[Insight-users] Re: itkmattesMutualInformationImagetoImageMetric

Wed May 31 15:16:12 EDT 2006

You are getting the expected behaviour. You cannot compare Mutual 
Information values across resolution levels. The MU content in two 
images at two different resolution levels is certainly to be very 
different.

Grace Chen wrote:

>Thanx for your inputs!
>
>However, because I use image pyramid in my registration. I prints the metric
>value out during each interation.  I get increased value everytime the
>program move to another level. (The print-out of the program is included
>below.)
>
>Does that mean image pyramid is not good for multi-modality registration??
>Or does it matter??
>
>Grace
>----------------------------------------------------------------------------
>-----
>0 -0.440609 [0.0559426, 0.104512, -0.0340259, 3.98082, -2.79831, 2.1402]
>
>1 -0.458077 [0.0237338, 0.074854, -0.0175932, 5.21787, -4.36703, 2.1402]
>
>2 -0.444355 [0.0407254, 0.043635, -0.0228596, 4.92513, -3.41354, 2.1402]
>
>3 -0.442927 [0.0535433, 0.000940102, -0.0426362, 5.67537, -2.75961, 2.1402]
>
>4 -0.447368 [0.046386, 0.0145486, -0.0167889, 5.19313, -2.8772, 2.1402]
>
>5 -0.450611 [0.0367333, 0.0255149, -0.00895245, 4.76924, -3.14028, 2.1402]
>
>6 -0.480234 [0.0273652, 0.0338326, 0.010831, 4.38405, -3.45562, 2.1402]
>
>7 -0.491004 [0.02514, 0.0276529, 0.0195927, 4.60897, -3.56254, 2.1402]
>
>8 -0.5167 [0.0196279, 0.0256835, 0.0341163, 4.70212, -3.79241, 2.1402]
>
>9 -0.517608 [0.0221036, 0.0190742, 0.0328978, 4.9453, -3.73621, 2.1402]
>
>10 -0.518191 [0.0211846, 0.02218, 0.0420499, 4.88552, -3.84425, 2.1402]
>
>11 -0.52135 [0.0208545, 0.0206294, 0.0415968, 4.94422, -3.86546, 2.1402]
>
>12 -0.523347 [0.0204449, 0.0191823, 0.0413636, 5.00101, -3.89139, 2.1402]
>
>13 -0.52123 [0.0211378, 0.0185994, 0.04133, 5.05607, -3.9209, 2.1402]
>
>14 -0.518093 [0.0197732, 0.018755, 0.0435616, 5.06313, -3.98278, 2.1402]
>
>15 -0.521895 [0.0201374, 0.0178134, 0.0462639, 5.12112, -4.00537, 2.1402]
>
>16 -0.52065 [0.0189278, 0.0179555, 0.0481575, 5.13278, -4.0666, 2.1402]
>
>17 -0.521602 [0.018905, 0.0169584, 0.049328, 5.1897, -4.09224, 2.1402]
>
>18 -0.52087 [0.019628, 0.017331, 0.0545369, 5.23946, -4.12855, 2.1402]
>
>19 -0.5204 [0.018108, 0.0172796, 0.0563701, 5.25343, -4.18928, 2.1402]
>
>20 -0.520973 [0.0177511, 0.0166056, 0.0577169, 5.30406, -4.2258, 2.1402]
>
>21 -0.521113 [0.0175278, 0.0177341, 0.0579758, 5.27651, -4.24036, 2.1402]
>
>22 -0.521249 [0.0201751, 0.0204605, 0.0574916, 5.27965, -4.22711, 2.1402]
>
>23 -0.522455 [0.0201017, 0.0202894, 0.0575649, 5.28657, -4.23073, 2.1402]
>
>24 -0.522845 [0.0200306, 0.0201182, 0.0576416, 5.2935, -4.2343, 2.1402]
>
>25 -0.523012 [0.0199572, 0.0199502, 0.0577323, 5.30037, -4.23801, 2.1402]
>
>26 -0.523017 [0.019885, 0.0197831, 0.0578181, 5.30723, -4.24171, 2.1402]
>
>27 -0.522595 [0.0197908, 0.0196344, 0.0579005, 5.31359, -4.24625, 2.1402]
>
>28 -0.522674 [0.0197011, 0.019487, 0.0579754, 5.31998, -4.25072, 2.1402]
>
>29 -0.521521 [0.0196008, 0.0193495, 0.0581019, 5.32608, -4.25558, 2.1402]
