Break plot up into block intervals in different colours

Hi, 

I have some EEG data. The 1st 45seconds of data is rest data, followed by some
stimulation data for 10 seconds. Again another 45 seconds of rest and another 10
seconds of stimulation data.

I want to have the first 45seconds of the plot in white and the next 10 seconds
in green to show the rest-to-stimulation transition (and again another 45seconds
of white and 10 seconds of white). I have seen this displayed in some papers but
i have no idea how to do it. 

e.g.
data with white backgrounddata with green background
	...where  is the transition between the rest and stimulation data
	Any help appreciated,
ChrisJ.
	

Hi Chris,

For once I'd say use MS Excel rather than Matlab. It's a piece of
cake.  The
graphics capability of Excel is often overlooked.

Cheers,

Jeff
	-----Original Message-----
From: matlab@matl... [mailto:matlab@matl...]On Behalf Of
chrissoraghan@chri...
Sent: 08 March 2006 17:16
To: matlab@matl...
Subject: [matlab] Break plot up into block intervals in different
colours
	Hi,

I have some EEG data. The 1st 45seconds of data is rest data, followed by
some stimulation data for 10 seconds. Again another 45 seconds of rest and
another 10 seconds of stimulation data.

I want to have the first 45seconds of the plot in white and the next 10
seconds in green to show the rest-to-stimulation transition (and again
another 45seconds of white and 10 seconds of white). I have seen this
displayed in some papers but i have no idea how to do it.

e.g.
data with white backgrounddata with green background
	...where  is the transition between the rest and stimulation data
	Any help appreciated,
ChrisJ.
	

Hi Jeff,

The Excel idea will work but i would rather try and keep it within Matlab.
There will be many tests like this and the protocols for each will not
always adhere to the same rest and stimulation periods of 45seconds and
10seconds as mentioned previously. I have to perform some pre and post
processing to the noisy data and apply other algorithms to SOME of the
results so I would like to find a way to control the graphics of matlab as
mentioned below and be able to change the lenght of time for the different
states, rest stimulation, semi stimulation. Really im looking for a way to
section off different states in the plot with colour ID, of green and white
as mentioned below. If i cannot get the info on how to do it I will use
Excel for the time being but it would be better to have a function m-file to
adjust the graphics, as with 'bar.m' etc. Thanks for the help Jeff,

ChrisJ
	On 3/9/06, Jeff Winter <jeff.winter@jeff...> wrote:
>
> Hi Chris,
>
> For once I'd say use MS Excel rather than Matlab. It's a piece of
> cake.  The
> graphics capability of Excel is often overlooked.
>
> Cheers,
>
> Jeff
>
>
> -----Original Message-----
> From: matlab@matl... [mailto:matlab@matl...]On Behalf Of
> chrissoraghan@chri...
> Sent: 08 March 2006 17:16
> To: matlab@matl...
> Subject: [matlab] Break plot up into block intervals in different
> colours
>
>
> Hi,
>
> I have some EEG data. The 1st 45seconds of data is rest data, followed by
> some stimulation data for 10 seconds. Again another 45 seconds of rest and
> another 10 seconds of stimulation data.
>
> I want to have the first 45seconds of the plot in white and the next 10
> seconds in green to show the rest-to-stimulation transition (and again
> another 45seconds of white and 10 seconds of white). I have seen this
> displayed in some papers but i have no idea how to do it.
>
> e.g.
> data with white backgrounddata with green background
>
>
> ...where  is the transition between the rest and stimulation data
>
>
> Any help appreciated,
> ChrisJ.
>
	

-----Original Message-----
From: Christopher Soraghan [mailto:c...@gmail.com]
Sent: 09 March 2006 17:39
To: j...@aeroflex.com
Cc: Subject: Re: [matlab] Break plot up into block intervals in different colours

Hi Jeff,

 

The Excel idea will work but i would rather try and keep it within Matlab. There will be many tests like this and the protocols for each will not always adhere to the same rest and stimulation periods of 45seconds and 10seconds as mentioned previously. I have to perform some pre and post processing to the noisy data and apply other algorithms to SOME of the results so I would like to find a way to control the graphics of matlab as mentioned below and be able to change the lenght of time for the different states, rest stimulation, semi stimulation. Really im looking for a way to section off different states in the plot with colour ID, of green and white as mentioned below. If i cannot get the info on how to do it I will use Excel for the time being but it would be better to have a function m-file to adjust the graphics, as with ' bar.m' etc. Thanks for the help Jeff,

 

ChrisJ

 

 

On 3/9/06, Jeff Winter <j...@aeroflex.com> wrote:

Hi Chris,

For once I'd say use MS Excel rather than Matlab. It's a piece of cake.  The
graphics capability of Excel is often overlooked.

