Various bug fixes and some new features

mrAlign/talairach.m

@@ -239,12 +239,12 @@ function talairachControlsHandler(params)
 function viewVol = vol2viewVol(vol,view2vol);
 % warp the volume
 if ~isequal(view2vol,eye(4))
-  disppercent(-inf,'Warping volume');
+  mlrDispPercent(-inf,'Warping volume');
   swapXY = [0 1 0 0;1 0 0 0;0 0 1 0;0 0 0 1];
   % warpAffine3 uses yx, not xy
   viewVol = warpAffine3(vol,swapXY*view2vol*swapXY,nan,[],'linear');
   %viewVol = warpAffine3(vol,swapXY*view2vol*swapXY,nan,[],'nearest');
-  disppercent(inf);
+  mlrDispPercent(inf);
 else
   viewVol = vol;
 end
@@ -371,14 +371,14 @@ function endHandler(ok)
     end
   end
   % read it
-  disppercent(-inf,sprintf('Loading %s',talinfo.filename));
+  mlrDispPercent(-inf,sprintf('Loading %s',talinfo.filename));
   if ~isempty(subset)
     [vol hdr] = mlrImageReadNifti(talinfo.filename,subset);
   else
     [vol hdr] = mlrImageReadNifti(talinfo.filename);
   end
   if doMean,vol = mean(vol,4);end
-  disppercent(inf);
+  mlrDispPercent(inf);
 else
   disp(sprintf('(talairach) Could not open file %s',talinfo.filename));
   return

mrLoadRet/Analysis/ConcatTSeries/concatTSeries.m

@@ -342,16 +342,16 @@
     % for means that are zero, divide by nan
     d.mean(d.mean==0) = nan;
 
-    disppercent(-inf, '(concatTSeries) Converting to percent signal change');
+    mlrDispPercent(-inf, '(concatTSeries) Converting to percent signal change');
     for i = 1:d.dim(4)
       d.data(:,:,:,i) = (d.data(:,:,:,i)./d.mean);
       if params.percentSignal == 2           % scale it to mean of 1,000
           params.scaleFactor = 10000;
           d.data(:,:,:,i) = d.data(:,:,:,i) * params.scaleFactor;
       end
-      disppercent(i/d.dim(4));
+      mlrDispPercent(i/d.dim(4));
     end
-    disppercent(inf);
+    mlrDispPercent(inf);
   end
   warning on
 
@@ -516,9 +516,9 @@
 %%%%%%%%%%%%%%%%%%%%%%%%%
 function d = detrendTSeries(d)
 
-disppercent(-inf,sprintf('(concatTSeries:detrendTSeries) Detrending data'));
+mlrDispPercent(-inf,sprintf('(concatTSeries:detrendTSeries) Detrending data'));
 d.data = reshape(eventRelatedDetrend(reshape(d.data,prod(d.dim(1:3)),d.dim(4))')',d.dim(1),d.dim(2),d.dim(3),d.dim(4));
-disppercent(inf);
+mlrDispPercent(inf);
 
 
 

