LSTM or any model on "Sequence Image Data"

Hi, Is there anybody who had used fastai to solve the following problem =>

  1. I have 10 sequenced images of one category and 10 other sequence images for second category and such thing i had for 32 samples.
  2. My aim is to get the model in which i pass input of sequences of image and it will classify for me and tell that ok this sequence of images is of one/second category.

Anybody who had gone through similar problem can throw some light on this issue?

What does the data look like?

Data is a medical image.
There are 32 samples
In each sample there are 2 category A and B
Each category have 10 images with 1 minute time interval.
My aim is to classify the image in either A Or B category.

Have a look at the Intracranial Haemhorrage competition solutions, similar situation there with different slices of the same brain

The 2nd place solution encoded each image with a CNN and then fed full sequences to a LSTM

Jeremy also had a really excellent series of notebooks showcasing (early) fastai v2 that you 100% should check out https://www.kaggle.com/c/rsna-intracranial-hemorrhage-detection/discussion/114214

2 Likes