2023
DOI: 10.3390/cancers15174389
|View full text |Cite
|
Sign up to set email alerts
|

Automatic Segmentation with Deep Learning in Radiotherapy

Lars Johannes Isaksson,
Paul Summers,
Federico Mastroleo
et al.

Abstract: This review provides a formal overview of current automatic segmentation studies that use deep learning in radiotherapy. It covers 807 published papers and includes multiple cancer sites, image types (CT/MRI/PET), and segmentation methods. We collect key statistics about the papers to uncover commonalities, trends, and methods, and identify areas where more research might be needed. Moreover, we analyzed the corpus by posing explicit questions aimed at providing high-quality and actionable insights, including:… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1

Citation Types

0
2
0

Year Published

2024
2024
2025
2025

Publication Types

Select...
5
1

Relationship

0
6

Authors

Journals

citations
Cited by 15 publications
(2 citation statements)
references
References 59 publications
0
2
0
Order By: Relevance
“…The extracted manuscripts covered various RT application spaces and disease sites. Auto-contouring was the most common application, aligning with its prevalence in AI-based RT [11,103,104]. Many studies focused on head and neck cancer, likely due to the complexity of this disease site, which requires precise delineation of numerous organs at risk (OARs) and challenging tumor-related target structures [105].…”
Section: Discussionmentioning
confidence: 99%
See 1 more Smart Citation
“…The extracted manuscripts covered various RT application spaces and disease sites. Auto-contouring was the most common application, aligning with its prevalence in AI-based RT [11,103,104]. Many studies focused on head and neck cancer, likely due to the complexity of this disease site, which requires precise delineation of numerous organs at risk (OARs) and challenging tumor-related target structures [105].…”
Section: Discussionmentioning
confidence: 99%
“…Due to the highly quantitative and structured nature of the RT clinical workflow, AI-based methodologies -namely, machine learning (ML) and deep learning (DL)have been increasingly investigated to automate and improve a variety of tasks [8]. Advances in DL algorithms trained on increasingly larger, diverse datasets have allowed for impressive performance in a variety of RT-related applications such as image synthesis [9], registration [10], contouring [11], dose prediction [12], and outcome prediction [13][14][15]. However, despite the impressive performance of these models in research studies, to date there are relatively few standard AI-based tools that are routinely used in RT workflows.…”
Section: Introductionmentioning
confidence: 99%