核医学影像中的数据处理

单击此处编辑母版标题样式,单击此处编辑母版文本样式,第二级,第三级,第四级,第五级,*,单击此处编辑母版标题样式,单击此处编辑母版文本样式,第二级,第三级,第四级,第五级,*,核医学影像中的数据处理,中国科学院高能物理研究所,北京市射线成像技术与装备工程技术研究中心,贠明凯,Modern Nuclear Medical Imaging,Acquire,Process,Apply,Scanners Computers Users,Outline,Data organization,Correction methods,Rebinning,Image reconstruction,Image registration and fusion,DICOM and PACS,Outline,Data organization,Correction methods,Rebinning,Image reconstruction,Image registration and fusion,DICOM and PACS,Data organization,List mode,Histgram,Sinogram,Linogram,SinogramPET,Sinogram,r,Projections and Sinogram,Sinogram,PET,Sinogram,r,Projections and Sinogram,Sinogram,SPECT,2D VS. 3D,Septa between crystal rings,Lower sensitivity,Lower random,Lower scatter,2D reconstruction,No septa,Higher sensitivity,Higher random,Higher scatter,3D reconstruction or hybrid reconstruction,Outline,Data organization,Correction methods,Rebinning,Image reconstruction,Image registration and fusion,DICOM and PACS,Scatter Coincidence,Trues Coincidence,Random Coincidence,True Counts & Noise,Normalization,A,B,C,D,Attenuation,A,B,C,D,Scatter,A,B,C,D,Need to correct the data,Correction methods,random,“dead time”,normalization,scatter,attenuation,decay,Arc correction,Depth of interaction,Motion correction,Partial volume,Axial of rotation,Camera head tilt,Random,Finite time window with,Energy window,Coincidence timing window,Activity,Random,Tail fitting,simplest,Small changes in tail, great changes in estimate,Estimation from singles rates,Measure the single count rate on each detector for a given time window,Subtracting from the prompts between detector pair,Singles rate is much larger than that of coincidence events,Single rates change in the same way over time,Delayed coincidence channel estimation,One channel is delayed before being sent to coincidence processing,Subtracted form prompt coincidences,Advantage,Accurate,Same dead time environment as prompt channel,Disadvantage,Increased system dead time,Doubling of the statistical noise due to random,Dead time correction,Decaying source experiment is performed,Dead time correction (con),Look up table,Uniform source,Known quantity,Short lived,Linear extrapolation from count rate for a given level of activity,Normalization,Causes of sensitivity variations,Summing of adjacent data elements,Detector efficiency variations,Geometric and solid angle effects,Rotational sampling,Time window alignment,Structural alignment,septa,Summing of adjacent data elements,Geometric and solid angle effects,Rotational sampling,LOR at the edge are sampled less than LOR close to the center,Crystal interface factors,Time window alignment factors,Normalization methods (con),Direct normalization,Simplest approach,Adequate statistical quality,Very uniform activity sources,Scatter in normalization should be substantially different from normal imaging,Normalization methods (con),Component-based normalization,Scatter correction,LORs recorded outside object boundary can only be explained by scatter,The scatter distribution is very broad,Scattered coincidences fall within the photo-peak window mainly due to scattered once,Scatter correction,Energy spectra distribution of scattered 511KeV photons according to the number of times each photon scatters,Scatter correction,Empirical scatter corrections,Fitting the scatter tails,Direct measurement technique,Energy window techniques,Dual energy window methods,Multiple energy window methods,Convolution and de-convolution,Simulation-based scatter correction,Analytical simulation,Monte Carlo simulation,Fitting the scatter tails,Simplest approach,Fit an analytical function to scatter tails,Second order polynomial or 1D Gaussian,Coincidences outside the object are entirely scatter events,Not always well approximated, particularly in thorax,Direct measurement technique,Only applicable to PET with retractable septa,Steps,Make a measurement of the same object with and without septa,Scaling septa extended projections for different efficiency,Subtract from projections of polar angle 0,Estimate the oblique scatter by interpolation of the direct plane scatter,Dual energy window methods,Dual energy window methods,Multiple energy window methods,Scatter Correction,Analytical simulation,Scatter Correction,A,B,Single Scatter - Model based correction,Calculate the contribution for an arbitrary scatter point using the Klein-Nishina equation,Before,Scatter,correction,After,Scatter,correction,Attenuation correction,Attenuation in the body is equal to that of source lying along the same LOR,Zaidi H, Hasegawa B. J Nucl Med 2003; 44:291-315.,SPECT,PET,Attenuation correction (con),Measured attenuation correction,Coincidence transmission data,Long-lived positron emitter,Normally more than one rod source are used,Sinogram