Cheers,

Jeff-----Original Message-----
From: m...@yahoogroups.com [mailto:m...@yahoogroups.com]On Behalf Of
c...@gmail.com
Sent: 08 March 2006 17:16
To: m...@yahoogroups.com
Subject: [matlab] Break plot up into block intervals in different
coloursHi,

I have some EEG data. The 1st 45seconds of data is rest data, followed by
some stimulation data for 10 seconds. Again another 45 seconds of rest and
another 10 seconds of stimulation data.

I want to have the first 45seconds of the plot in white and the next 10
seconds in green to show the rest-to-stimulation transition (and again
another 45seconds of white and 10 seconds of white). I have seen this
displayed in some papers but i have no idea how to do it.

e.g.
data with white backgrounddata with green background...where is the transition between the rest and stimulation dataAny help appreciated,
ChrisJ.

<*> To visit your group on the web, go to:
   http://groups.yahoo.com/group/matlab/

<*> To unsubscribe from this group, send an email to:
   m...@yahoogroups.com

<*> ">http://docs.yahoo.com/info/terms/

Hi Jeff,
	What a quick reply and what a good advice, thanks a million! This is what I
will use,
	Cheers,

ChrisJ
	On 3/9/06, Jeff Winter <jeff.winter@jeff...> wrote:
>
>  Hi Chris,
>
> One way of plotting the points in a different colour is to create a number
> of arrays (equal to the number of transitions) that are of the same length
> as the total data you want to plot.
>
> If your data is: rest1  stim1  rest2  stim2 which for arguments sake is
> 110 samples long, then the aim is to have 4 arrays (of the same length so
> that plot doesn't moan) that contain some of the data:
> rest1  -----  -----  -----
> -----  stim1  -----  -----
> -----  -----  rest2  -----
> -----  -----  -----  stim2
>
> To plot each of these segments you'll need 4 zero'd arrays the
same length
> as your data:
> array1 = zeros(1,110); array2 = array1 ; array1 ; array4 = array1 ;
>
> Then all you need to do is copy the correct part of each data segment into
> one of the new arrays. However, you will want to offset the start point of
> your data entry into each new array. e.g.:
> If rest1 is 45 samples long, then array1(1:45) = data(1:45);
> If stim1 is 10 samples long, then array2(46:55) = data(46:55);
> If rest2 is 45 samples long, then array2(56:100) = data(56:100);
> If stim2 is 10 samples long, then array2(101:110) = data(101:110);
>
> Now you should be able to plot all four arrays of any colour you want on
> top of each other so that each transition is in a different colour.
> This does assume that you know the sample rate the data was recorded at,
> so that you can work out what the array indexing should be.
>
> The down side to this is that the zeros will also be plotted.  As Matlab
> plots all the points in a standard plot this will make the intercept points
> on your graph look odd.  However, you could plot the points in different
> colours and symbols using this technique on top of a grey line for example.
>
> In any case this is one way to segment your data.
>
> Cheers,
>
> Jeff
>
> -----Original Message-----
> *From:* Christopher Soraghan [mailto:chrissoraghan@chri...]
> *Sent:* 09 March 2006 17:39
> *To:* jeff.winter@jeff...
> *Cc:* *Subject:* Re: [matlab] Break plot up into block intervals in
> different colours
>
> Hi Jeff,
>
> The Excel idea will work but i would rather try and keep it within Matlab.
> There will be many tests like this and the protocols for each will not
> always adhere to the same rest and stimulation periods of 45seconds and
> 10seconds as mentioned previously. I have to perform some pre and post
> processing to the noisy data and apply other algorithms to SOME of the
> results so I would like to find a way to control the graphics of matlab as
> mentioned below and be able to change the lenght of time for the different
> states, rest stimulation, semi stimulation. Really im looking for a way to
> section off different states in the plot with colour ID, of green and white
> as mentioned below. If i cannot get the info on how to do it I will use
> Excel for the time being but it would be better to have a function m-file
to
> adjust the graphics, as with ' bar.m' etc. Thanks for the help
Jeff,
>
> ChrisJ
>
>
>
> On 3/9/06, Jeff Winter <jeff.winter@jeff...> wrote:
> >
> > Hi Chris,
> >
> > For once I'd say use MS Excel rather than Matlab. It's a
piece of
> > cake.  The
> > graphics capability of Excel is often overlooked.
> >
> > Cheers,
> >
> > Jeff
> >
> >
> > -----Original Message-----
> > From: matlab@matl... [mailto:matlab@matl...]On Behalf Of
> >
> > chrissoraghan@chri...
> > Sent: 08 March 2006 17:16
> > To: matlab@matl...
> > Subject: [matlab] Break plot up into block intervals in different
> > colours
> >
> >
> > Hi,
> >
> > I have some EEG data. The 1st 45seconds of data is rest data, followed
> > by
> > some stimulation data for 10 seconds. Again another 45 seconds of rest
> > and
> > another 10 seconds of stimulation data.
> >
> > I want to have the first 45seconds of the plot in white and the next
10
> > seconds in green to show the rest-to-stimulation transition (and again
> > another 45seconds of white and 10 seconds of white). I have seen this
> > displayed in some papers but i have no idea how to do it.
> >
> > e.g.
> > data with white backgrounddata with green background
> >
> >
> > ...where  is the transition between the rest and stimulation data
> >
> >
> > Any help appreciated,
> > ChrisJ.
> >
	

