SATr: Slice Attention with Transformer for Universal Lesion Detection
This is an incremental improvement for computer-aided diagnosis in medical imaging, enhancing detection accuracy in a domain-specific context.
The paper tackles the problem of Universal Lesion Detection (ULD) in CT scans by addressing limitations in global representation across slices, proposing a Slice Attention Transformer (SATr) block that integrates with convolutional backbones to improve accuracy without extra hyperparameters.
Universal Lesion Detection (ULD) in computed tomography plays an essential role in 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 in 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 a novel Slice Attention Transformer (SATr) block which can be easily plugged into convolution-based ULD backbones to form hybrid network structures. Such newly formed hybrid backbones can better model long-distance feature dependency via the cascaded self-attention modules in the Transformer block while still holding a strong power of modeling local features with the convolutional operations in the original backbone. Experiments with five state-of-the-art methods show that the proposed SATr block can provide an almost free boost to lesion detection accuracy without extra hyperparameters or special network designs.