>
>30 -0.521485 [0.019483, 0.0192236, 0.0582602, 5.33178, -4.26091, 2.1402]
>
>31 -0.522264 [0.0193449, 0.0191302, 0.058358, 5.33658, -4.26705, 2.1402]
>
>32 -0.522681 [0.0191987, 0.0190598, 0.0584676, 5.34094, -4.27352, 2.1402]
>
>33 -0.522735 [0.019052, 0.0190024, 0.0585873, 5.34506, -4.28015, 2.1402]
>
>34 -0.522813 [0.0188757, 0.019, 0.0586345, 5.34733, -4.28762, 2.1402]
>
>35 -0.522514 [0.0187408, 0.0190399, 0.0584218, 5.34984, -4.295, 2.1402]
>
>36 -0.521768 [0.0187707, 0.0191056, 0.0580022, 5.35557, -4.30025, 2.1402]
>
>37 -0.52162 [0.0187711, 0.0189871, 0.0578692, 5.36285, -4.30304, 2.1402]
>
>38 -0.521197 [0.0187743, 0.0189535, 0.0577763, 5.36943, -4.30726, 2.1402]
>
>39 -0.521503 [0.0187491, 0.0189253, 0.0576104, 5.37551, -4.31215, 2.1402]
>
>40 -0.52184 [0.0187632, 0.0188649, 0.0576716, 5.3825, -4.31564, 2.1402]
>
>41 -0.52203 [0.0188227, 0.0189083, 0.0572787, 5.38909, -4.31975, 2.1402]
>
>42 -0.521454 [0.0187143, 0.0189169, 0.0573131, 5.39284, -4.3266, 2.1402]
>
>43 -0.521827 [0.0186828, 0.0189044, 0.0573726, 5.39863, -4.33185, 2.1402]
>
>44 -0.521805 [0.0186981, 0.018911, 0.0573623, 5.40497, -4.33641, 2.1402]
>
>45 -0.521772 [0.0187554, 0.0189015, 0.0573842, 5.41202, -4.33977, 2.1402]
>
>46 -0.521918 [0.0188327, 0.0188701, 0.0573479, 5.41947, -4.34214, 2.1402]
>
>47 -0.522521 [0.018739, 0.0188427, 0.0574597, 5.42425, -4.34831, 2.1402]
>
>48 -0.521881 [0.0187046, 0.0190295, 0.0572338, 5.42603, -4.35589, 2.1402]
>
>49 -0.521567 [0.0186666, 0.0190674, 0.0573927, 5.4308, -4.36207, 2.1402]
>
>0 -0.389391 [0.0187093, 0.0191248, 0.0573509, 5.42544, -4.35649, 2.14127]
>
>1 -0.389505 [0.0187492, 0.0191828, 0.0573081, 5.41998, -4.35101, 2.14234]
>
>2 -0.389436 [0.018788, 0.0192422, 0.0572655, 5.41448, -4.34557, 2.14341]
>
>3 -0.389401 [0.0188278, 0.0193, 0.0572215, 5.40901, -4.34009, 2.14446]
>
>4 -0.389451 [0.018868, 0.0193577, 0.0571799, 5.40352, -4.33464, 2.1455]
>
>5 -0.389498 [0.0189076, 0.0194122, 0.0571426, 5.39805, -4.32915, 2.14653]
>
>6 -0.389495 [0.0189478, 0.0194658, 0.0571099, 5.39261, -4.32365, 2.14757]
>
>7 -0.389489 [0.0189853, 0.0195208, 0.0570757, 5.38713, -4.31818, 2.14864]
>
>8 -0.38948 [0.0190242, 0.0195751, 0.0570417, 5.38172, -4.31265, 2.14971]
>
>9 -0.389535 [0.019064, 0.0196318, 0.056998, 5.37632, -4.30711, 2.15077]
>
>10 -0.389525 [0.0191047, 0.0196881, 0.0569549, 5.37095, -4.30154, 2.15183]
>
>11 -0.389573 [0.0191432, 0.0197446, 0.0569186, 5.36554, -4.29601, 2.15292]
>
>12 -0.389567 [0.0191819, 0.0198004, 0.0568823, 5.36006, -4.29055, 2.154]
>
>13 -0.389627 [0.019221, 0.0198535, 0.0568449, 5.35467, -4.285, 2.15508]
>
>14 -0.389533 [0.0192599, 0.0199053, 0.0568056, 5.34927, -4.27946, 2.15615]
>
>15 -0.389808 [0.0193003, 0.0199556, 0.0567747, 5.34391, -4.27387, 2.15722]
>
>16 -0.390001 [0.0193408, 0.0200036, 0.0567476, 5.3386, -4.26826, 2.15831]
>
>17 -0.390044 [0.0193779, 0.0200513, 0.0567296, 5.33321, -4.26271, 2.15943]
>
>18 -0.390089 [0.019414, 0.0201015, 0.056709, 5.32774, -4.25725, 2.16053]
>