You can also try using the 'patch' function. Say for example the
data is stored as  a vector, data = [segment45 segment10 ... ]. You plot this
out and then use the patch function in a for loop over the number of segments,
in every iteration of the for loop, use patch([x1 x2 x2 x1],[y1 y1 y2 y2],[0 1
0]) for segment45 where (x1,y1),(x2,y2),(x1,y2) and (x2,y1) are the end points
of rectangular patch and the end parameter is for green color. To plot the patch
in the background so you can see the plot, use a =
get(gca,'child');set(gca,'child',flipud(a)). Similarly plot
a patch for segment10.
   
  Premal

Jeff Winter <jeff.winter@jeff...> wrote:
      Hi Chris,
   
  One way of plotting the points in a different colour is to create a number of
arrays (equal to the number of transitions) that are of the same length as the
total data you want to plot.
   
  If your data is: rest1  stim1  rest2  stim2 which for arguments sake is 110
samples long, then the aim is to have 4 arrays (of the same length so that plot
doesn't moan) that contain some of the data:
  rest1  -----  -----  ----- 
  -----  stim1  -----  -----
  -----  -----  rest2  ----- 
  -----  -----  -----  stim2
   
  To plot each of these segments you'll need 4 zero'd arrays the same
length as your data:
  array1 = zeros(1,110); array2 = array1 ; array1 ; array4 = array1 ; 
   
  Then all you need to do is copy the correct part of each data segment into one
of the new arrays. However, you will want to offset the start point of your data
entry into each new array. e.g.:
  If rest1 is 45 samples long, then array1(1:45) = data(1:45);
  If stim1 is 10 samples long, then array2(46:55) = data(46:55);
  If rest2 is 45 samples long, then array2(56:100) = data(56:100);
  If stim2 is 10 samples long, then array2(101:110) = data(101:110);
   
  Now you should be able to plot all four arrays of any colour you want on top
of each other so that each transition is in a different colour.
  This does assume that you know the sample rate the data was recorded at, so
that you can work out what the array indexing should be. 
   
  The down side to this is that the zeros will also be plotted.  As Matlab plots
all the points in a standard plot this will make the intercept points on your
graph look odd.  However, you could plot the points in different colours and
symbols using this technique on top of a grey line for example.
   
  In any case this is one way to segment your data.
   
  Cheers,
   
  Jeff
    -----Original Message-----
From: Christopher Soraghan [mailto:chrissoraghan@chri...]
Sent: 09 March 2006 17:39
To: jeff.winter@jeff...
Cc: Subject: Re: [matlab] Break plot up into block intervals in different
colours
	  Hi Jeff, 
   
  The Excel idea will work but i would rather try and keep it within Matlab.
There will be many tests like this and the protocols for each will not always
adhere to the same rest and stimulation periods of 45seconds and 10seconds as
mentioned previously. I have to perform some pre and post processing to the
noisy data and apply other algorithms to SOME of the results so I would like to
find a way to control the graphics of matlab as mentioned below and be able to
change the lenght of time for the different states, rest stimulation, semi
stimulation. Really im looking for a way to section off different states in the
plot with colour ID, of green and white as mentioned below. If i cannot get the
info on how to do it I will use Excel for the time being but it would be better
to have a function m-file to adjust the graphics, as with ' bar.m'
etc. Thanks for the help Jeff,
   
  ChrisJ
	  On 3/9/06, Jeff Winter <jeff.winter@jeff...> wrote:   Hi Chris,

For once I'd say use MS Excel rather than Matlab. It's a piece of
cake.  The
graphics capability of Excel is often overlooked. 

