SATr: Slice Attention with Transformer for Universal Lesion Detection

Universal Lesion Detection (ULD) in computed tomography plays an essential role computer-aided diagnosis. Promising ULD results have been reported by multi-slice-input detection approaches which model 3D context from multiple adjacent CT slices, but such methods still experience difficulty obtaining a global representation among different slices and within each individual slice since they only use convolution-based fusion operations. In this paper, we propose novel Slice Attention Transformer (SATr) block can be easily plugged into backbones to form hybrid network structures. Such newly formed better long-distance feature dependency via the cascaded self-attention modules while holding strong power of modeling local features with convolutional operations original backbone. Experiments five state-of-the-art show that proposed SATr provide almost free boost lesion accuracy without extra hyperparameters or unique designs. Code: https://github.com/MIRACLE-Center/A3D_SATr .