ࡱ > B D = > ? @ A ` jbjb S |c |c \ 8 \ <
" * * * * * * ,
R D
6 * * * * * * * * * * D 0 , * T O (
0 <
z z \ \ $ $ \ \ Tiffany Beres
April 19, 2005
Viewing fMRI images
CS24: Field Writing Assignment
Functional magnetic resonance imaging (fMRI) is a modern noninvasive imaging technology for mapping the braindetermining which parts of the brain are activated by different kinds of stimuli, e.g. physical sensation or activity, visual stimulation, sound or the movement of a subject's fingers. In contrast to conventional MRI scans showing brain structure alone; fMRI provides information about brain function based on measured blood/oxygen flow in activated areas of the brain. For example, Fig. 1 shows the activation of the visual cortex in nine different participants in response to the same visual stimuli. Such an image is the product of a complex fMRI-imaging process where temporal and spatial data from the MRI scanner generate raw input images, which are then mathematically transformed and statistically analyzed to reconstruct activation maps of the brain, generally denoted with colored blobs as shown in Fig. 1.
From the standpoint of computer graphics, fMRI images are very interesting because computers are used to visualize data that corresponds to radio waves formed from magnetic fields generated around a persons head. Although the 2D slices (cross-sections) seen in Fig. 1 are geometrically created using computer algorithms, like typical vector-base graphics, these images are better described as raster graphics, sampled from discrete data and converted into a density map, a reconstructed image of the brain that is formed by assigning a specific density or color based on the raw data. As these cross-sections are layered one on top of each other fMRI data can generate images like Fig. 2, a reconstructed 3D coordinate space where each pixel, or in this case voxel (Volumetric Pixels), is represented by x-, y- and z- coordinates. Thus, while the original data set is created by scanning a real human head, unlike geometry-based modeling programs, the fMRI images themselves are not rendered from shape primitives, but rather these images are a highly sophisticated synthesis of layers of sampled-based pixels, which also allows for the inter-mapping/direct-comparison of different scans and easy manipulation of colors and values.
Perceptually, looking at 2D fMRI images like those in Fig.1 is strange (to the visual system) because each is a flat representation of a 3D object not generally encountered in day-to-day life, i.e. each picture does not correspond to a typical optic array that replicates the light of an original scene. Instead, these representations capture a sample of picture primitives (marks), such as occluding contours (lines), and volumes (regions). Our visual systems try to interpret these images by segmenting them in two ways: grouping, or grouping elements into regions based on common elements (Gestalt laws of proximity, similarity, closure, and good continuation), and also by boundary formation, using local differences in brightness, color and texture to determine which areas are of equal importance. While these processes are useful in breaking down the flat image, we can also look at these contrast borders themselves to suggest how these images appear to us. The edges of monochrome areas of these images tend to be soft edges, minimizing contrast and lateral inhibition of the photoreceptors that sense such boundaries, such as our rods; the result explains why these areas look rather dull and ambiguous. Oppositely, the colored blobs, have hard edges and color contrast boundaries making these regions pop out (activation of edge-detecting cones). Thus, while these fMRI images do not have a point-to-point correspondence of brightness and color of the actual brain areas, this imaging process enables a perceptual highlighting as it were, in those areas that are under examination through contrast and color boundaries.
With regards to design and the visual display of information, each fMRI image is really a diagram that delivers information for the doctor or scientist to interpret. Both Fig. 1 and Fig.2 are essentially just information graphics conveying active regions, areas where there is a great difference between the average intensity in the task stimulus condition as compared to rest. These high correlations areas are then delineated using an arbitrary color scale, where monochrome represents a resting state, and colored blobs represents activation from red to yellow, with the later representing the greatest amount of activation. As graphics it is important to note that there are many variables to contrast within each image: the scale, value, line thickness, shapes, spaces and as mentioned previously the selected colors. While Fig. 1 and Fig. 2 focus on the color and value information, it is possible to imagine computerized representations where the data is manipulated along any of these other contrast variables. Still, along the lines of the perceptual interpretations mentioned earlier, the heuristic rules of design (CRAP), also relevant here, suggest that these images are well designed to convey information. For example, the color regions greatly contrasts with the monochromatic surround, the repetitious color palate seen in all nine scans again draws attention to particular patterns and areas of interest, and finally, the proximal information, conveys information about the brain anatomy because of the relative positioning of each shape along the coordinate axis.