>19 -0.390178 [0.0194489, 0.0201502, 0.0566886, 5.32228, -4.25178, 2.16166]
>
>20 -0.390225 [0.0194831, 0.0201989, 0.0566697, 5.31684, -4.24628, 2.1628]
>
>21 -0.390266 [0.0195176, 0.0202466, 0.0566484, 5.31149, -4.24071, 2.16393]
>
>22 -0.390314 [0.0195499, 0.0202953, 0.0566288, 5.3061, -4.23517, 2.16508]
>
>23 -0.390361 [0.0195833, 0.0203445, 0.0566087, 5.30071, -4.22963, 2.16621]
>
>24 -0.390561 [0.0196143, 0.0203937, 0.0565863, 5.29526, -4.22415, 2.16735]
>
>25 -0.390608 [0.0196454, 0.0204433, 0.0565666, 5.28983, -4.21866, 2.16851]
>
>26 -0.390654 [0.0196803, 0.0204966, 0.0565403, 5.28438, -4.21317, 2.16963]
>
>27 -0.390701 [0.0197127, 0.02055, 0.0565067, 5.27897, -4.20765, 2.17075]
>
>28 -0.390748 [0.0197464, 0.020604, 0.0564753, 5.27358, -4.20211, 2.17187]
>
>29 -0.390799 [0.019781, 0.0206587, 0.0564385, 5.26825, -4.19652, 2.17301]
>
>30 -0.390924 [0.0198148, 0.020714, 0.0564045, 5.26291, -4.19093, 2.17415]
>
>31 -0.390978 [0.0198506, 0.0207689, 0.0563671, 5.25765, -4.18527, 2.1753]
>
>32 -0.391097 [0.0198857, 0.020823, 0.0563262, 5.25239, -4.17961, 2.17645]
>
>33 -0.391153 [0.0199238, 0.0208752, 0.0562792, 5.24712, -4.17396, 2.17758]
>
>34 -0.391209 [0.0199613, 0.0209244, 0.0562253, 5.24197, -4.16818, 2.17865]
>
>35 -0.39122 [0.0199967, 0.0209734, 0.0561743, 5.23681, -4.16243, 2.17975]
>
>36 -0.391208 [0.0200331, 0.0210228, 0.0561208, 5.23162, -4.15669, 2.18084]
>
>37 -0.391263 [0.0200683, 0.0210748, 0.0560653, 5.22649, -4.15091, 2.18197]
>
>38 -0.391314 [0.0201025, 0.0211254, 0.0560108, 5.22138, -4.14512, 2.18313]
>
>39 -0.391315 [0.0201379, 0.021176, 0.0559565, 5.21633, -4.13928, 2.1843]
>
>40 -0.391372 [0.020176, 0.0212252, 0.0559052, 5.21137, -4.13336, 2.18546]
>
>41 -0.39142 [0.0202143, 0.0212756, 0.0558562, 5.20641, -4.12743, 2.18663]
>
>42 -0.391472 [0.0202572, 0.0213236, 0.0558088, 5.20158, -4.12141, 2.18776]
>
>43 -0.391519 [0.0202995, 0.0213715, 0.0557598, 5.19674, -4.11538, 2.18891]
>
>44 -0.391574 [0.0203443, 0.0214211, 0.0557124, 5.19194, -4.10932, 2.19005]
>
>45 -0.39162 [0.0203885, 0.0214708, 0.0556644, 5.18715, -4.10327, 2.19121]
>
>46 -0.39167 [0.0204329, 0.0215207, 0.0556168, 5.18235, -4.09721, 2.19238]
>
>47 -0.391716 [0.0204762, 0.0215718, 0.0555651, 5.17761, -4.09111, 2.19353]
>
>48 -0.391721 [0.0205174, 0.0216242, 0.0555157, 5.17278, -4.08509, 2.19469]
>
>49 -0.391766 [0.0205587, 0.0216747, 0.0554653, 5.1679, -4.0791, 2.19585]
>
>0 -0.293629 [0.0206299, 0.0217326, 0.0554804, 5.16151, -4.07467, 2.19658]
>
>1 -0.293697 [0.0206962, 0.0217884, 0.055503, 5.1551, -4.07027, 2.1973]
>
>2 -0.293755 [0.0207634, 0.0218441, 0.0555272, 5.14872, -4.06583, 2.19804]
>
>3 -0.293865 [0.0208278, 0.0218952, 0.0555421, 5.14225, -4.0615, 2.19877]
>
>4 -0.29399 [0.0208918, 0.0219462, 0.0555567, 5.13578, -4.05718, 2.19944]
>
>5 -0.294044 [0.0209524, 0.0219982, 0.055583, 5.12937, -4.05278, 2.20014]
>
>6 -0.2941 [0.021012, 0.022049, 0.0556144, 5.12296, -4.04837, 2.20085]
>
>7 -0.294231 [0.0210701, 0.0221001, 0.0556433, 5.11657, -4.04393, 2.20157]
>
>8 -0.294284 [0.0211241, 0.0221552, 0.0556723, 5.1101, -4.03961, 2.20229]
