Added IAFDB Dataset and IAFClassification sample task.

[jakecumber1]

Students: Jake Cumberland (jakepc3), Swaroop Potdar (spotd)
Paper Title: Interpretation of Intracardiac Electrograms Through Textual Representations
Paper Link: https://arxiv.org/abs/2402.01115

Highlights

There are two new files, one in datasets called "IAFDBDataset.py" which contains the code for our new dataset object IAFDBDataset, and the other file is in tasks as "IAFClassificationTask.py" which contains the code for a custom task to test the IAFDBDataset. Changes have also been made to init in both tasks and datasets to recognize these classes.

The IAFDBDataset is meant to be used with the external dataset in the original paper available here: https://physionet.org/static/published-projects/iafdb/intracardiac-atrial-fibrillation-database-1.0.0.zip

The IAFDBDataset handles and preps the data in the same way as the original paper but some updates have been made to make it compatible with pyhealth and the BaseDataset class it derives from.

The IAFClassificationTask is a simple task to make sure IAFDBDataset is functioning correctly, it preps a sample for training.

Testing

You can test both the dataset and the task are functioning properly by running the following code:

dataset = IAFDBDataset("./intracardiac-atrial-fibrillation-database-1.0.0")
task = IAFClassificationTask(dataset)

x, y = task(task[0])
print(x.shape, y)

Which should print out (1000, 1)