Reading .czi files into image arrays

I have multichannel Z-stacks taken on a Zeiss LSM700, and wish to process and analyze them in MATLAB. The files are in .czi format, so I installed the bio-formats importer and use bfopen('foo.czi') to attempt to import the images. The example I've been working with has 2 channels, 20 z-slices each, although other files may have up to 4 channels and 60+ slices. When using the bfopen command, I get a warning, but then a window pops up allowing me to manually find the file and select it. output shown here:

>> bfopen('20191001_WM_231_gel1.czi')
log4j:WARN No appenders could be found for logger (loci.formats.ClassList).
log4j:WARN Please initialize the log4j system properly.
Reading series #1
    ........................................
ans =
  1×4 cell array
  Columns 1 through 3
    {40×2 cell}    {327 java.util.Hashtable}    {40×40 cell}
  Column 4
    {1×1 loci.formats.ome.OMEXMLMetadataImpl}

Note that the window asking me to locate the file pops up after the two log4j:.... lines, after which "Reading Series #1..." shows after selecting the file.

Ideally, I want to load these .czi files as matrices with dimension (H x W x Z x C), so for my example Z = 20 z-slices, C = 2 channels, H x W = 1192x1192 pixels per image I'd end up with a workspace array (1192 x 1192 x 20 x 2) or something similar. Other things I've tried include using Fiji/ImageJ to save the stacks as .tifs, but then MATLAB only recognizes the first slice in the stack, returning a single 1192x1192 matrix (which is of no use to me). I've used MATLAB for image processing and analysis before, but never using z-stacks, and the images were already in a common format like .png or .jpeg. While I could manually separate each channel/slice and save them as .tifs or something in fiji, I'd really like to be able to automate the process and use only MATLAB.

I apologize for the generality of the question, but I'm at a loss here. All of my peers do their image processing in Fiji/ImageJ, and the few that use something like MATLAB do so only for strictly numeric applications. I know I can process my images much better by applying linear algebra techniques like PCA and SVD in matlab (c.f. Fiji), but I need to be able to read these stacks into matrices before I can work with them.

V/R

WM