mrLoadRet/Analysis/EventRelated/eventRelatedHighpass.m

@@ -60,7 +60,7 @@
   end  
 
   % notch out highest frequency that we get with sense processing
-  if d.notchFilterForTSense
+  if ~fieldIsNotDefined(d,'notchFilterForTSense') && d.notchFilterForTSense
     if iseven(n)
       if d.notchFilterForTSense == 2
 	hipassfilter((n/2)+1) = 0;
@@ -119,13 +119,13 @@
   hipassfilter = d.hipassfilter;
   if isfield(d,'hipasscutoff')
     % go through the data, detrend and apply filter in fourier domain
-    disppercent(-inf,sprintf('(eventRelatedHighpass) Applying temporal hipass filter (cutoff=%0.03f Hz)',d.hipasscutoff));
+    mlrDispPercent(-inf,sprintf('(eventRelatedHighpass) Applying temporal hipass filter (cutoff=%0.03f Hz)',d.hipasscutoff));
   else
-    disppercent(-inf,sprintf('(eventRelatedHighpass) Applying temporal hipass filter'))
+    mlrDispPercent(-inf,sprintf('(eventRelatedHighpass) Applying temporal hipass filter'))
   end    
   if (0)
     for i = 1:d.dim(1)
-      disppercent(i/d.dim(1));
+      mlrDispPercent(i/d.dim(1));
       for j = 1:d.dim(2)
 	for k = 1:d.dim(3)
 	  timecourse = squeeze(d.data(i,j,k,:));
@@ -143,7 +143,7 @@
     end
     % detrend and high pass filter
     for k = 1:d.dim(3)
-      disppercent(k/d.dim(3));
+      mlrDispPercent(k/d.dim(3));
       for j = 1:d.dim(2)
 	timecourses = squeeze(d.data(:,j,k,:));
 	timecourses = eventRelatedDetrend(timecourses')';
@@ -153,14 +153,14 @@
     end
   elseif (isfield(d,'roidata'))
     for i = 1:size(d.roidata,1)
-      disppercent(i/size(d.roidata,1))
+      mlrDispPercent(i/size(d.roidata,1))
       timecourse = squeeze(d.roidata(i,:));
       timecourse = eventRelatedDetrend(timecourse);
       timecourse = ifft(fft(timecourse) .* hipassfilter');
       d.roidata(i,:) = real(timecourse);
     end
   end
-  disppercent(inf);
+  mlrDispPercent(inf);
 end
 
 

mrLoadRet/Analysis/EventRelated/eventRelatedMultiple.m

@@ -225,13 +225,13 @@
 d.unexplainedVariance = zeros(d.dim(1),d.dim(2),d.dim(3));
 d.totalVariance = zeros(d.dim(1),d.dim(2),d.dim(3));
 % display string
-disppercent(-inf,'Calculating goodness of fit');
+mlrDispPercent(-inf,'Calculating goodness of fit');
 % cycle through images calculating the estimated hdr and r^s of the 
 % estimate.
 %
 onesmatrix = ones(length(d.volumes),1);
 for j = yvals
-    disppercent(max((j-min(yvals))/yvaln,0.1));
+    mlrDispPercent(max((j-min(yvals))/yvaln,0.1));
     for k = slices
         ehdr = squeeze(d.ehdr(:,j,k,:))';
         % get the time series we are working on
@@ -244,7 +244,7 @@
         tv{j,k} = sum(timeseries.^2);
     end
 end
-disppercent(inf);
+mlrDispPercent(inf);
 % reshape matrix. 
 for j = yvals
   for k = slices
@@ -289,12 +289,12 @@
 d.meanintensity = zeros(d.dim(1),d.dim(2),d.dim(3));
 
 % display string
-disppercent(-inf,'Calculating hdr');
+mlrDispPercent(-inf,'Calculating hdr');
 % cycle through images calculating the estimated hdr and r^s of the 
 % estimate.
 onesmatrix = ones(length(d.volumes),1);
 for j = yvals
-  disppercent(max((j-min(yvals))/yvaln,0.1));
+  mlrDispPercent(max((j-min(yvals))/yvaln,0.1));
   for k = slices
     % get the time series we are working on
     % this includes all the rows of one column from one slice
@@ -310,15 +310,15 @@
     d.meanintensity(:,j,k)=colmeans(:);
   end
 end
-disppercent(inf);
+mlrDispPercent(inf);
 % reshape matrix. this also seems the fastest way to do things. we
 % could have made a matrix in the above code and then reshaped here
 % but the reallocs needed to continually add space to the matrix
 % seems to be slower than the loops needed here to reconstruct
 % the matrix from the {} arrays.
-disppercent(-inf,'Reshaping matrices');
+mlrDispPercent(-inf,'Reshaping matrices');
 for i = xvals
-    disppercent((i-min(xvals))/xvaln);
+    mlrDispPercent((i-min(xvals))/xvaln);
     for j = yvals
         for k = slices
             % now reshape into a matrix
@@ -328,5 +328,5 @@
 end
 
