Automatic Segmentation of Pelvic Cancers Using Deep Learning: State-of-the-Art Approaches and Challenges

Reza Kalantar; Gigin Lin; Jessica M Winfield; Christina Messiou; Susan Lalondrelle; Matthew D Blackledge; Dow-Mu Koh

doi:10.3390/diagnostics11111964

Automatic Segmentation of Pelvic Cancers Using Deep Learning: State-of-the-Art Approaches and Challenges

Diagnostics (Basel). 2021 Oct 22;11(11):1964. doi: 10.3390/diagnostics11111964.

Authors

Reza Kalantar¹, Gigin Lin², Jessica M Winfield^{1

3}, Christina Messiou^{1

3}, Susan Lalondrelle^{1

3}, Matthew D Blackledge¹, Dow-Mu Koh^{1

3}

Affiliations

¹ Division of Radiotherapy and Imaging, The Institute of Cancer Research, London SM2 5NG, UK.
² Department of Medical Imaging and Intervention, Chang Gung Memorial Hospital at Linkou and Chang Gung University, 5 Fuhsing St., Guishan, Taoyuan 333, Taiwan.
³ Department of Radiology, The Royal Marsden Hospital, London SW3 6JJ, UK.

Abstract

The recent rise of deep learning (DL) and its promising capabilities in capturing non-explicit detail from large datasets have attracted substantial research attention in the field of medical image processing. DL provides grounds for technological development of computer-aided diagnosis and segmentation in radiology and radiation oncology. Amongst the anatomical locations where recent auto-segmentation algorithms have been employed, the pelvis remains one of the most challenging due to large intra- and inter-patient soft-tissue variabilities. This review provides a comprehensive, non-systematic and clinically-oriented overview of 74 DL-based segmentation studies, published between January 2016 and December 2020, for bladder, prostate, cervical and rectal cancers on computed tomography (CT) and magnetic resonance imaging (MRI), highlighting the key findings, challenges and limitations.

Keywords: deep learning; pelvic cancer segmentation; radiation oncology; radiology; radiotherapy planning.

Publication types

Review

Abstract

Publication types

Grants and funding