S2S-ARSNet: Sequence-to-Sequence automatic renal segmentation network. (January 2023)

• Record Type: Journal Article
• Authors: Cao, Gaoyu; Sun, Zhanquan; Wang, Chaoli; Geng, Hongquan; Fu, Hongliang; Sun, Lin; Nan, Jiao
• Is Part Of: Biomedical signal processing and control. Volume 79(2023)Part 1
• Publication Date: 2023-01
• Subjects: ConvLSTM -- Deep learning -- Diuretic renography -- Sequence-to-Sequence image segmentation; Signal processing -- Periodicals; Biomedical engineering -- Periodicals; Signal Processing, Computer-Assisted -- Periodicals; Image Processing, Computer-Assisted -- Periodicals; Biomedical Engineering -- Periodicals; 610.28
• DOI: 10.1016/j.bspc.2022.104121
• Languages: English
• ISSNs: 1746-8094

Abstract

Highlights: A Sequence-to-Sequence Automatic Renal Segmentation Network (S2S-ARSNet) is proposed for diuretic renography. Through embedding the ConvLSTM into the Unet structure, inter-slice sequence information and intra-slice spatial information can be automatically learned. The spatial and temporal information is enriched and the segmentation results are further optimized. A pre-trained Unet was used to generate an auxiliary mask at the intermediate moment to supervise each frame, which eliminated the interference of displacement caused by patients in clinical diagnosis. Compared with the existing methods, the proposed method achieves a better result with a DSC of 0.9470 and an IOU of 0.9004. Abstract: Accurate segmentation of kidney contours in diuretic renography has important implications for clinical diagnosis and treatments. However, the lack of clear boundaries and high-quality images makes automatic segmentation challenging. This paper proposes a novel automatic renal segmentation network, S2S-ARSNet, combining Convolutional Long Short-Term Memory (ConvLSTM) with the Unet structure. Unet is used to learn the spatial information of each sequence, and ConvLSTM is used to discover the temporal information between sequences and automatically update the temporal state of the sequence. Moreover, an additional pre-trained Unet is applied to generate coarse masks at different times to simulate the displacement that may occur during the detection process. In this way, the spatiotemporal information is modelled, and all the information of the 3D data is fully utilized to eliminate false positives and improve the segmentation accuracy. Extensive experiments were performed on the diuretic renography dataset. The experimental results show that the proposed method can significantly enhance the kidney segmentation performance compared with other single-image-based deep learning segmentation methods.