An efficient pyramid scene parsing network with multi-scale feature fusion for liver segmentation in magnetic resonance imaging

1

The PSP-EffB0-MSFF model is proposed for automated liver segmentation from MRI images, addressing the need for reliable clinical applications.
2

EfficientNetB0 replaces ResNet50 in the model to reduce computational costs while enhancing feature representation through multi-scale feature fusion.
3

The model demonstrates high performance on the DLDS dataset with an intersection over union of 0.905 and a Dice score of 0.913.
4

On the CirrMRI600+ dataset, the model achieves an intersection over union of 0.86 and a Dice score of 0.90, indicating consistent performance.
5

The proposed model requires 14.91 GFLOPs, showcasing its efficiency in processing MRI scans for liver segmentation.

Frontiers In Medicine

1