Skip to main content
Fig. 3 | BMC Medical Informatics and Decision Making

Fig. 3

From: Automated segmentation and diagnosis of pneumothorax on chest X-rays with fully convolutional multi-scale ScSE-DenseNet: a retrospective study

Fig. 3

The concurrent spatial and channel squeeze and excitation (scSE) module. The input feature maps of a dense block \({\mathrm {U}}\) can be recalibrated to the output feature maps \({{\mathrm {U}}}_{{\mathrm{scSE}}}\) through the two branches of \({{\mathrm {U}}}_{{\mathrm{sSE}}}\) and \({{\mathrm {U}}}_{{\mathrm{cSE}}}\). The top branch is the spatial recalibrating (\({\mathrm {U}}_{{\mathrm {sSE}}}\)), and the bottom branch is channel-wise recalibrating (\({\mathrm {U}}_{{\mathrm {cSE}}}\)), and then \({\mathrm {U}}_{{\mathrm {sSE}}}\) and \({\mathrm {U}}_{{\mathrm {cSE}}}\) are merged into the output

Back to article page