Structure boundary-preserving U-Net for prostate ultrasound image segmentation

Hui Bi; Jiawei Sun; Yibo Jiang; Xinye Ni; Huazhong Shu

doi:10.3389/fonc.2022.900340

Structure boundary-preserving U-Net for prostate ultrasound image segmentation

Front Oncol. 2022 Jul 28:12:900340. doi: 10.3389/fonc.2022.900340. eCollection 2022.

Authors

Hui Bi^{1

2

3}, Jiawei Sun⁴, Yibo Jiang⁵, Xinye Ni^{1

4}, Huazhong Shu^{6

7

8}

Affiliations

¹ Department of Radiation Oncology, The Affiliated Changzhou No. 2 People's Hospital of Nanjing Medical University, Changzhou, China.
² School of Computer Science and Artificial Intelligence, Changzhou University, Changzhou, China.
³ Key Laboratory of Computer Network and Information Integration, Southeast University, Nanjing, China.
⁴ Jiangsu Province Engineering Research Center of Medical Physics, Changzhou, China.
⁵ School of Electrical and Information Engineering, Changzhou Institute of Technology, Changzhou, China.
⁶ Laboratory of Image Science and Technology, Southeast University, Nanjing, China.
⁷ Centre de Recherche en Information Biomédicale Sino-francais, Rennes, France.
⁸ Jiangsu Provincial Joint International Research Laboratory of Medical Information Processing, Southeast University, Nanjing, China.

Abstract

Prostate cancer diagnosis is performed under ultrasound-guided puncture for pathological cell extraction. However, determining accurate prostate location remains a challenge from two aspects: (1) prostate boundary in ultrasound images is always ambiguous; (2) the delineation of radiologists always occupies multiple pixels, leading to many disturbing points around the actual contour. We proposed a boundary structure-preserving U-Net (BSP U-Net) in this paper to achieve precise prostate contour. BSP U-Net incorporates prostate shape prior to traditional U-Net. The prior shape is built by the key point selection module, which is an active shape model-based method. Then, the module plugs into the traditional U-Net structure network to achieve prostate segmentation. The experiments were conducted on two datasets: PH2 + ISBI 2016 challenge and our private prostate ultrasound dataset. The results on PH2 + ISBI 2016 challenge achieved a Dice similarity coefficient (DSC) of 95.94% and a Jaccard coefficient (JC) of 88.58%. The results of prostate contour based on our method achieved a higher pixel accuracy of 97.05%, a mean intersection over union of 93.65%, a DSC of 92.54%, and a JC of 93.16%. The experimental results show that the proposed BSP U-Net has good performance on PH2 + ISBI 2016 challenge and prostate ultrasound image segmentation and outperforms other state-of-the-art methods.

Keywords: ASM-based key points selection; U-Net architecture; deep learning; prostate ultrasound image segmentation; shape prior.