 % display time took
-disppercent(inf);
+mlrDispPercent(inf);
 

mrLoadRet/Analysis/EventRelated/eventRelatedPlot.m

@@ -156,7 +156,7 @@ function eventRelatedPlot(view,overlayNum,scan,x,y,s,roi)
   % first go for the quick and dirty way, which is
   % to load up the computed hemodynamic responses
   % and average them. 
-  disppercent(-inf,'(eventRelatedPlot) Computing mean hdr');
+  mlrDispPercent(-inf,'(eventRelatedPlot) Computing mean hdr');
   for voxnum = 1:size(roi{roinum}.scanCoords,2)
     x = roi{roinum}.scanCoords(1,voxnum);
     y = roi{roinum}.scanCoords(2,voxnum);
@@ -171,7 +171,7 @@ function eventRelatedPlot(view,overlayNum,scan,x,y,s,roi)
 	end
       end
     end
-    disppercent(voxnum/size(roi{roinum}.scanCoords,2));
+    mlrDispPercent(voxnum/size(roi{roinum}.scanCoords,2));
   end
   % plot the average of the ehdrs that beat the r2 cutoff
   if roin
@@ -196,7 +196,7 @@ function eventRelatedPlot(view,overlayNum,scan,x,y,s,roi)
   % put up button whose call back will be to compute the error bars
   figpos = get(fignum,'position');
   gEventRelatedPlot.computeErrorBarsHandle = uicontrol('Parent',fignum,'Style','pushbutton','Callback',@eventRelatedPlotComputeErrorBars,'String','Compute error bars','Position',[figpos(3)/2+figpos(3)/20 figpos(4)/24 figpos(3)/2-figpos(3)/8 figpos(4)/14]);
-  disppercent(inf);
+  mlrDispPercent(inf);
 end
 
 drawnow;
@@ -305,7 +305,7 @@ function eventRelatedSaveToWorkspace(varargin)
   return
 end
 
-disppercent(-inf,'(eventRelatedPlot) Plotting time series');
+mlrDispPercent(-inf,'(eventRelatedPlot) Plotting time series');
 gEventRelatedPlot.loadingTimecourse = 1;
 subplot(2,2,1:2)
 tSeries = squeeze(loadTSeries(gEventRelatedPlot.v,gEventRelatedPlot.scan,gEventRelatedPlot.vox(3),[],gEventRelatedPlot.vox(1),gEventRelatedPlot.vox(2)));
@@ -346,7 +346,7 @@ function eventRelatedSaveToWorkspace(varargin)
 gEventRelatedPlot.tSeries = tSeries;
 
 % done
-disppercent(inf);
+mlrDispPercent(inf);
 gEventRelatedPlot.loadingTimecourse = 0;
 
 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

mrLoadRet/Analysis/EventRelated/getGlmContrast.m

@@ -73,7 +73,7 @@
 warning('off','MATLAB:divideByZero');
 
 % display string
-disppercent(-inf,'(getGlmContrast) Calculating r2');
+mlrDispPercent(-inf,'(getGlmContrast) Calculating r2');
 % cycle through images calculating the estimated hdr and r^2s of the 
 % estimate.
 %
@@ -85,7 +85,7 @@
 % was by far the faster by a factor of about 2-3. 
 onesmatrix = ones(length(d.volumes),1);
 for j = yvals
-  disppercent(max((j-min(yvals))/yvaln,0.1));
+  mlrDispPercent(max((j-min(yvals))/yvaln,0.1));
   for k = slices
     % get the time series we are working on
     % this includes all the rows of one column from one slice
@@ -121,16 +121,16 @@
     r2{j,k} = (1-sumOfSquaresResidual./sum(timeseries.^2));
   end
 end
-disppercent(inf);
+mlrDispPercent(inf);
 
