CT 成像原理介紹History of Computed Tomography  Early 1970s Concept of X-ray Attenuation A n X-ray beam passing through the body is attenuated (loses its energy) by : Absorption Scattering  CT Generations & Design “Generation”is used to label CT tube-detector designs Slip-ring Technology Power is transmitted through parallel sets of conductive rings instead of electrical cables  Continuous Gantry Rotation  Prerequisite for Spiral CT What is Spiral Scan? -- just 4“C”C ontinuously rotating tube/detector system Continuously generating X-ray Continuously table feed Continuously data acquisition Scan Parameters X-ray Tube Voltage (kVp) X-ray Tube Current (mA) Scan Time (s) Slice thickness or Collimation (mm) Table Speed & Pitch  T able Speed is defined as distance traveled in mm per 360o rotation Slice Profile (SP) Effective slice thickness of an image  S lice Profile Comparison S mallest Possible Effective Slice Thickness Peripheral vein eg. antecubital vein 19-20 gauge needle or IV catheter T ailoring Scan & Injection Protocols T est Bolus Procedure Dynamic Evaluation Time-density curve  Scan Delay  Peak Enhancement Time  CT Angiography 3D Post-processing Slice Profile		Resolution Factors influencing Slice Profile  Collimation  Pitch  Interpolation algorithm (360° or 180°) Factors influencing SSP Collimator width collimation => SSP Spiral CT Table speed or Pitch Interpolation Algorithm => mathematical process required to reconstruct axial images from the spiral volume data set Pitch & Slice Profile Slim vs Wide –SSP Comparison Slice Profile Slim  %Broaden Wide %Braden Pitch One  5.0 mm 0 6.3 mm  26 Pitch Two 6.5 mm  30 10.8 mm 116 WIDE 720 degree More photons SLIM 464 degree Less photons SSP Spatial resolution SSP Spatial resolution Smoother image Noisier image Slim - Advantages Improved Z –Resolution Reduced partial volume artifacts Slim + extended Pitch Longer coverage Same coverage with shorter scan time or thinner slices Less radiation dose Wide - Advantages Noise Reduction Smoother image Useful for scanning huge patient Only for scanning at Pitch One Lesion smaller than 1cm SCAN RANGE = 150mm 10/10/10 (15s) 5/10/5 (15s) Slice Profile = 10mm Slice Profile = 6.5mm Optimizing the Scanning Parameters Scan Length (mm) Scan Duration (s) Table Speed (mm/s) Smallest Collimation (mm) Scan Duration Depends on the scan length & patient’s breath-hold compliance Table Speed Pitch Factor 1 < Pitch < 2 to cover the whole volume in one breath-hold Computed Tomography 	CT Basics 	Principle of Spiral CT 	Scan Parameter & Image Quality 	Optimizing Injection Protocols 	Clinical Applications Volume Flow Rate Concentration Injection Protocols 300 mg I/ml non-ionic contrast 2 - 5 ml/s  cardiac output 80 - 150 ml patients’weight & region of interest Site Scan Delay Delay between injection initiation & the start of the scan sequence Injection Duration must be equal to or greater than Scan Time 50 100 150 200 250 Time[s] HU CONTRAST 50 100 150 200 250 Time[s] HU CONTRAST NaCl Bolus Duration < scan time Insufficient, inhomogeneous opacification Bolus Duration = scan time Uniform, maximum opacification E nhancement Curve of the Target Region 50 100 150 200 250 Time[s] HU Optimal Window Early Time-density curve of the target region Late Contrast Bolus Timing Determines optimal scan delay for spiral CTA sequence CONTRAST NaCl Imaging Volume for spiral CTA 10-20 ml of contrast is injected at the chosen rate for spiral After a delay of 8-10s, low-dose, single-level axial images are  acquired every 2s at the starting point of the imaging volume Dynamic scans at this position Dynamic Evaluation to generate  a Time-density curve Dynamic Scans ROI placed in the Aorta Computed Tomography 	CT Basics 	Principle of Spiral CT 	Scan Parameter & Image Quality 	Optimizing Injection Protocols 	Clinical Applications  Arterial Phase Venous Phase Dual Phase Liver Exam Liver Metastases Single Plane Imaging with Multiplanar Results 2D reconstruction based on a serial of axial images along a certain axis Sagittal Coronal Oblique recon. of Aorta Spine 3D image: AVM Max. Intensity Projection Surface Shaded Display (3D) Femoral Arteries CT Angiogram * * Computed Tomography 	CT Basics 	Principle of Spiral CT 	Scan Parameter & Image Quality 	Optimizing Injection Protocols 	Clinical Applications X-Ray Discovery  X-ray was discovered by a German scientist Roentgen 100 years ago. This made people for the