How to set a Matrix as Output for a Neural Network

Hello, we have a question regarding Outputs of a neural network.

The Situation:

We got 7 (scalar) Inputs and 3 (scalar) Outputs working using something like this:

custom_net = fitnet([20, 20, 20, 20], 'trainbr');
trainedNet = train(custom_net, inputs, outputs);

The results we got from this are within our goal regarding the error.

The Problem:

The 3 (scalar) Outputs are just mean values regarding 3 different distributions, which are measured each in a 30x30 Matrix. While we are happy with the current situation we are wondering wether we can use the whole field as an Output and if so, how we can get Matlab to recognize this, since all we read thus far is, that Inputs as well as Outputs both need to given in a NxM Matrix, where N is the number of Elements in this In-/Outputvector and M is the number of samples.

Thank you for any help regarding this topic even if it is just "No, this is not possible" we just want to know, if there is an easy solution for this problem.