distinguish between malignant and ⎜ ⎜ body weight ⎟ ( ⎟ × dose calibratio n factor × 10 6 ) ⎝ ⎠benign tumor. The cut-off value for malignantlesion is 2.5 and It have largely been proven tobe an oversimplification. Where Pixel Value is Pixel Intensity Value in In this study, the DICOM files are the Region of Interest (ROI),main source to analyze and calculate SUV.The Component of the DICOM file can bedivided into 2 main parts which are the image , Actual activity is Injected activity at the timeand metadata. The image is the part that of scan, Body Weight is Patient Body Weightdisplays the information by image and the (kg) and Dose Calibration factor must bemetadata is the part that keep the detail of the multiplied by 10-6 for MBq/ml to Bq/mldata such as dose of injection, type of scan, conversion.patient weight, patient name, etc. We calculate the Standardized Uptake For SUV is practically in the definedValue (SUV) and the brief meaning of SUV is Volume Of Interest (VOI). We usedthe Differential Uptake Value, Dose Uptake MATLAB as a tool in the process. The SUVRatio, or Dose Absorption Ratio, For this formula(2) have shown the requirement of datastudy we used PET DICOM file that has been which are from image part including metadatathrough the attenuation correction. part. Here we used MATLAB to extract these The Uptake Value is represented by Pixel data.intensity value in the image and the GEDICOM file. It collects the data in 16 bit so The steps of the SUV calculation scheme arethat the pixel intensity value is between 0- summarized as follow :32767.Therefore, the first process of the SUV Step 1: Extract Pixel Intensity Value incalculation is the conversion of the pixel Defined ROI (region of interest).intensity value to the activity concentration. The purpose of extraction is to noteThe related attributed tag* for the conversion the obtained pixel intensity value foris Rescale Slope tag, and Rescale Intercept tag the next step of calculation, So we(there are from metadata part). These tags extract Mean and Maximum pixelISSN: 1792-460X 414 ISBN: 978-960-474-216-5
    • NEW ASPECTS of APPLIED INFORMATICS, BIOMEDICAL ELECTRONICS & INFORMATICS and COMMUNICATIONS intensity because we need them to Fig. 2. Integral ROIs to VOI calculate the SUVmean and SUVmaxStep 2: Convert Intensity Pixel Value toactivity concentration (Bq). After we extracted the desired value, the conversion of Intensity Pixel Value to Activity Concentration (Bq) is the next step. The data in this conversion based on some of the detail part (metadata). See formula (2) then become Where The related tag in this part are Rescale Slope, Rescale Intercept and Units. Only Rescale slope is individual in each slice.Step 3: Calculate SUV in each slice. Calculate SUV by formula (3) which the related tag in this part are Total Dose (0018,1074), Series Date (0008,0021), Series Time (0008,0031), Radiophamaceutical Start Time (0018,1072), Radiophamaceutical Half Life (0018,1075) and Patient Weight (0010,1010)Step 4: Integral SUV of ROI to VOI In practical , SUV calculation is analyze in the Volume of Interest (VOI). So In this study we calculated SUV in each slice, summarized them Fig. 3. The SUV calculation Scheme which is depending on number of slices and projected the region of interest into every slices in the 3 Scheme Testing. volume. Mean that every slices under This SUV calculation Scheme is operated on the volume has the same size and the PC with installing MATLAB. In order to position. evaluate the performance of the scheme, it is compared to the SUV that obtained from the Well-known application software from GE healthcare (Xeleris workstation) Xeleris and MATLAB all allow interpretation of ROI analysis. Three circular ROIs were drawn surrounding the region of FDG uptake. see the demonstration in the fig. 4.ISSN: 1792-460X 415 ISBN: 978-960-474-216-5
    • NEW ASPECTS of APPLIED INFORMATICS, BIOMEDICAL ELECTRONICS & INFORMATICS and COMMUNICATIONS [2] Gustavo S.P. Meirelles, Peter kijewski, Timothy, “Correlation of PET/CT standardized Uptake Value measurements between Dedicated workstations and PACS-integrated workstation system” Journal of Digital Imaging, Vol 20, NO.3(September)2007, p.307-313 [3] S.Chicklore, P.Jaye, S.Vijayanatham, Fig. 4. Example of the 3 sizes ROI drawn D.Gallacher, “Comparing diagnostic on Phantom transaxial images accuracy of digital images viewed on a Picture Archiving and CommunicationThe maximum SUV was recorded and mean System(PACS) diagnostic workstation andSUVs of the malignant lesion were calculated on a Personal Computer(PC)-Basedand recorded by the Xeleris® workstation and imaging system” St Thomas’ Hospital,MATLAB. And the Xeleris and MATLAB London, UKwere analyzed together for further studying the [4] ACR-NEMA, “Digital Imaging andcorrelation of both system. Communication in Medicine (DICOM) Data were analyzed using SPSS software. Positron emission tomography imageCorrelation of the SUV measurement between objects Supplement 12 ”the dedicated Xeleris workstations and [5] Aparna Kanakatte, Jayavardhana Gubbi,MATLAB was measured with Pearson Nallasamy Mani, Tomas Kron, Davidcorrelation coefficient. The comparison was Binns, “A Pilot Study of Automatic Lungperform with 3 sized-ROI in 45 phantom slice. Tumor Segmentation from Position Emission Tomography Images using Standard Uptake Values”, Proceedings of4 Results the 2007 IEEE symposium onThe result shown that the Correlation between Computational Intelligence in Image andXeleris Workstation and MATLAB was very Signal Processing (CIISP 2007), p 363-good for the SUVmaximum and SUVmean. This is 368reassuring, Because it mean that the proposed [6] M.Teras, T.Tovanen, J.J.Johnasson,of the SUV calculation scheme can interpret J.J.Williams, J.Knunti, “Performance ofPET/CT images even though the worker is not the new generation of whole-body PET/CTon the workstation site. scanners: Discovery STE and Discovery VCT”, Eur journal Nuclear Molecular Imaging (2007)34, p.1683-16925 Conclusion [7] Mark. T Madsen, “Selecting New PET/CTThese Result indicate that the SUVs from System”, Technology talk, 2004 AmericanMATLAB analysis could be used interchange College of Ragiology, p.356-358with PET/CT Xeleris® workstation for [8] GE Medical System, “Technicalinterpretation of PET/CT image without losing Publication Advance TM 5.1 Conformancethe capacity to accurately measure the SUV. Statement for DICOM V3.0”, GE Healthcare, p. 3_17 – 3_21References: [9] http://qibawiki.rsna.org/[1] Joseph A. Thie, PhD, “Understanding the [10] GE Medical System, “DiscoverySTE Standardized Uptake Value, Its Methods, Quick Manual” , GE Healthcare MISM And Implication for Usage” The journal of Tip application, p. 28-38 Nuclear Medicine, Vo. 45, No.9 p. 1431- 1434ISSN: 1792-460X 416 ISBN: 978-960-474-216-5