Cheers,

Jeff
	-----Original Message-----
From: matlab@matl... [mailto:matlab@matl...]On Behalf Of 
chrissoraghan@chri...
Sent: 08 March 2006 17:16
To: matlab@matl...
Subject: [matlab] Break plot up into block intervals in different 
colours
	Hi,

I have some EEG data. The 1st 45seconds of data is rest data, followed by
some stimulation data for 10 seconds. Again another 45 seconds of rest and
another 10 seconds of stimulation data. 

I want to have the first 45seconds of the plot in white and the next 10
seconds in green to show the rest-to-stimulation transition (and again
another 45seconds of white and 10 seconds of white). I have seen this 
displayed in some papers but i have no idea how to do it.

e.g.
data with white backgrounddata with green background
	...where  is the transition between the rest and stimulation data
	Any help appreciated,
ChrisJ.
	

Well, If I understand your correctly, all you will do
simply should be something like: 

----
plot(1:45, eeg(1:45));
hold on;
plot(45:100, eeg(45:100),'r')
hold on;
plot(100:150, eeg(100:150),'g')
-----

the indices will definitely change depending on the
sampling frequency. hope it helps.
	tolga esat ozkurt

--- Jeff Winter <jeff.winter@jeff...> wrote:

> Hi Chris,
> 
> One way of plotting the points in a different colour
> is to create a number
> of arrays (equal to the number of transitions) that
> are of the same length
> as the total data you want to plot.
> 
> If your data is: rest1  stim1  rest2  stim2 which
> for arguments sake is
> 110 samples long, then the aim is to have 4 arrays
> (of the same length so
> that plot doesn't moan) that contain some of the
> data:
> rest1  -----  -----  -----
> -----  stim1  -----  -----
> -----  -----  rest2  -----
> -----  -----  -----  stim2
> 
> To plot each of these segments you'll need 4 zero'd
> arrays the same length
> as your data:
> array1 = zeros(1,110); array2 = array1 ; array1 ;
> array4 = array1 ;
> 
> Then all you need to do is copy the correct part of
> each data segment into
> one of the new arrays. However, you will want to
> offset the start point of
> your data entry into each new array. e.g.:
> If rest1 is 45 samples long, then array1(1:45) > data(1:45);
> If stim1 is 10 samples long, then array2(46:55) > data(46:55);
> If rest2 is 45 samples long, then array2(56:100) > data(56:100);
> If stim2 is 10 samples long, then array2(101:110) > data(101:110);
> 
> Now you should be able to plot all four arrays of
> any colour you want on top
> of each other so that each transition is in a
> different colour.
> This does assume that you know the sample rate the
> data was recorded at, so
> that you can work out what the array indexing should
> be.
> 
> The down side to this is that the zeros will also be
> plotted.  As Matlab
> plots all the points in a standard plot this will
> make the intercept points
> on your graph look odd.  However, you could plot the
> points in different
> colours and symbols using this technique on top of a
> grey line for example.
> 
> In any case this is one way to segment your data.
> 
> Cheers,
> 
> Jeff
>   -----Original Message-----
>   From: Christopher Soraghan
> [mailto:chrissoraghan@chri...]
>   Sent: 09 March 2006 17:39
>   To: jeff.winter@jeff...
>   Cc: Subject: Re: [matlab] Break plot up into block
> intervals in different
> colours
> 
> 
>   Hi Jeff,
> 
>   The Excel idea will work but i would rather try
> and keep it within Matlab.
> There will be many tests like this and the protocols
> for each will not
> always adhere to the same rest and stimulation
> periods of 45seconds and
> 10seconds as mentioned previously. I have to perform
> some pre and post
> processing to the noisy data and apply other
> algorithms to SOME of the
> results so I would like to find a way to control the
> graphics of matlab as
> mentioned below and be able to change the lenght of
> time for the different
> states, rest stimulation, semi stimulation. Really
> im looking for a way to
> section off different states in the plot with colour
> ID, of green and white
> as mentioned below. If i cannot get the info on how
> to do it I will use
> Excel for the time being but it would be better to
> have a function m-file to
> adjust the graphics, as with ' bar.m' etc. Thanks
> for the help Jeff,
> 
>   ChrisJ
> 
> 
> 
>   On 3/9/06, Jeff Winter <jeff.winter@jeff...>
> wrote:
>     Hi Chris,
> 
>     For once I'd say use MS Excel rather than
> Matlab. It's a piece of cake.
> The
>     graphics capability of Excel is often
> overlooked.
> 
>     Cheers,
> 
>     Jeff
> 
> 
>     -----Original Message-----
>     From: matlab@matl...
> [mailto:matlab@matl...]On Behalf Of
>     chrissoraghan@chri...
>     Sent: 08 March 2006 17:16
>     To: matlab@matl...
>     Subject: [matlab] Break plot up into block
> intervals in different
>     colours
> 
> 
>     Hi,
> 
>     I have some EEG data. The 1st 45seconds of data
> is rest data, followed
> by
>     some stimulation data for 10 seconds. Again
> another 45 seconds of rest
> and
>     another 10 seconds of stimulation data.
> 
>     I want to have the first 45seconds of the plot
> in white and the next 10
>     seconds in green to show the rest-to-stimulation
> transition (and again
>     another 45seconds of white and 10 seconds of
> white). I have seen this
>     displayed in some papers but i have no idea how
> to do it.
> 
>     e.g.
>     data with white backgrounddata with green
> background
> 
> 
>     ...where  is the transition between the rest
> and stimulation data
> 
> 
>     Any help appreciated,
>     ChrisJ.
>
	