first time be able to view the anatomy structure of human body without operation But it's superimposed And we couldn't view soft tissue  My name is Godfrey Hounsfield  I work for the Central Research  Labs. of EMI, Ltd in England  I developed the the first clinically  useful CT scanner in 1971 1963 - Alan Cormack developed a mathematical method of reconstructing images from x-ray projections For the first time we could view: CT Broke the Barrier - Tomographic or “Slice”anatomy- Density difference But it's time consuming And resolution needs to be improved Incident X-ray Transmitted ray SCATTERED X-RAYS BODY TISSUE Absorption by the tissue is proportional to the density Less dense tissue More dense tissue MORE ATTENUATION LESS ATTENUATION How does CT Work? Recon. & postpro. Data acquisition X-ray generation X-ray goes through collimator therefore penetrate only an axial layer of the object, called "slice" How does CT Work? Patient is placed in the center of the measurement field X-ray is passed through the patient’s slice from many direction along a 360° path The transmitted beams are captured by the detectors which digitizes these signals These digitized signals called raw data are sent to a computer which create the CT image How does CT Work? The object slice is divided into small volume elements called voxels. Each voxel is assigned a value which is dependent on the average amount of attenuation How is CT Image generated? The attenuation values are transferred to the computer where they are coded & used to create a slice image How is CT Image generated? 3rd Generation Design Rotating tube & detector 4th Generation Design Fixed ring detector Non Slip-ring Scanner Slip-ring Scanner Computed Tomography 	CT Basics 	Principle of Spiral CT 	Scan Parameter & Image Quality 	Optimizing Injection Protocols 	Clinical Applications Reconstruction of arbitrary slices (either contiguous or overlapping) within the scanned volume Distance between the slices is called Increment A B Volume Data Continuous data acquisition Increment Slice Thickness Increment = Slice Thickness  No Overlap  No Gaps Contiguous Image Reconstruction Increment  Overlap Slice Thickness Overlapping Image Reconstruction Increment < Slice Thickness  Overlap of slices  Closer image interval  More images created Increment > Slice Thickness  Gaps between slices  Images are further apart  Less images created Image Reconstruction with Gaps Increment Slice Thickness Standard CT / Slice Imaging Deep Inspiration Shallow Inspiration  Misregistration due to different  respiratory levels between slices Partial Volume Effect  Unable to resconstruct images at  arbitrary position  Slice imaging is slow Scan the whole region of  interest in one breath hold Reconstruction of overlapping  images without additional dose Retrospective reconstruction of slices in arbitrary position within the scanned volume No gaps since radiation always transmits the whole volume Spiral CT / Volume Imaging Computed Tomography 	CT Basics 	Principle of Spiral CT 	Scan Parameter & Image Quality 	Optimizing Injection Protocols 	Clinical Applications Table Speed (mm/rot) or Feed per 360 rotation Pitch Interpolation Process Increment (mm) Pitch => 		 Table Feed per rotation Collimation 10mm P1 10mm P2 30s 30s More Coverage in the same time with extended Pitch!! Pitch 2 covers 2x distance as Pitch 1 Scan Range = 300mm 10mm P1 10 mm/s 10mm P2 20 mm/s 30s 15s Cover the same volume in shorter time with extended Pitch To reduce artifacts due to table motion during spiral scanning, we use a special reconstruction process called INTERPOLATION  Interpolation Converts volume data into slice images Interpolation Algorithm  Wide Algorithm Slim Algorithm 2 x 360° = 720° raw data 2 (180+52) = 464° raw data Wide algorithm produces a broader image thickness Wide algorithm uses more raw data => less image noise PITCH 2 PITCH 1 Pitch 2 scanning produces a broader image thickness Pitch 2 scanning does not increase image noise 30% increase in image thickness with Pitch 2 Slice Sensitivity Profile ( SSP ) SSP describes the effective slice thickness of an image and to what extent anatomy within that slice contribute to the signal SSP RESOLUTION All points within the slice contribute equally & points outside of the slice do not contribute to the image at all . Image signal Ideal SSP Z-axis (mm) Collimation = width of x-ray beam =slice profile * * *