While writing or editing a card, add an image to the front, the back, or both. You can take the photo with the camera there and then, or pick one from your photo library.
Both sides accepting an image matters more than it sounds. A photo on the front makes the image the question: here is the slide, what is labelled at the arrow. A photo on the back makes it the answer: name the structure, then check yourself against the picture.
Photographing a whole slide and calling it a card. A card with six bullet points on it has no single answer, so there is no way to say whether you got it right, and the rating you give it is meaningless. If the image contains several facts, it wants to be several cards.
If the thing you need to know is the structure itself, drawing it from memory is a harder retrieval than recognising a photo of it, and it exposes gaps a photo lets you bluff past. Reps has a drawing pad for this, and pressure controls line thickness on Apple Pencil models that support it.
More answers on the help page, or write to [email protected].