Together, each of these perspectives contributes a layer of understanding and appreciation to these fMRI images. Taken out of the realm of illusionist aims and by focusing on the computer-based/technological, perceptual and information-graphical nature of these images we are able to see how informative these images really are. While these brain-maps are perhaps not the most intuitive images to interpret and understand, they succeed at doing their job: organizing information to highlight particular differences in a way that can be easily manipulated in both 2D and 3D space, as well as along the dimension of timea significant tool for better understanding the cognitive functions of our human brains.
Figure 1 - Nine Horizonal Cross-Sections Obtained Using fMRI during a Visual Task
Figure 2- Three-dimensional View of the Activated Brain Region from Visual Task Above
PAGE
PAGE 2
1 7 & y z ſſſŴſ h[X 0J mH nH u
h[X 0J j h[X 0J Uj h[X Uj h[X Uh7 h[X h[X h( h[X h1@ h[X 5h1@ h[X 1 R S + ] # $ & y { | &`#$gd[X gd[X
dh `gd[X dh gd[X dh `gd[X `gd[X $a$gd[X $a$gd[X gd[X &`#$gd[X # 0 / =!"#$% " XX ujVg( H H ( d h
com.apple.print.PageFormat.FormattingPrinter
com.apple.print.ticket.creator
com.apple.printingmanager
com.apple.print.ticket.itemArray
com.apple.print.PageFormat.FormattingPrinter
Lexmark_Z55
com.apple.print.ticket.client
com.apple.printingmanager
com.apple.print.ticket.modDate
2005-02-09T20:12:35Z
com.apple.print.ticket.stateFlag
0
com.apple.print.PageFormat.PMHorizontalRes
com.apple.print.ticket.creator
com.apple.printingmanager
com.apple.print.ticket.itemArray
com.apple.print.PageFormat.PMHorizontalRes
600
com.apple.print.ticket.client
com.apple.printingmanager
com.apple.print.ticket.modDate
2005-02-09T20:12:30Z
com.apple.print.ticket.stateFlag
0
com.apple.print.PageFormat.PMOrientation
com.apple.print.ticket.creator
com.apple.printingmanager
com.apple.print.ticket.itemArray
com.apple.print.PageFormat.PMOrientation
1
com.apple.print.ticket.client
com.apple.printingmanager
com.apple.print.ticket.modDate
2005-02-09T20:12:35Z
com.apple.print.ticket.stateFlag
0
com.apple.print.PageFormat.PMScaling
com.apple.print.ticket.creator
com.apple.printingmanager
com.apple.print.ticket.itemArray
com.apple.print.PageFormat.PMScaling
1
com.apple.print.ticket.client
com.apple.printingmanager
com.apple.print.ticket.modDate
2005-02-09T20:12:35Z
com.apple.print.ticket.stateFlag
0
com.apple.print.PageFormat.PMVerticalRes
com.apple.print.ticket.creator
com.apple.printingmanager
com.apple.print.ticket.itemArray
com.apple.print.PageFormat.PMVerticalRes
600
com.apple.print.ticket.client
com.apple.printingmanager
com.apple.print.ticket.modDate
2005-02-09T20:12:30Z
com.apple.print.ticket.stateFlag
0
com.apple.print.PageFormat.PMVerticalScaling
com.apple.print.ticket.creator
com.apple.printingmanager
com.apple.print.ticket.itemArray
com.apple.print.PageFormat.PMVerticalScaling
1
com.apple.print.ticket.client
com.apple.printingmanager
com.apple.print.ticket.modDate
2005-02-09T20:12:35Z
com.apple.print.ticket.stateFlag
0
com.apple.print.subTicket.paper_info_ticket
com.apple.print.PageFormat.PMAdjustedPageRect
com.apple.print.ticket.creator
com.apple.printingmanager
com.apple.print.ticket.itemArray
com.apple.print.PageFormat.PMAdjustedPageRect
