Turku PET Centre Vesa Oikonen http users utu

Turku PET Centre Vesa Oikonen http: //users. utu. fi/vesoik/ MYOCARDIAL BLOOD FLOW WITH [15 O]H 2 O PET: ERROR ESTIMATES TPC modelling workshop 2005 -09 -14

Model parameters • MBF myocardial blood flow in perfusable tissue volume • α perfusable tissue fraction, PTF • VA myocardial blood volume fraction and spill-over from LV cavity

MBF model is fitted… 1. to ROI curves drawn on LV cavity (“lv”) and on entire myocardial tissue surrounding the cavity (“whole”) to estimate virtual arterial blood concentration, CA(t) 2. To smaller myocardial ROIs using CA(t) as model input.

Results of MBF model fitmbf 2. 1. 3 (c) 1999 -2005 by Turku PET Centre Date: 2005 -09 -13 Study: s 05851 dy Data file: S 05851s 05851. dft Reference region: lv All Data range: 0 - 360 sec (N=20) Beta: 0. 85 Data was not weighted. Region lat ant sept whole flow. . All All 0. 8120 0. 8613 0. 7151 0. 8353 alpha Va 0. 6564 0. 6391 0. 5245 0. 6329 (Weighted) Sum-of-squares describing goodness-of-fit 0. 2671 0. 2036 0. 4841 0. 2863 Estimates of model parameters SS AIC 9. 5215 e+001 3. 8708 e+001 1. 2060 e+002 4. 3434 e+001 1. 9348 e+002 5. 2888 e+001 4. 8313 e+001 2. 5140 e+001 Akaike information criteria for comparison between models

Error estimate for each model parameter • 95% confidence limits and/or standard deviations • With command-line options -CL and –SD • Bootstrap method (Efron and Tibshirani, 1993; Turkheimer et al. , 1998) • Starting from fitmbf version 2. 1. 0, 2005 -07 -15

Measured and fitted TACs

Results with error estimates fitmbf 2. 1. 3 (c) 1999 -2005 by Turku PET Centre Date: 2005 -09 -13 Study: s 05851 dy Data file: S 05851s 05851. dft Reference region: lv All Data range: 0 - 360 sec (N=20) Beta: 0. 85 Data was not weighted. Region lat SD CL CL ant SD CL CL sept SD CL CL whole SD CL CL flow. All. 95% 95%. . 95% . Lower Upper All. Lower Upper 0. 8120 0. 0456 0. 7241 0. 9012 0. 8613 0. 0529 0. 7701 0. 9690 0. 7151 0. 0758 0. 5683 0. 8756 0. 8353 0. 0401 0. 7561 0. 9169 alpha Va 0. 6564 0. 0164 0. 6228 0. 6849 0. 6391 0. 0193 0. 6073 0. 6768 0. 5245 0. 0205 0. 4827 0. 5612 0. 6329 0. 0126 0. 6076 0. 6594 0. 2671 0. 0163 0. 2364 0. 2955 0. 2036 0. 0191 0. 1674 0. 2376 0. 4841 0. 0202 0. 4439 0. 5236 0. 2863 0. 0123 0. 2601 0. 3097 SS AIC 9. 5215 e+001 3. 8708 e+001. . . 1. 2060 e+002 4. 3434 e+001. . . 1. 9348 e+002 5. 2888 e+001. . . 4. 8313 e+001 2. 5140 e+001. . .

Larger errors with higher MBF BASAL study Region lat. SD ant. SD sept. SD whole. SD Stimulation Region lat. SD ant. SD sept. SD whole. SD flow All. 0. 7259 0. 0493 0. 6751 0. 0596 0. 5756 0. 1109 0. 6823 0. 0627 . All . . flow All . . 4. 1351 0. 4198 2. 8656 0. 8369 4. 9285 1. 0812 4. 0911 0. 7307 alpha Va 0. 6577 0. 0176 0. 5503 0. 0186 0. 5703 0. 0401 0. 5924 0. 0193 0. 1498 0. 0176 0. 2290 0. 0187 0. 4679 0. 0370 0. 2555 0. 0197 SS AIC 6. 3335 e+001 3. 0554 e+001. . 7. 2341 e+001 3. 3213 e+001. . 3. 4841 e+002 6. 4653 e+001. . 7. 0570 e+001 3. 2717 e+001. . alpha Va 0. 7403 0. 0486 0. 4734 0. 0922 0. 7674 0. 1251 0. 6560 0. 0717 0. 0500 0. 0461 0. 2891 0. 0906 0. 2390 0. 1183 0. 1708 0. 0682 SS AIC 7. 1618 e+001 3. 3012 e+001. . 1. 3123 e+002 4. 5125 e+001. . 1. 6528 e+002 4. 9738 e+001. . 4. 9344 e+001 2. 5562 e+001. .