Hi Chris,
Try using 'hold on'.
Say,your data is in the arrays rest1  stim1  rest2  stim2 .. 

lr1 = length(rest1);
ls1 = length(stim1);
.... 

hold on;
plot(1:lr1,rest1,'r');
plot(lr1+1:lr1+ls1,stim1,'b');
.... 

regards,
Ravi
	Jeff Winter writes: 

> Hi Chris, 
> 
> One way of plotting the points in a different colour is to create a number
> of arrays (equal to the number of transitions) that are of the same length
> as the total data you want to plot. 
> 
> If your data is: rest1  stim1  rest2  stim2 which for arguments sake is
> 110 samples long, then the aim is to have 4 arrays (of the same length so
> that plot doesn't moan) that contain some of the data:
> rest1  -----  -----  -----
> -----  stim1  -----  -----
> -----  -----  rest2  -----
> -----  -----  -----  stim2 
> 
> To plot each of these segments you'll need 4 zero'd arrays the
same length
> as your data:
> array1 = zeros(1,110); array2 = array1 ; array1 ; array4 = array1 ; 
> 
> Then all you need to do is copy the correct part of each data segment into
> one of the new arrays. However, you will want to offset the start point of
> your data entry into each new array. e.g.:
> If rest1 is 45 samples long, then array1(1:45) = data(1:45);
> If stim1 is 10 samples long, then array2(46:55) = data(46:55);
> If rest2 is 45 samples long, then array2(56:100) = data(56:100);
> If stim2 is 10 samples long, then array2(101:110) = data(101:110); 
> 
> Now you should be able to plot all four arrays of any colour you want on
top
> of each other so that each transition is in a different colour.
> This does assume that you know the sample rate the data was recorded at, so
> that you can work out what the array indexing should be. 
> 
> The down side to this is that the zeros will also be plotted.  As Matlab
> plots all the points in a standard plot this will make the intercept points
> on your graph look odd.  However, you could plot the points in different
> colours and symbols using this technique on top of a grey line for example.