0.0
0.0
6260.0000000000009
4800
com.apple.print.ticket.client
com.apple.printingmanager
com.apple.print.ticket.modDate
2005-02-09T20:12:35Z
com.apple.print.ticket.stateFlag
0
com.apple.print.PageFormat.PMAdjustedPaperRect
com.apple.print.ticket.creator
com.apple.printingmanager
com.apple.print.ticket.itemArray
com.apple.print.PageFormat.PMAdjustedPaperRect
-40
-150
6560.0000000000009
4950
com.apple.print.ticket.client
com.apple.printingmanager
com.apple.print.ticket.modDate
2005-02-09T20:12:35Z
com.apple.print.ticket.stateFlag
0
com.apple.print.PaperInfo.PMPaperName
com.apple.print.ticket.creator
com.lexmark.print.z55.pm
com.apple.print.ticket.itemArray
com.apple.print.PaperInfo.PMPaperName
na-letter
com.apple.print.ticket.client
com.lexmark.print.z55.pm
com.apple.print.ticket.modDate
2003-11-25T00:39:40Z
com.apple.print.ticket.stateFlag
1
com.apple.print.PaperInfo.PMUnadjustedPageRect
com.apple.print.ticket.creator
com.lexmark.print.z55.pm
com.apple.print.ticket.itemArray
com.apple.print.PaperInfo.PMUnadjustedPageRect
0.0
0.0
751.20000000000005
576
com.apple.print.ticket.client
com.lexmark.print.z55.pm
com.apple.print.ticket.modDate
2003-11-25T00:39:40Z
com.apple.print.ticket.stateFlag
1
com.apple.print.PaperInfo.PMUnadjustedPaperRect
com.apple.print.ticket.creator
com.lexmark.print.z55.pm
com.apple.print.ticket.itemArray
com.apple.print.PaperInfo.PMUnadjustedPaperRect
-4.7999999999999998
-18
787.20000000000005
594
com.apple.print.ticket.client
com.lexmark.print.z55.pm
com.apple.print.ticket.modDate
2003-11-25T00:39:40Z
com.apple.print.ticket.stateFlag
1
com.apple.print.ticket.APIVersion
00.20
com.apple.print.ticket.privateLock
com.apple.print.ticket.type
com.apple.print.PaperInfoTicket
com.lexmark.print.PaperInfo.PaperID
com.apple.print.ticket.creator
com.lexmark.print.z55.pm
com.apple.print.ticket.itemArray
com.apple.print.ticket.client
com.lexmark.print.z55.pm
com.apple.print.ticket.modDate
2003-11-25T00:39:40Z
com.apple.print.ticket.stateFlag
1
com.lexmark.print.PaperInfo.PaperID
0
com.lexmark.print.PaperInfo.PaperType
com.apple.print.ticket.creator
com.lexmark.print.z55.pm
com.apple.print.ticket.itemArray
com.apple.print.ticket.client
com.lexmark.print.z55.pm
com.apple.print.ticket.modDate
2003-11-25T00:39:40Z
com.apple.print.ticket.stateFlag
1
com.lexmark.print.PaperInfo.PaperType
1
com.apple.print.ticket.APIVersion
00.20
com.apple.print.ticket.privateLock
com.apple.print.ticket.type
com.apple.print.PageFormatTicket
D d =d7 0
# A B X63 D T X63R[
X e xeǕYNU<5E1ь1G A\HnQ$#
h|hԕ_.gU&.7E۹߷XXiZ4i&]Ҥɲ&;4B0 ipZs\j}6W5w.w-:^'BSOÜƘ4
{z9[l-'-]9yvaDz@~>MY_Tւ\z8,('K/!eI-Фb)/>scҤr h)p敕&=$%#;Ҥ_ EPϵs?mke&nu=+,_sfpqڦZ)M:0CC,t4i61K|I_ɐVӰPiI=i^~߱rXX(o/ýi]4~({Gs}.yw o;)LYi5¯oҲiX^wXi҃YiRʽ24鯽5ӤSӤɃ[ p5VtiI^;MjZD+Γ՟NRk^J8[m֛g)27e
Y,Uڷ&배hYbir7Vlr9Ju՛]Y&-|:8@&Umn':,̝r"зϽ4ʹQv2ҤQYiRdIfYiR_V`KA45ѦIy"b\=Ҙ^CEaa?FӤkiJͪd'NyzbaSca,c?~fZ(M,X>=7%6G&sL]5,|J.lew4B9f\BjrϱP;boY2k9 z9t٤{[V`;i.&QYi҃Y:rӤ3*+Mz0dK6]~Wtv͉h9wEaaAW2LbVte`;bYi.,6bk)Xرc6Y✮l\ns0{8[=gzhsO*N۩%vr@ ԲUzI7o=U7:u2M..'|o'NO_I_]vc{p1w.٥nYnOmГ\0cDynЭ"|UpUK_#c/~
_|.5æl^sp`|а]p'zD@ /P'ǁIeMs,s܀b[`pJMKP֜_;s_Ҥ].kҤS7CU+M<{RD聰Ҥ'Ѥ]<Ωݴv,"?@0)*>o%D>O|7OSOqL1sOJrS+sLaYhS sV/s7P_O>R?c
D ҫ9'NMlmݱП6
'zr'>ˍvoK/Kg?+` ?Q<= 4 W"_91u~7/;ca4N&MrK}4G6+MZ|}6Sş.iؽe