[Retracted] Automated Detection of Rehabilitation Exercise by Stroke Patients Using 3-Layer CNN-LSTM Model

Algorithm 1

Data preprocessing for training.

	Input: rehabilitation exercise, labelled dataset categories
	Preprocessing:
(1)	Categories ⟵ Dataset categories
(2)	Apply data augmentation on Categories
(3)	for each_Image in Categories
	a: Image_array ⟵ (Image, “gray”)
	b: Drop null values
	c: New_array = Resize (Image_array, (256, 256)) end for
	Preparation: Training_data = [ ]
(4)	For each_class in Categories
	a: class_num ⟵ Categories (each_class)
	b: Training_data ⟵ (New_array, class_num) end for
(5)	Create Training_data
(6)	Shuffle randomly (Training_data)