Automated Image Analysis Software for Quality Assurance of a Radiotherapy CT Simulator

Transcription

1 Automated Image Analysis Software for Quality Assurance of a Radiotherapy CT Simulator Andrew J Reilly Imaging Physicist Oncology Physics Edinburgh Cancer Centre Western General Hospital EDINBURGH EH4 2XU Phone: Fax: andrew.reilly@luht.scot.nhs.uk Web:

2 Overview Radiotherapy imaging RT Imaging QA: problems and solution Describe features of auto analysis software Demonstrate application to CT-Sim and Sim-CT Outline experience to date

3 Imaging Modalities for RT Common Simulator (fluoroscopy) CT-simulator Digitally Reconstructed Radiographs (DRRs) Simulator-CT (single slice and cone-beam) Electronic Portal Imaging Devices (EPIDs) Emerging Ultrasound MRI PET On treatment cone-beam CT and kv radiography Integrated System

4 RT Imaging QA: Essential Tests Geometric Accuracy in 3D In and out of image plane (pixel size, couch travel) Mechanical alignments Laser alignment Image quality Sufficient for purpose? Consistent over time Accurate physical information CT number / HU calibration -> electron density Testing of overall system Geometrical co-registration Transfer of image data

5 The Problems Different tests are specified for different modalities Range of equivalent test objects Most tests are only semi-quantitative Operator dependency Frequent (daily/fortnightly) comprehensive testing is required BUT most tests are time-consuming Some imaging equipment performs too well! Difficult to test integrated system.

6 The Solution Develop single, uniform approach for all RT imaging modalities + display devices, film processors, etc. Robust, fully objective and quantitative Analysis performed by computer Results automatically stored in database for trend analysis, etc.

7 The Approach 1. Develop Appropriate Phantom 2. Acquire Image of Phantom Signal s 1 s 2 Signal s 1 s 2 SNR in = s 1 / s 2 SNR out = s 1 / s 2 2 SNR out = SNRin DQE( f )

8 Determining the DQE Modulation Transfer Function (Phantom) DQE ( f ) = K D MTF 2 NPS Dose and acquisition setting dependent. Noise Power Spectrum (Phantom)

9

10 Additional Collimators Varian Ximatron EX Sim-CT

11 Varian Performance Phantom A A A A WATER 1 2 R L INNER INNER BONE BONE MTF WATER LUNG R L R R 3 θ L L CORT BONE MTF AIR θ P P P P

12 Varian Uniformity Phantoms 44 cm 34 cm Polyurethane Casting HU -580

13 Geometry: Phantom Alignment Detect phantom edge Threshold at 580 Trace edges and choose largest contour Calculate COM Compare against CT zero position

14 Geometry: Pixel Size Measure distance between holes Use centre of phantom A and expected pixel size to identify seek area Local minimum is centre of hole R θ L P

15 Hounsfield Unit Calibration Baseline Values Measured During Commissioning ICRU 42 Ax, 80kV, 150mA Ax, 80kV, 300mA Ax, 120kV, 150mA Ax, 120kV, 300mA Ax, 140kV, 150mA Ax, 140kV, 250mA 1000 CT Number Electron Density Rel to Water

16 Hounsfield Unit Calibration A WATER R SOFT BONE MTF LUNG L HARD BONE AIR P

17 Modulation Transfer Function Calculate from impulse object ( f ) FT{ PSF( x) } MTF = OSF ( x) = PSF( x) DSF( x) Finite size (DSF) { OSF( x) } = FT{ PSF( x) } FT{ DSF( x) } FT ( f ) MTF = FT FT { OSF( x) } { DSF( x) }

18 Calculation from Impulse Object Object Spread Function (From ALL pixels in ROI)

19

20 Uniformity Phantom Analysis Define Useful FOV (UFOV) as 90% FOV Calculate: std dev Coefficien t of Variation, CoV = mean pmax mean Integral Uniformity, U + = mean pmin mean Integral Uniformity, U = mean pmax Differential Uniformity,U d = mean centre periphery Uniformity Index, U CT = 1000

21 Uniformity Phantom Analysis

22 Uniformity Profiles CT Sim: 50 cm FOV Sim-CT Urethane Norm Air Norm

23 Noise Power Spectrum Region of Interest from Uniformity Phantom Remove DC component (subtract mean value) Perform 2D FFT NPS ( u, v) = Re ( ) 2 u, v + Im( u, v) area 2 Separation of stochastic noise NPS s = n NPS NPS n n ROI n

24 NPS Example 100 images of Uniformity Phantom, 50 cm FOV

25

26 Production of DRRs Ray trace from virtual source of x-rays through stack of CT slices and model attenuation of beam. X-ray source SAD 100 cm isocentre Imaging Plane Reference: Milickovic et al, Physics in Medicine and Biology (2000) 45:10; Projected back to isocentre

27 DRR Production Example CT Slices 3D array of voxels DRR

28 Edinburgh DRR Phantom

29 Software Demo

30 Experience & Conclusions New approach appears complicated, but Significantly faster than previous methods More robust, fully objective and quantitative Greater confidence in results New ability to follow trends Need to finalise DRR phantom Expand to include other RT imaging modalities