Segmenting Medical MRI via Recurrent Decoding Cell
This work addresses segmentation challenges in medical MRI for healthcare applications, but it is incremental as it builds on existing encoder-decoder architectures with a novel fusion unit.
The authors tackled the problem of long-term dependency and multi-modality underutilization in encoder-decoder networks for medical MRI segmentation by proposing a Recurrent Decoding Cell (RDC) and Convolutional Recurrent Decoding Network (CRDN), resulting in improved segmentation accuracy and reduced model size as demonstrated on BrainWeb, MRBrainS, and HVSMR datasets.
The encoder-decoder networks are commonly used in medical image segmentation due to their remarkable performance in hierarchical feature fusion. However, the expanding path for feature decoding and spatial recovery does not consider the long-term dependency when fusing feature maps from different layers, and the universal encoder-decoder network does not make full use of the multi-modality information to improve the network robustness especially for segmenting medical MRI. In this paper, we propose a novel feature fusion unit called Recurrent Decoding Cell (RDC) which leverages convolutional RNNs to memorize the long-term context information from the previous layers in the decoding phase. An encoder-decoder network, named Convolutional Recurrent Decoding Network (CRDN), is also proposed based on RDC for segmenting multi-modality medical MRI. CRDN adopts CNN backbone to encode image features and decode them hierarchically through a chain of RDCs to obtain the final high-resolution score map. The evaluation experiments on BrainWeb, MRBrainS and HVSMR datasets demonstrate that the introduction of RDC effectively improves the segmentation accuracy as well as reduces the model size, and the proposed CRDN owns its robustness to image noise and intensity non-uniformity in medical MRI.