>
>9 -0.294336 [0.0211772, 0.022211, 0.0557008, 5.10361, -4.03534, 2.20301]
>
>10 -0.294362 [0.0212304, 0.0222671, 0.0557319, 5.09703, -4.03118, 2.20374]
>
>11 -0.294546 [0.0212832, 0.0223233, 0.0557612, 5.09049, -4.02698, 2.20447]
>
>12 -0.29457 [0.0213338, 0.022381, 0.0557899, 5.08392, -4.02283, 2.20521]
>
>13 -0.294619 [0.0213839, 0.0224376, 0.0558198, 5.07738, -4.01861, 2.20592]
>
>14 -0.294669 [0.0214307, 0.0224928, 0.0558477, 5.07081, -4.01445, 2.20664]
>
>15 -0.29465 [0.0214796, 0.0225477, 0.0558771, 5.06424, -4.01028, 2.20734]
>
>16 -0.294699 [0.0215272, 0.0226027, 0.0559079, 5.05764, -4.00618, 2.2081]
>
>17 -0.294962 [0.0215761, 0.0226595, 0.0559332, 5.05102, -4.0021, 2.20883]
>
>18 -0.295011 [0.0216243, 0.0227165, 0.0559536, 5.04442, -3.99798, 2.20959]
>
>19 -0.295061 [0.0216733, 0.0227738, 0.0559782, 5.03782, -3.99387, 2.21035]
>
>20 -0.295111 [0.0217199, 0.0228307, 0.0560019, 5.03123, -3.98976, 2.21113]
>
>21 -0.295158 [0.021766, 0.0228869, 0.0560243, 5.02464, -3.98564, 2.2119]
>
>22 -0.295178 [0.02181, 0.0229439, 0.0560421, 5.01805, -3.98152, 2.21266]
>
>23 -0.295273 [0.0218512, 0.0230004, 0.0560632, 5.0114, -3.97749, 2.21345]
>
>24 -0.29532 [0.0218946, 0.0230556, 0.0560855, 5.00476, -3.97346, 2.21427]
>
>25 -0.295368 [0.0219363, 0.023112, 0.0561085, 4.99813, -3.96941, 2.2151]
>
>26 -0.295483 [0.0219783, 0.0231686, 0.0561281, 4.99155, -3.9653, 2.21596]
>
>27 -0.295485 [0.0220189, 0.0232239, 0.0561398, 4.98503, -3.96109, 2.21684]
>
>28 -0.295535 [0.0220623, 0.0232788, 0.0561568, 4.97855, -3.95681, 2.21771]
>
>29 -0.295583 [0.0220982, 0.023335, 0.0562, 4.97201, -3.95265, 2.21866]
>
>30 -0.295633 [0.0221333, 0.0233902, 0.0562436, 4.9655, -3.94844, 2.21962]
>
>31 -0.295679 [0.0221686, 0.0234457, 0.0562842, 4.95903, -3.94417, 2.22058]
>
>32 -0.295695 [0.0222031, 0.0235014, 0.0563223, 4.95256, -3.9399, 2.22155]
>
>33 -0.295744 [0.0222382, 0.0235583, 0.0563625, 4.94598, -3.9358, 2.22248]
>
>34 -0.295791 [0.0222705, 0.023615, 0.0564006, 4.93936, -3.93175, 2.2234]
>
>35 -0.295842 [0.0223052, 0.0236725, 0.0564318, 4.93278, -3.92765, 2.22434]
>
>36 -0.295987 [0.0223366, 0.0237306, 0.0564639, 4.92614, -3.92366, 2.2253]
>
>37 -0.296163 [0.0223679, 0.0237889, 0.0564979, 4.91948, -3.91969, 2.22627]
>
>38 -0.296211 [0.0224004, 0.0238459, 0.0565363, 4.91287, -3.91564, 2.22724]
>
>39 -0.29621 [0.022432, 0.0239043, 0.0565776, 4.90623, -3.91164, 2.22825]
>
>40 -0.296259 [0.0224627, 0.0239626, 0.056617, 4.89959, -3.90766, 2.22925]
>
>41 -0.296308 [0.0224945, 0.0240195, 0.0566568, 4.89299, -3.9036, 2.23027]
>
>42 -0.296357 [0.0225259, 0.024077, 0.0566959, 4.88635, -3.89962, 2.23129]
>
>43 -0.29642 [0.0225552, 0.0241341, 0.0567328, 4.87972, -3.89563, 2.23233]
>
>44 -0.296469 [0.0225819, 0.0241912, 0.0567734, 4.8731, -3.89162, 2.2334]
>
>45 -0.296519 [0.0226042, 0.0242463, 0.0568092, 4.86641, -3.88774, 2.23447]
>
>46 -0.296568 [0.0226277, 0.0243018, 0.0568419, 4.8597, -3.88389, 2.23555]
>
>47 -0.296619 [0.0226534, 0.0243579, 0.056879, 4.85302, -3.87999, 2.23663]
