How to best chose a model and hyper parameters for an unbalanced small dataset

I'm working on a deep learning problem where I was told to use LSTM or GRU to predict if a patient will die within 4hrs in the future. For the dataset, each patient has several measurements that were measured at different times. Please refer to the image attached to this post. The real measurements were changed to use a more constant values like very low(-3,-2), low(-1), no measurement/normal(0), and so on. The dataset is very unbalanced. I only have 12% of the records that has the positive class.

I have tried training LSTM and GRU but it always get stuck on the on validation accuracy and does not change the value. It seems like it gets stuck.

My questions are:

Am I using the wrong models?

How can improve the performance?

Is there anything I can to make my models perform better. Any guidance?