 % reshape matrix. this also seems the fastest way to do things. we
 % could have made a matrix in the above code and then reshaped here
 % but the reallocs needed to continually add space to the matrix
 % seems to be slower than the loops needed here to reconstruct
 % the matrix from the {} arrays.
-disppercent(-inf,'(getGlmContrast) Reshaping matrices');
+mlrDispPercent(-inf,'(getGlmContrast) Reshaping matrices');
 for i = xvals
-  disppercent((i-min(xvals))/xvaln);
+  mlrDispPercent((i-min(xvals))/xvaln);
   for j = yvals
     for k = slices
       % get the ehdr
@@ -144,6 +144,6 @@
 end
 
 % display time took
-disppercent(inf);
+mlrDispPercent(inf);
 
 warning('on','MATLAB:divideByZero');

mrLoadRet/Analysis/EventRelated/getr2.m

@@ -60,7 +60,7 @@
 warning('off','MATLAB:divideByZero');
 
 % display string
-if verbose,disppercent(-inf,'(getr2) Calculating r2');end
+if verbose,mlrDispPercent(-inf,'(getr2) Calculating r2');end
 % cycle through images calculating the estimated hdr and r^2s of the 
 % estimate.
 %
@@ -101,16 +101,16 @@
     % calculate variance accounted for by the estimated hdr
     r2{j,k} = (1-sumOfSquaresResidual./sum(timeseries.^2));
   end
-  if verbose,disppercent(max((j-min(yvals))/yvaln,0.1));end
+  if verbose,mlrDispPercent(max((j-min(yvals))/yvaln,0.1));end
 end
-if verbose,disppercent(inf);end
+if verbose,mlrDispPercent(inf);end
 
 % reshape matrix. this also seems the fastest way to do things. we
 % could have made a matrix in the above code and then reshaped here
 % but the reallocs needed to continually add space to the matrix
 % seems to be slower than the loops needed here to reconstruct
 % the matrix from the {} arrays.
-if verbose,disppercent(-inf,'(getr2) Reshaping matrices');end
+if verbose,mlrDispPercent(-inf,'(getr2) Reshaping matrices');end
 for i = xvals
   for j = yvals
     for k = slices
@@ -122,10 +122,10 @@
       d.r2(i,j,k) = r2{j,k}(i);
     end
   end
-  if verbose,disppercent((i-min(xvals))/xvaln);end
+  if verbose,mlrDispPercent((i-min(xvals))/xvaln);end
 end
 
 % display time took
-if verbose,disppercent(inf);end
+if verbose,mlrDispPercent(inf);end
 
 warning('on','MATLAB:divideByZero');

mrLoadRet/Analysis/makeCorrelationMap.m

@@ -38,17 +38,17 @@
     % keep name of roi
     roiName = roi{1}.name;
     % for this case, we get all the ehdrs for the roi
-    disppercent(-inf,sprintf('(makeCorrelationMap) Using roi %s as source',roiName));
+    mlrDispPercent(-inf,sprintf('(makeCorrelationMap) Using roi %s as source',roiName));
     roi{1}.scanCoords = getROICoordinates(v,roi{1});
     roiN = size(roi{1}.scanCoords,2); 
     for i = 1:roiN
       xi = roi{1}.scanCoords(1,i);
       yi = roi{1}.scanCoords(2,i);
       si = roi{1}.scanCoords(3,i);
       sourceHDR(i,:) = reshape(squeeze(d.ehdr(xi,yi,si,:,:)),1,d.nhdr*d.hdrlen);
-      disppercent(i/roiN);
+      mlrDispPercent(i/roiN);
     end
-    disppercent(inf);
+    mlrDispPercent(inf);
   end
 
   % now reshape the ehdr matrix as a matrix where the first dimension