Deep learning for segmentation in radiation therapy planning: a review

Abstract

Summary Segmentation of organs and structures, as either targets or organs‐at‐risk, has a significant influence on the success of radiation therapy. Manual segmentation is a tedious and time‐consuming task for clinicians, and inter‐observer variability can affect the outcomes of radiation therapy. The recent hype over deep neural networks has added many powerful auto‐segmentation methods as variations of convolutional neural networks (CNN). This paper presents a descriptive review of the literature on deep learning techniques for segmentation in radiation therapy planning. The most common CNN architecture across the four clinical sub sites considered was U‐net, with the majority of deep learning segmentation articles focussed on head and neck normal tissue structures. The most common data sets were CT images from an inhouse source, along with some public data sets. N‐fold cross‐validation was commonly employed; however, not all work separated training, test and validation data sets. This area of research is expanding rapidly. To facilitate comparisons of proposed methods and benchmarking, consistent use of appropriate metrics and independent validation should be carefully considered.

Deep Learning Model for Classifying Metastatic Epidural Spinal Cord Compression on MRI

“…DL in spinal oncology imaging is limited with most researchers focusing on the detection of metastases ( 30 ), or automated spinal cord segmentation as an organ at risk for radiotherapy planning ( 31 ). Average Dice similarity coefficients for spinal cord segmentation are as high as 0.9 for automated lung cancer radiotherapy planning using DL on CT studies ( 32 , 33 ). Automated detection of spinal cord compression on MRI has currently only been assessed in the cervical spine.…”

“…Our DL model is focused on Bilsky classification and currently does not have the ability to segment or outline tumors. DL auto-segmentation of tumors in MR images could optimize and reduce the time taken for radiotherapy planning ( 32 ). Future research will focus on developing a DL model for this application, which will be especially useful for SBRT.…”

Pelvic U-Net: multi-label semantic segmentation of pelvic organs at risk for radiation therapy anal cancer patients using a deeply supervised shuffle attention convolutional neural network

“…Expanding on ideas from general semantic segmentation, recent proposed neural network models have shown great performance in medical image segmentation tasks, and are now even used for RT applications, such as OAR and target segmentation for head and neck, prostate, and breast cancer patients [15][16][17][18][19]. In addition, neural network models have shown to outperform ABS methods [20,21] and can reduce the overall manual intervention time [21,22].…”

“…Even though many researchers have incorporated neural network-based segmentation methods for RT purposes, only a few models for automatic segmentation of pelvic OAR and target structures have been proposed [18,[23][24][25]. Notably, automated deep learning-based segmentation methods have rarely been applied to complex pelvic OAR structures, like small and large bowel, and resulted in relatively unsatisfactory segmentation metrics [23][24][25].…”

Auto-detection and segmentation of involved lymph nodes in HPV-associated oropharyngeal cancer using a convolutional deep learning neural network

“…Among patients with early-stage HPV-associated OPC dispositioned to definitive radiotherapy, accurate identification of involved lymph nodes is paramount to ensuring adequate dose delivery to all sites of regional disease. Although lymph node segmentation has traditionally been performed manually by a clinician, there is an evolving role for deep learning algorithms in the automation of target volume segmentation for cancers of the head and neck [6] , [7] . Within the context of OPC, deep learning algorithms have been used to auto-segment clinical target volumes (CTVs) inclusive of areas at risk for clinical and subclinical disease [8] .…”