|
Project 3: Feature Detection
Data Output (Blobs.xml file)
The Blobs.xml will contain in the following XML formatted information an example shown below.. Note the items in red do not appear in the file and are only there for description. NOTE: you are encouraged to use better/more descriptive names than info1, info2, etc. For example if you feature is the shape of the boundary of your blob you could use the tag <shape> rather than <info1>
<CancerAnalysis>
<image filename="XXX"> XXX here is the input filename
<width>WWW</width> WWW is width in pixels
<height>HHH</height> HHH is height in pixels
</image>
<numBlobs>NNN</numBlobs> NNN is number of blobs detected
NOW you will have NNN instances of <Blob>...</Blob> for the NNN blobs found
<Blob>
<id> 1 </id> give a unique id - I suggest starting from number 1
<location>
<start> row,column</start> will be an actual number for row and column
<end> row,column </end> will be an actual number for row and column
<center> row, column</center> will be an actual number for row and column
</location>
<area>
<upperleft> row,column</upperleft> will be an actual number for row and column
<lowerright> row,column</lowerright> will be an actual number for row and column
<width> value</width> width in terms of number of pixels across
<height>value</height>
<pixels>
<pixel> row,column</pixel>
<pixel> row,column</pixel>
<pixel> row,column</pixel>
........KEEP listing for all pixels in blob
</pixels>
</area>
<feature1>
<name> descriptive name of feature</name>
<location>row,colum</location> row,column is location (center or start) of this feature
<area> ZZZ</area> either bounding box information or list of pixels
that belong to this feature
<certainty> PPP </certainty> PPP=0 to 100% , certainty of this feature
<importance> SSS</importance> SSS=0 to 1 , level of importance.
<info1> XXXXXX</info1> make up XML describing YOUR feature
<info2> XXXXXX</info2> make up XML describing YOUR feature
<info3> XXXXXX</info3> make up XML describing YOUR feature
YOU may name 5 or 1 or 20 pieces of info describing YOUR feature1
</feature1>
<feature2>
<name> descriptive name of feature</name>
<location>row,colum</location> row,column is location (center or start) of this feature
<area> ZZZ</area> either bounding box information or list of pixels
that belong to this feature
<certainty> PPP </certainty> PPP=0 to 100% , certainty of this feature
<importance> SSS</importance> SSS=0 to 1 , level of importance.
<info1> XXXXXX</info1> make up XML describing YOUR feature
<info2> XXXXXX</info2> make up XML describing YOUR feature
<info3> XXXXXX</info3> make up XML describing YOUR feature
YOU may name 5 or 1 or 20 pieces of info describing YOUR feature2
</feature2>
<feature3>
<name> descriptive name of feature</name>
<location>row,colum</location> row,column is location (center or start) of this feature
<area> ZZZ</area> either bounding box information or list of pixels
that belong to this feature
<certainty> PPP </certainty> PPP=0 to 100% , certainty of this feature
<importance> SSS</importance> SSS=0 to 1 , level of importance.
<info1> XXXXXX</info1> make up XML describing YOUR feature
<info2> XXXXXX</info2> make up XML describing YOUR feature
<info3> XXXXXX</info3> make up XML describing YOUR feature
YOU may name 5 or 1 or 20 pieces of info describing YOUR feature3
</feature3>
<feature4>
<name> descriptive name of feature</name>
<location>row,colum</location> row,column is location (center or start) of this feature
<area> ZZZ</area> either bounding box information or list of pixels
that belong to this feature
<certainty> PPP </certainty> PPP=0 to 100% , certainty of this feature
<importance> SSS</importance> SSS=0 to 1 , level of importance.
<info1> XXXXXX</info1> make up XML describing YOUR feature
<info2> XXXXXX</info2> make up XML describing YOUR feature
<info3> XXXXXX</info3> make up XML describing YOUR feature
YOU may name 5 or 1 or 20 pieces of info describing YOUR feature4
</feature4>
<feature5>
<name> descriptive name of feature</name>
<location>row,colum</location> row,column is location (center or start) of this feature
<area> ZZZ</area> either bounding box information or list of pixels
that belong to this feature
<certainty> PPP </certainty> PPP=0 to 100% , certainty of this feature
<importance> SSS</importance> SSS=0 to 1 , level of importance.
<info1> XXXXXX</info1> make up XML describing YOUR feature
<info2> XXXXXX</info2> make up XML describing YOUR feature
<info3> XXXXXX</info3> make up XML describing YOUR feature
YOU may name 5 or 1 or 20 pieces of info describing YOUR feature5
</feature5
</Blob> |
...........REPEAT for the other NNN-4 BLOBS !!!!! ...............
</CancerAnalysis> |
.
Note that when calculating each feature, you will not only specify
it but, also give an certainty metric on a scale of 0 to 100% indicating
how true this feature is. Finally, for each feature you will give a metric on a scale of 0 to 1 indicating the importance of this feature. There is no right or wrong way to calculate certainty and importance but, should make some common sense for your feature.
- Feature Certainty = this metric should be calcuated automatically by the program and a function of the certainty of the pixels used in its calculation. The certainty of a pixel represents the certainty that a pixel belongs to the tumor under consideration instead of belonging to the non-tumor or background part of the ROI. Think creatively about how you could come up with certainty (i.e. the variance of a metric being smaller could be said to make it more certain).
- Feature Importance = this metric may be calculated automatically but, when this is not feasible can be determined by a human operator. In the later case, the feature, its location, its value should be given in a pop-up box and a prompt for a "feature importance" value from 0 to 1 requested of the operator as their input. You can be creative and rule based in saying something is important (i.e.as an example consider looking at the shape of the boundary of the mass as a feature if certainty > 80 and shape feature value > 2 then is very strong indicator of this cancer thus important is high = 1.0).
|