I have a question about the use of deep learning techniques with time-fixed features and images (setting 1) and time-dependent features (setting 2). (I am pretty new to the deep learning world so please excuse me if it's a basic question.)
Setting 1: Imagine having a training dataset composed of
- some time-fixed features such as height, weight, and age of an individual at the first medical visit (these features are recorded once and therefore time-fixed, i.e., they do not change in time in the dataset).
- some medical images for each individual, such as for example a CT scan.
- a label defining if the patient has or not a specific disease.
Setting 2: Same as setting 1 but with some features that are repeated over time (time-dependent, longitudinal), such as for example blood pressure recorded twice a day for each individual for several days.
Let say that the goal is to classify if an individual has or not a specific disease given the aforementioned features.
I have seen zillions of papers and blogs talking about convolutional neural networks to classify images and a few million about recurrent neural networks for time-dependent features. However, I am not very aware of what to use in case I have time-fixed, time-dependent, and imaging features altogether.
I am wondering how you would attach this problem.
