Fig. 4From: Effective automatic detection of anterior cruciate ligament injury using convolutional neural network with two attention mechanism modulesStructure of attention mechanism module 1(ATM1). Green block and yellow block are the symmetrical features in the encoding and decoding process. To reduce the dimension of the channel, a convolutional layer is connected after the green block and the yellow block, obtaining the grey block. Purple block is the finally result of enhanced feature extraction. R denotes reshape, S denotes softmax activation function, □denotes matrix multiplication, ⊕denotes matrix additionBack to article page