windowing is applied provide location of rod,Impractical in 3D,Singles transmission data,Shielded point transmission source,Separate blank scan is needed,Significant scatter and broad beam,Measured attenuation correction,Coincidence measurement using rod source,Transmission measurement using point source,CT scan,Advantage,High statistical quality,High spatial resolution,Significant reduction in scan time,Disadvantage,Faster CT, slower PET,Smaller FOV of CT,Difficulty in registration,values do not scale linearly,Attenuation correction for PET,Types of transmission images,Coincident photon Ge-68/Ga-68,(511 keV),high noise,15-30 min scan time,low bias,low contrast,Single photon Cs-137,(662 keV),lower noise,5-10 min scan time,some bias,lower contrast,X-ray,(30-140kVp),no noise,1 min scan time,potential for bias,high contrast,Other attenuation correction methods,Calculated attenuation,Regular geometric outline,Constant tissue,Segmented attenuation,Segment transmission image according to tissue type,Assigning known attenuation coefficients,Forward projection,attenuation correction,Attenuation/Scatter correction,University of Pennsylvania PET Center,No AC or Scatter Corr,AC and Scatter Corr,Philips,Allegro,Arc correction,Different sampling distance at different radial position,Equal sampling distance is required in analytical method,Interpolation method,Nearest interpolation,Linear interpolation,B-spline interpolation (negative values!),DOIdepth of interaction,DOIdepth of interaction,（,con,）,Dual Layer,A Point Spread Function (PSF) describes the response of an imaging system to a point source or point object. A system that knows the response of a point source from everywhere in its field of view can use this information to recover the original shape and form of imaged objects.,PSFs are used in precision imaging instruments, such as microscopy, ophthalmology, and astronomy (e.g. the Hubble telescope) to make geometric corrections to the final image.,Point Spread Function (PSF),Motion correction,Cardiac motion and respiration,Motion correction,（,con,）,Gated frames,List mode,Respiratory motion is distributed throughout the whole body,Impact is rarely on detection, but often affects quantitation,Static wholebody,Single respiratory phase,(1 of 7, so noisier), 1 cc lesion on CT,Whole-body respiratory gated PET/CT: Patients,Partial volume effect,Characters,Object or structure being imaged only partially occupies the sensitive volume of scanner,Signal amplitude becomes diluted with signals from surrounding structures,The degree of underestimation of radioactivity concentration will depend not only on its size but also on the relative concentration in surrounding structures,Correction methods,Resolution recovery,Use of anatomical imaging data,A Point Spread Function (PSF) describes the response of an imaging system to a point source or point object. A system that knows the response of a point source from everywhere in its field of view can use this information to recover the original shape and form of imaged objects.,PSFs are used in precision imaging instruments, such as microscopy, ophthalmology, and astronomy (e.g. the Hubble telescope) to make geometric corrections to the final image.,Point Spread Function (PSF),Partial volume effectMAP,assumptions:,camera moves along circular orbit,orbit is reproducible,calibration method finds system geometry,problem 1: tilting detector,assumption: camera moves along circular orbit,AORAxial of rotation,Offset of AOR,Rotation of AOR,Nutation of AOR,Camera head tilt,Heads need to be exactly parallel to axis of rotation,Correct alignment,Head tilt,pinhole calibration,Dirk Bequ, Kathleen Vunckx,circular orbit,circular orbit + new model,extension 2: circular orbit +,arbitrary,small deviations,measurement,model,Michel Defrise, Chris Vanhove,extension 2: circular orbit +,arbitrary,small deviations,old,new,translations,rotations,1mm,-3mm,1.5mm,-1.5mm,1.5mm,-1mm,1,o,-2,o,1.5,o,-1.5,o,3.5,o,-2.5,o,1mm,1.2,1.4,1.6,1.8,2mm,Outline,Data organization,Correction methods,Rebinning,Image reconstruction,Image registration and fusion,DICOM and PACS,Rebinning,Convert 3D data to 2D,SSRB and MSRB,SSRB- Single-slice rebinning,Detection: center slice,Simple,Fast,Resolution loss,MSRB- Multi-slice rebinning,Distribute along all intermediate slices,De-blurring along z-axis,Fourier rebinning,Outline,Data organization,Correction methods,Rebinning,Image reconstruction,Image registration and fusion,DICOM and PACS,Image reconstruction,Analytical,FBP,BPF,FDK,3D RP,Iterative,ART,MLEM,OSEM,OSLS,MAP,Analytical algorithms,For example, FBP (Filtered Back-projection),Treat the unknown image as continuous,Point-by-point reconstruction,Regular grid points are commonly chosen,Treat projection process