>
>48 -0.296666 [0.0226835, 0.0244134, 0.0569233, 4.84643, -3.87592, 2.23767]
>
>49 -0.296714 [0.0227186, 0.0244661, 0.056984, 4.83996, -3.87168, 2.23874]
>
>----------------------------------------------------------------------------
>-----------------------------
>
>
>
>
>
>----- Original Message ----- 
>From: "Karthik Krishnan" <Karthik.Krishnan at kitware.com>
>To: "Grace Chen" <Grace.Chen at swri.ca>
>Cc: "Kajetan Berlinger" <kaje at kaje.info>; <insight-users at itk.org>
>Sent: Wednesday, May 31, 2006 10:53 AM
>Subject: Re: [Insight-users] Re:
>itkmattesMutualInformationImagetoImageMetric
>
>
>  
>
>>Grace,
>>
>>Yes, As Kajetan mentioned, the Mattes metric needs to be minimized. This
>>may sound counterintuitive since Mutual information defined as :
>>
>>I(X,Y) = H(X) + H(Y) - H(X,Y)  has got to be positive,
>>
>>Yet the mattes yeilds values that are negative. Read the last line of
>>the GetValue() method in
>>itkMattesMutualInformationImageToImageMetric.txx. It reads :
>>     return static_cast<MeasureType>( -1.0 * sum );
>>In other words the classical MI value is negated.
>>
>>The other mutual information metrics though, need to be maximized. For
>>instance, if you use itk::MutualInformationHistogramImageToImageMetric,
>>you will have to maximize it.
>>
>>HTH
>>-karthik
>>
>>Kajetan Berlinger wrote:
>>
>>    
>>
>>>Hi Grace,
>>>
>>>does your optimizer maximize or minimize?
>>>Use optimizer->SetMinimize(true) to make sure.
>>>
>>>cu
>>>kaj
>>>
>>>
>>>Grace Chen wrote:
>>>
>>>      
>>>
>>>>Hi,
>>>>
>>>>        
>>>>
>>>>>From the ITK software guide, it seems to indicate that the metric
>>>>>value should get smaller (more negative) as the iteration
>>>>>increases.  However, that's not the case for me....
>>>>>          
>>>>>
>>>>I am very confused.  I've posted many questions here but no
>>>>answer.... Please help!!
>>>>
>>>>Thanx!!
>>>>
>>>>
>>>>Grace
>>>>
>>>>  ----- Original Message -----   From: Grace Chen   To:
>>>>insight-users at itk.org   Sent: Tuesday, May 30, 2006 4:44 PM
>>>>  Subject: itkmattesMutualInformationImagetoImageMetric
>>>>
>>>>
>>>>  Hi there,
>>>>
>>>>  I am using itkmattesMutualInformationImagetoImageMetric, and I am
>>>>wondering if the mutual information value approaching toward 0 (it's
>>>>always negative), does that mean a better regirtration transformation?
>>>>
>>>>  Thanx!
>>>>
>>>>  Grace
>>>>
>>>>
>>>>        
>>>>
>>>------------------------------------------------------------------------
>>>      
>>>
>>>>_______________________________________________
>>>>Insight-users mailing list
>>>>Insight-users at itk.org
>>>>http://www.itk.org/mailman/listinfo/insight-users
>>>>        
>>>>
>>>_______________________________________________
>>>Insight-users mailing list
>>>Insight-users at itk.org
>>>http://www.itk.org/mailman/listinfo/insight-users
>>>
>>>      
>>>
>
>_______________________________________________
>Insight-users mailing list
>Insight-users at itk.org
>http://www.itk.org/mailman/listinfo/insight-users
>
>  
>