> 
> In any case this is one way to segment your data. 
> 
> Cheers, 
> 
> Jeff
>   -----Original Message-----
>   From: Christopher Soraghan [mailto:chrissoraghan@chri...]
>   Sent: 09 March 2006 17:39
>   To: jeff.winter@jeff...
>   Cc: Subject: Re: [matlab] Break plot up into block intervals in different
> colours 
> 
> 
>   Hi Jeff, 
> 
>   The Excel idea will work but i would rather try and keep it within
Matlab.
> There will be many tests like this and the protocols for each will not
> always adhere to the same rest and stimulation periods of 45seconds and
> 10seconds as mentioned previously. I have to perform some pre and post
> processing to the noisy data and apply other algorithms to SOME of the
> results so I would like to find a way to control the graphics of matlab as
> mentioned below and be able to change the lenght of time for the different
> states, rest stimulation, semi stimulation. Really im looking for a way to
> section off different states in the plot with colour ID, of green and white
> as mentioned below. If i cannot get the info on how to do it I will use
> Excel for the time being but it would be better to have a function m-file
to
> adjust the graphics, as with ' bar.m' etc. Thanks for the help
Jeff, 
> 
>   ChrisJ 
> 
>  
> 
>   On 3/9/06, Jeff Winter <jeff.winter@jeff...> wrote:
>     Hi Chris, 
> 
>     For once I'd say use MS Excel rather than Matlab. It's a
piece of cake.
> The
>     graphics capability of Excel is often overlooked. 
> 
>     Cheers, 
> 
>     Jeff 
> 
> 
>     -----Original Message-----
>     From: matlab@matl... [mailto:matlab@matl...]On Behalf Of
>     chrissoraghan@chri...
>     Sent: 08 March 2006 17:16
>     To: matlab@matl...
>     Subject: [matlab] Break plot up into block intervals in different
>     colours 
> 
> 
>     Hi, 
> 
>     I have some EEG data. The 1st 45seconds of data is rest data, followed
> by
>     some stimulation data for 10 seconds. Again another 45 seconds of rest
> and
>     another 10 seconds of stimulation data. 
> 
>     I want to have the first 45seconds of the plot in white and the next 10
>     seconds in green to show the rest-to-stimulation transition (and again
>     another 45seconds of white and 10 seconds of white). I have seen this
>     displayed in some papers but i have no idea how to do it. 
> 
>     e.g.
>     data with white backgrounddata with green background 
> 
> 
>     ...where  is the transition between the rest and stimulation data 
> 
> 
>     Any help appreciated,
>     ChrisJ. 
>
	

Tolga you're a star!

Thanks for pointing that out.

Regards,

Jeff

-----Original Message-----
From: matlab@matl... [mailto:matlab@matl...]On Behalf Of
Tolga kurt
Sent: 09 March 2006 19:52
To: matlab@matl...
Subject: RE: [matlab] Break plot up into block intervals in different
colours
	Well, If I understand your correctly, all you will do
simply should be something like:

----
plot(1:45, eeg(1:45));
hold on;
plot(45:100, eeg(45:100),'r')
hold on;
plot(100:150, eeg(100:150),'g')
-----

the indices will definitely change depending on the
sampling frequency. hope it helps.
	tolga esat ozkurt

--- Jeff Winter <jeff.winter@jeff...> wrote:

> Hi Chris,
>
> One way of plotting the points in a different colour
> is to create a number
> of arrays (equal to the number of transitions) that
> are of the same length
> as the total data you want to plot.
>
> If your data is: rest1  stim1  rest2  stim2 which
> for arguments sake is
> 110 samples long, then the aim is to have 4 arrays
> (of the same length so
> that plot doesn't moan) that contain some of the
> data:
> rest1  -----  -----  -----
> -----  stim1  -----  -----
> -----  -----  rest2  -----
> -----  -----  -----  stim2
>
> To plot each of these segments you'll need 4 zero'd
> arrays the same length
> as your data:
> array1 = zeros(1,110); array2 = array1 ; array1 ;
> array4 = array1 ;
>
> Then all you need to do is copy the correct part of
> each data segment into
> one of the new arrays. However, you will want to
> offset the start point of
> your data entry into each new array. e.g.:
> If rest1 is 45 samples long, then array1(1:45) > data(1:45);
> If stim1 is 10 samples long, then array2(46:55) > data(46:55);
> If rest2 is 45 samples long, then array2(56:100) > data(56:100);
> If stim2 is 10 samples long, then array2(101:110) > data(101:110);
>
> Now you should be able to plot all four arrays of
> any colour you want on top
> of each other so that each transition is in a
> different colour.
> This does assume that you know the sample rate the
> data was recorded at, so
> that you can work out what the array indexing should
> be.
>
> The down side to this is that the zeros will also be
> plotted.  As Matlab
> plots all the points in a standard plot this will
> make the intercept points
> on your graph look odd.  However, you could plot the
> points in different
> colours and symbols using this technique on top of a
> grey line for example.
>
> In any case this is one way to segment your data.
>
> Cheers,
>
> Jeff
>   -----Original Message-----
>   From: Christopher Soraghan
> [mailto:chrissoraghan@chri...]
>   Sent: 09 March 2006 17:39
>   To: jeff.winter@jeff...
>   Cc: Subject: Re: [matlab] Break plot up into block
> intervals in different
> colours
>
>
>   Hi Jeff,
>
>   The Excel idea will work but i would rather try
> and keep it within Matlab.
> There will be many tests like this and the protocols
> for each will not
> always adhere to the same rest and stimulation
> periods of 45seconds and
> 10seconds as mentioned previously. I have to perform
> some pre and post
> processing to the noisy data and apply other
> algorithms to SOME of the
> results so I would like to find a way to control the
> graphics of matlab as
> mentioned below and be able to change the lenght of
> time for the different
> states, rest stimulation, semi stimulation. Really
> im looking for a way to
> section off different states in the plot with colour
> ID, of green and white
> as mentioned below. If i cannot get the info on how
> to do it I will use
> Excel for the time being but it would be better to
> have a function m-file to
> adjust the graphics, as with ' bar.m' etc. Thanks
> for the help Jeff,
>
>   ChrisJ
>
>
>
>   On 3/9/06, Jeff Winter <jeff.winter@jeff...>
> wrote:
>     Hi Chris,
>
>     For once I'd say use MS Excel rather than
> Matlab. It's a piece of cake.
> The
>     graphics capability of Excel is often
> overlooked.
>
>     Cheers,
>
>     Jeff
>
>
>     -----Original Message-----
>     From: matlab@matl...
> [mailto:matlab@matl...]On Behalf Of
>     chrissoraghan@chri...
>     Sent: 08 March 2006 17:16
>     To: matlab@matl...
>     Subject: [matlab] Break plot up into block
> intervals in different
>     colours
>
>
>     Hi,
>
>     I have some EEG data. The 1st 45seconds of data
> is rest data, followed
> by
>     some stimulation data for 10 seconds. Again
> another 45 seconds of rest
> and
>     another 10 seconds of stimulation data.
>
>     I want to have the first 45seconds of the plot
> in white and the next 10
>     seconds in green to show the rest-to-stimulation
> transition (and again
>     another 45seconds of white and 10 seconds of
> white). I have seen this
>     displayed in some papers but i have no idea how
> to do it.
>
>     e.g.
>     data with white backgrounddata with green
> background
>
>
>     ...where  is the transition between the rest
> and stimulation data
>
>
>     Any help appreciated,
>     ChrisJ.
>
	

Thank you all for the help. The explanation by Tolga has been the simplest

and quickest but very efficient. Here is the code:

****

seg1_len = (1:(rest_per*fs));

seg2_len = (((rest_per)*fs):((rest_per+stim_per)*fs));

seg3_len = (((rest_per+stim_per)*fs):((rest_per*2+stim_per)*fs));

seg4_len = (((rest_per*2+stim_per)*fs):((rest_per*2+stim_per*2)*fs));

plot( seg1_len, oxy(seg1_len), 'b');

hold on

plot(seg2_len, oxy(seg2_len),'k');

plot(seg3_len, oxy(seg3_len), 'b');

plot(seg4_len, oxy(seg4_len), 'k');