as line integral theoretically,解析重建,-FBP,FBP,back projection (BP) = summation of projections,filtered back projection (FBP),FDK,Feldkamp,、,Davis,、,Kress,FDK,3D RPRe-projection,Steps of 3D RP,Extract 2D sinograms,Reconstruct each with 2D FBP and stack to form 3D image,Forward project to calculate missing LORs,Extract 2D projection data of all oblique slices,Take 2D Fourier transform,Back project data through 3D image matrix,Repeat for all angles and oblique slices,What is iterative reconstruction,Discrete measurements, discrete image,Optimization,Attractions of iterative methods,Either consistent or inconsistent is OK,Complex geometry,Physical effects and detection processes can be modeled,Non-negativity,Great reducing streaking artifacts,Better contrast recovery,Classification of iteration reconstruction methods,ART (algebraic reconstruction techniques),MART (multiplicative ART),AART (additive ART),SIRT (simultaneous iterative reconstruction),SMART (simultaneously MART),BI-ART (block iterative ART),BI-SMART (block iterative SMART),RBI-SMART (rescaled BI-SMART),Statistical algorithms,MAP:,Maximize the conditional probability,P(image|data),MLEM:,Maximize the probability,P(data|image),Statistical algorithmsGaussian assumption,P is projection column matrix, A is system matrix, F image column matrix, C is the covariance matrix of the data,Assumed all standard deviations are identical and equal to 1, idealized parallel projection, perfect resolution and no attenuation or other degrading affects,Statistical algorithmsPoisson assumption,实测数据,迭代重建,-MLEM&OSEM,正投影,比较,更新,重建,MLEM, OSEM, .,likelihood,iteration,Sinogram,r,Subset 1,Subset 2,Subset 3,Subset 4,1, 3 2 4,Subset order,0,1,2,3,4,10,40,ordered subsets,1 iteration of 40 subsets,(2 proj per subset),System matrix,Scan geometry,Collimator/detector response,Attenuation,Scatter (object, collimator, scintillator),Duty cycle (dwell time at each angle),Detector efficiency,Dead-time losses,Positron range,Non-colinearity,Crystal penetration,Considerations of system matrix,Quantitative accuracy,Spatial accuracy,Computation time,Storage space,Model uncertainties,Artifacts due to over simpleifications,System matrix tricks,Factorize,Symmetry,Sparseness,Approximation,Partial Monte Carlo,System matrix model,Reconstruction image of uniform source,FBP VS. OSEM,FBPanalytical,Pros,：,Single pass,Linear,Fast,Cons,：,Streak artifact,Poor resolution,Correction not built-in,OSEMiteration,Pros,：,Better resolution,Better contrast,Lower noise,Cons,：,Extensive time consuming,Memory consuming,Required user training,FBP VS. OSEM,Phantom test (left),Clinical results (right),Outline,Data organization,Correction methods,Rebinning,Image reconstruction,Image registration and fusion,DICOM and PACS,Image Registration,PET,CT,PET/CT,Voxel based image registration,Image Registration,Image Registration,算法流程图,相似性测量一般用到的函数有,：,相同模态图像：残差（,sum of square difference,）,不同模态图像：互信息（,mutual information,）,一般用来做配准的优化算法有：,六参数或十二参数的优化一般使用,Powell,优化算法,多参数优化一般使用,LBFGS,（,limited-memory BroydenFletcherGoldfarbShanno,）优化算法（由牛顿算法演变而来）,Image Fusion,Alpha Blending based,Adaptive alpha blending,Alpha blending,Adaptive Alpha blending,Outline,Data organization,Correction methods,Rebinning,Image reconstruction,Image registration and fusion,DICOM and PACS,DICOM and PACS,DICOM,Digital image and Communication in Medicine,Created by ACR (American College of Radiology) and NEMA (National Electrical Manufacturers Association) in 1985,Two components:,Communication protocols and image format standards,DICOM image VS. General image,Structure:,DICOM contains header and image data sections;,Other image file such as BMP, JPG, TIFF, which contain also two sections;,Size:,The header size of DICOM is variable;,The header size of many general image is constant,Contents:,DICOM contains additional patients data such as basic information, study information and so on;,General image header describes basic image parameters, such as image size, compression type,DICOM logical layers,Patient information:,Patients name, patients ID, patients birth date, hospital information system,Study information:,Study information such as dose, injection time and additional examination information such as study name study date;,Series information:,Series ID, manufacturer and institution name,Image information:,The size of pixel, image size, pixel value and how it is encoded,PACS,PACS,The Picture Archival and Communication System,A system for storage of images and transferring images between computers in different facilities through networks,PACS is helpful to provide comparative studies among different image modalities,PACS (con),Thanks,Our Group is growing up,TOF-PET,TOF,信息,PET,原始数据,偶然符合校正,发射图像,衰变校正,死时间校正,弓形几何校正,衰减校正,散射校正,归一化校正,图像重建,数据重组,数据校正,