machine-learning - CNN or RNN which one should i use?

Question

I want to do my thesis about the diagnosis of heart attack. I have already known that CNN is the best for diagnosis. But one of my seniors, he recommended me to use RNN because in my case, I will use video files, moving files as an input dataset. But I have no idea about what RNN can do in diagnosis. Can anybody give me some ideas about it? Thank in advance.

标签： machine-learning

Answer 1

Despite the outstanding performance that Neural Networks have shown us over the last decade, they still have a big big limitation: they can’t understand the sequence, in which the current state is affected by its previous states. And Recurrent Neural Networks came out as a promising solution for that. So if you are working on video it's better to use RNN because RNN can remember previous states. But you can also take a look at "Long Short-term Memory (LSTM)" in some cases which you need to remember previous states LSTM works better.