****
Thanks again,
I second that, fair play Tolga, thanks too Jeff,

Chris.
	On 3/10/06, Jeff Winter <jeff.winter@jeff...> wrote:
>
> Tolga you're a star!
>
> Thanks for pointing that out.
>
> Regards,
>
> Jeff
>
> -----Original Message-----
> From: matlab@matl... [mailto:matlab@matl...]On Behalf Of
> Tolga kurt
> Sent: 09 March 2006 19:52
> To: matlab@matl...
> Subject: RE: [matlab] Break plot up into block intervals in different
> colours
>
>
> Well, If I understand your correctly, all you will do
> simply should be something like:
>
> ----
> plot(1:45, eeg(1:45));
> hold on;
> plot(45:100, eeg(45:100),'r')
> hold on;
> plot(100:150, eeg(100:150),'g')
> -----
>
> the indices will definitely change depending on the
> sampling frequency. hope it helps.
>
>
> tolga esat ozkurt
>
> --- Jeff Winter <jeff.winter@jeff...> wrote:
>
> > Hi Chris,
> >
> > One way of plotting the points in a different colour
> > is to create a number
> > of arrays (equal to the number of transitions) that
> > are of the same length
> > as the total data you want to plot.
> >
> > If your data is: rest1  stim1  rest2  stim2 which
> > for arguments sake is
> > 110 samples long, then the aim is to have 4 arrays
> > (of the same length so
> > that plot doesn't moan) that contain some of the
> > data:
> > rest1  -----  -----  -----
> > -----  stim1  -----  -----
> > -----  -----  rest2  -----
> > -----  -----  -----  stim2
> >
> > To plot each of these segments you'll need 4 zero'd
> > arrays the same length
> > as your data:
> > array1 = zeros(1,110); array2 = array1 ; array1 ;
> > array4 = array1 ;
> >
> > Then all you need to do is copy the correct part of
> > each data segment into
> > one of the new arrays. However, you will want to
> > offset the start point of
> > your data entry into each new array. e.g.:
> > If rest1 is 45 samples long, then array1(1:45) > > data(1:45);
> > If stim1 is 10 samples long, then array2(46:55) > > data(46:55);
> > If rest2 is 45 samples long, then array2(56:100) > >
data(56:100);
> > If stim2 is 10 samples long, then array2(101:110) > >
data(101:110);
> >
> > Now you should be able to plot all four arrays of
> > any colour you want on top
> > of each other so that each transition is in a
> > different colour.
> > This does assume that you know the sample rate the
> > data was recorded at, so
> > that you can work out what the array indexing should
> > be.
> >
> > The down side to this is that the zeros will also be
> > plotted.  As Matlab
> > plots all the points in a standard plot this will
> > make the intercept points
> > on your graph look odd.  However, you could plot the
> > points in different
> > colours and symbols using this technique on top of a
> > grey line for example.
> >
> > In any case this is one way to segment your data.
> >
> > Cheers,
> >
> > Jeff
> >   -----Original Message-----
> >   From: Christopher Soraghan
> > [mailto:chrissoraghan@chri...]
> >   Sent: 09 March 2006 17:39
> >   To: jeff.winter@jeff...
> >   Cc: Subject: Re: [matlab] Break plot up into block
> > intervals in different
> > colours
> >
> >
> >   Hi Jeff,
> >
> >   The Excel idea will work but i would rather try
> > and keep it within Matlab.
> > There will be many tests like this and the protocols
> > for each will not
> > always adhere to the same rest and stimulation
> > periods of 45seconds and
> > 10seconds as mentioned previously. I have to perform
> > some pre and post
> > processing to the noisy data and apply other
> > algorithms to SOME of the
> > results so I would like to find a way to control the
> > graphics of matlab as
> > mentioned below and be able to change the lenght of
> > time for the different
> > states, rest stimulation, semi stimulation. Really
> > im looking for a way to
> > section off different states in the plot with colour
> > ID, of green and white
> > as mentioned below. If i cannot get the info on how
> > to do it I will use
> > Excel for the time being but it would be better to
> > have a function m-file to
> > adjust the graphics, as with ' bar.m' etc. Thanks
> > for the help Jeff,
> >
> >   ChrisJ
> >
> >
> >
> >   On 3/9/06, Jeff Winter <jeff.winter@jeff...>
> > wrote:
> >     Hi Chris,
> >
> >     For once I'd say use MS Excel rather than
> > Matlab. It's a piece of cake.
> > The
> >     graphics capability of Excel is often
> > overlooked.
> >
> >     Cheers,
> >
> >     Jeff
> >
> >
> >     -----Original Message-----
> >     From: matlab@matl...
> > [mailto:matlab@matl...]On Behalf Of
> >     chrissoraghan@chri...
> >     Sent: 08 March 2006 17:16
> >     To: matlab@matl...
> >     Subject: [matlab] Break plot up into block
> > intervals in different
> >     colours
> >
> >
> >     Hi,
> >
> >     I have some EEG data. The 1st 45seconds of data
> > is rest data, followed
> > by
> >     some stimulation data for 10 seconds. Again
> > another 45 seconds of rest
> > and
> >     another 10 seconds of stimulation data.
> >
> >     I want to have the first 45seconds of the plot
> > in white and the next 10
> >     seconds in green to show the rest-to-stimulation
> > transition (and again
> >     another 45seconds of white and 10 seconds of
> > white). I have seen this
> >     displayed in some papers but i have no idea how
> > to do it.
> >
> >     e.g.
> >     data with white backgrounddata with green
> > background
> >
> >
> >     ...where  is the transition between the rest
> > and stimulation data
> >
> >
> >     Any help appreciated,
> >     ChrisJ.
> >