… with confidence limits fitmbf 2. 1. 3 (c) 1999 -2005 by Turku PET Centre Date: 2005 -09 -13 Study: s 06003 dy Data file: S 06002s 06003. dft Reference region: lv All Data range: 0 - 360 sec (N=20) Beta: 0. 85 Data was not weighted. Region lat CL CL ant CL CL sept CL CL whole CL CL flow. 95% 95% All Lower Upper 4. 1351 3. 3282 4. 9667 2. 8656 1. 5818 4. 7635 4. 9285 2. 9679 6. 9406 4. 0911 2. 8782 5. 7352 alpha Va 0. 7403 0. 6074 0. 7831 0. 4734 0. 3160 0. 6941 0. 7674 0. 5418 0. 9908 0. 6560 0. 5174 0. 7852 0. 0500 0. 0228 0. 1776 0. 2891 0. 0794 0. 4426 0. 2390 0. 0292 0. 4528 0. 1708 0. 0445 0. 2994 SS AIC 7. 1618 e+001 3. 3012 e+001. . 1. 3123 e+002 4. 5125 e+001. . 1. 6528 e+002 4. 9738 e+001. . 4. 9344 e+001 2. 5562 e+001. .

Reduce SD by constraining α • α during stimulation study is constrained to the estimate from basal conditions • With command-line options -f. Alpha and -f=basal. res, when using fitmbf 2. 1. 0 and later versions • Also VA can be constrained with option -f. Va.

… after constraints for α fitmbf 2. 1. 3 (c) 1999 -2005 by Turku PET Centre Date: 2005 -09 -13 Study: s 06003 dy Data file: S 06002s 06003. dft Reference file: resultss 06002 mbf 1. res Reference region: lv All Data range: 0 - 360 sec (N=20) Beta: 0. 85 Data was not weighted. Region lat CL CL ant CL CL sept CL CL whole CL CL flow. 95% 95% All Lower Upper 3. 5534 2. 8580 4. 3003 3. 4219 2. 5938 4. 6877 3. 2420 2. 3396 4. 5089 3. 5087 2. 9207 4. 2717 alpha Va 0. 6577 0. 5503 0. 5703 0. 5924 0. 1283 0. 1096 0. 1494 0. 2140 0. 1865 0. 2379 0. 4294 0. 4071 0. 4600 0. 2319 0. 2190 0. 2447 SS AIC 7. 6282 e+001 3. 1480 e+001. . 1. 3465 e+002 4. 2845 e+001. . 1. 7751 e+002 4. 8372 e+001. . 5. 0833 e+001 2. 3362 e+001. .

… constraints for α and VA fitmbf 2. 1. 3 (c) 1999 -2005 by Turku PET Centre Date: 2005 -09 -13 Study: s 06003 dy Data file: . . S 06002s 06003. dft Reference file: s 06002 mbf 1. res Reference region: lv All Data range: 0 - 360 sec (N=20) Beta: 0. 85 Data was not weighted. Region lat CL CL ant CL CL sept CL CL whole CL CL flow. 95% 95% All Lower Upper 3. 0499 2. 6776 3. 4477 2. 9320 2. 3330 3. 6193 2. 3688 1. 9069 2. 9013 2. 8962 2. 4452 3. 3895 alpha Va 0. 6577 0. 5503 0. 5703 0. 5924 0. 1498 0. 2290 0. 4679 0. 2555 SS AIC 9. 4997 e+001 3. 3384 e+001. . 1. 4408 e+002 4. 1715 e+001. . 2. 2999 e+002 5. 1068 e+001. . 7. 2143 e+001 2. 7881 e+001. .

Constraints or not? • Calculate results without and with constraints (α, α and VA, VA) • Select the method that gives constantly smaller AIC

References 1. 2. 3. 4. 5. 6. Araujo LI, Lammertsma AA, Rhodes CG, Mc. Falls EO, Iida H, Rechavia E, Galassi A, De Silva R, Jones T, Maseri A. Noninvasive quantification of regional myocardial blood flow in coronary artery disease with oxygen-15 -labeled carbon dioxide inhalation and positron emission tomography. Circulation 1991; 83: 875 -885. Efron B, Tibshirani RJ. An introduction to the bootstrap. (1993) Chapman & Hall, New York. Iida H, Rhodes CG, de Silva R, Yamamoto Y, Araujo LI, Maseri A, Jones T. Myocardial tissue fraction – correction for partial volume effects and measure of tissue viability. J. Nucl. Med. 1991; 32: 2169 -2175. Iida H, Rhodes CG, de Silva R, Araujo LI, Bloomfield P, Lammertsma AA, Jones T. Use of the left ventricular time-activity curve as a noninvasive input function in dynamic oxygen-15 -water positron emission tomography. J. Nucl. Med. 1992; 33: 1669 -1677. Turkheimer F, Sokoloff L, Bertoldo A, Lucignani G, Reivich M, Jaggi JL, Schmidt K. Estimation of component and parameter distributions in spectral analysis. J. Cereb. Blood Flow Metabol. 1998; 18: 1211 -1222. Watabe H, Jino H, Kawachi N, Teramoto N, Hayashi T, Ohta Y, Iida H. Parametric imaging of myocardial blood flow with 15 O-water and PET using the basis function method. J. Nucl. Med. 2005; 46: 1219 -1224.

More information • Detailed report TPCMOD 0005 http: //www. turkupetcentre. net/reports/ • Analysis instructions http: //www. turkupetcentre. net/analysis/doc/tracer/mbf_h 2 o. html • Bootstrap method in estimation of variance of model parameters http: //www. turkupetcentre. net/modelling/workshop 1/bootstrap. ppt