Validation of an established deep learning auto-segmentation tool for cardiac substructures in 4D radiotherapy planning scans

Walls, Gerard; Giacometti, Valentina; Apte, Aditya; Thor, Maria; McCann, Conor; Hanna, Gerard G.; O’Connor, John D.; Deasy, Joseph O.; Hounsell, Alan R.; Butterworth, Karl T.; Cole, Aidan; Jain, Suneil; McGarry, Conor K.

doi:10.1016/j.phro.2022.07.003

Cited by 19 publications

(7 citation statements)

References 56 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Compared to many other studies presenting models for automatic segmentation of cardiac substructures, this work uses far less training data to achieve similar results, requiring only 10 images with manually contoured cardiac substructures compared to other approaches which used 41 [ 19 ], 127 [ 50 ], and 217 [ 14 ] cases for training. An overview of published tools for cardiac substructure segmentation can be found in a recent publication by Walls et al (Supplementary Table 11) [ 18 ]. The model proposed in this work provides automatic definitions of 18 independent structures, more than any other method currently available.…”

Section: Discussionmentioning

confidence: 99%

“…Haq et al [ 14 ] used a training set of 217 thoracic CT scans to develop a model for the heart, cardiac chambers, and great vessels (including the inferior vena cava and complete aorta extending down to the most inferior slice of the heart), and achieved better accuracy when measured with the DSC and 95th percentile of the Hausdorff distance. This method has also been validated on an independent dataset [ 18 ], with results suggesting reduced accuracy when applied to new data as well as systematic variations. A DL model developed by Garrett Fernandes et al [ 50 ] to delineate these same cardiac substructures from a training dataset 127 CT scans was validated on an independent dataset and also achieved higher DSC values, however a substantial reduction in performance was observed on CT imaging acquired without contrast enhancement.…”

Section: Discussionmentioning

confidence: 99%

“…Recent studies have demonstrated the potential of machine learning methods to automatically and accurately delineate cardiac substructures [12][13][14][15][16][17][18][19][20]. Existing methods can require large training datasets, fail to provide practicable segmentations for smaller structures, or rely on imaging not typically used in routine RT.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Open-source, fully-automated hybrid cardiac substructure segmentation: development and optimisation

et al. 2023

View full text Add to dashboard Cite

Radiotherapy for thoracic and breast tumours is associated with a range of cardiotoxicities. Emerging evidence suggests cardiac substructure doses may be more predictive of specific outcomes, however, quantitative data necessary to develop clinical planning constraints is lacking. Retrospective analysis of patient data is required, which relies on accurate segmentation of cardiac substructures. In this study, a novel model was designed to deliver reliable, accurate, and anatomically consistent segmentation of 18 cardiac substructures on computed tomography (CT) scans. Thirty manually contoured CT scans were included. The proposed multi-stage method leverages deep learning (DL), multi-atlas mapping, and geometric modelling to automatically segment the whole heart, cardiac chambers, great vessels, heart valves, coronary arteries, and conduction nodes. Segmentation performance was evaluated using the Dice similarity coefficient (DSC), mean distance to agreement (MDA), Hausdorff distance (HD), and volume ratio. Performance was reliable, with no errors observed and acceptable variation in accuracy between cases, including in challenging cases with imaging artefacts and atypical patient anatomy. The median DSC range was 0.81–0.93 for whole heart and cardiac chambers, 0.43–0.76 for great vessels and conduction nodes, and 0.22–0.53 for heart valves. For all structures the median MDA was below 6 mm, median HD ranged 7.7–19.7 mm, and median volume ratio was close to one (0.95–1.49) for all structures except the left main coronary artery (2.07). The fully automatic algorithm takes between 9 and 23 min per case. The proposed fully-automatic method accurately delineates cardiac substructures on radiotherapy planning CT scans. Robust and anatomically consistent segmentations, particularly for smaller structures, represents a major advantage of the proposed segmentation approach. The open-source software will facilitate more precise evaluation of cardiac doses and risks from available clinical datasets. Graphical abstract

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Open-source, fully-automated hybrid cardiac substructure segmentation: development and optimisation

et al. 2023

View full text Add to dashboard Cite

show abstract

“…The 4D-AVE was used for all segmentations, and an auto-segmented LA contour based was provided for the observers, which they were permitted to modify. The LA structure was an autosegmentation using the Haq tool (21), which is based on the Feng definitions ( 16), aligning with how contemporary treatment planning workflows are commonly furnished with automated tools for organs at risk including the cardiac substructures (22). Geometric and dosimetric comparisons of the observer structures were then made with reference structures previously jointly created by GW, CM and PB, using a range of metrics as per recommendations for inter-observer variation studies (23,24).…”

Section: Atlas Evaluationmentioning

confidence: 99%

A pulmonary vein atlas for radiotherapy planning

Walls¹,

McCann²,

Ball³

et al. 2023

Radiotherapy and Oncology

Self Cite

View full text Add to dashboard Cite

“…Some models have also been validated on datasets from other institutions 23 demonstrating their robustness and reproducibility. NTCP and TCP models are based on radiotherapy dose and radiomics features.…”

Section: Introductionmentioning

confidence: 99%

Technical Note: STRATIS: A Cloud-enabled Software Toolbox for Radiotherapy and Imaging Analysis

Apte

LoCastro

Iyer

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Purpose: Recent advances in computational resources, including software libraries and hardware, have enabled the use of high-dimensional, multi-modal datasets to build Artificial Intelligence (AI) models and workflows for radiation therapy and image analysis. The purpose of Software Toolbox for RAdioTherapy and Imaging analysiS (STRATIS) is to provide cloud-enabled, easy-to-share software workflows to train and deploy AI models for transparency and multi-institutional collaboration. Method: STRATIS leverages open source medical image informatics software for application-specific analysis. Jupyter notebooks for AI modeling workflows are provided with Python language as the base kernel. In addition to Python, workflows use software written in other languages, such as MATLAB, GNU-Octave, R, and C++, with the help of bridge libraries. The workflows can be run on a cloud platform, local workstation, or an institutional HPC cluster. Computational environments are provided in the form of publicly available docker images -and build scripts for local Anaconda environments. Utilities provided with STRATIS simplify bookkeeping of associations between imaging objects and allow chaining data processing operations defined via a setting file for AI models. Results: Workflows available on STRATIS can be broadly categorized into image segmentation, deformable image registration, and outcomes modeling for radiotherapy toxicity and tumor control using radiomics and dosimetry features. The STRATIS-forge GitHub organization https://www.github.com/stratis-forge hosts build-scripts for Docker and Anaconda as well as Jupyter notebooks for analysis workflows. The software for building environments and workflow notebooks has open source-GNU-GPL copyright, and AI models retain the copyright chosen by their original developers. Conclusion: STRATIS enables researchers to deploy and share AI modeling workflows for radiotherapy and image analysis. STRATIS is publicly available on Terra.bio FireCloud platform with a pre-deployed computational environment and on GitHub organization for users pursuing local deployment.

show abstract

Validation of an established deep learning auto-segmentation tool for cardiac substructures in 4D radiotherapy planning scans

Cited by 19 publications

References 56 publications

Open-source, fully-automated hybrid cardiac substructure segmentation: development and optimisation

Open-source, fully-automated hybrid cardiac substructure segmentation: development and optimisation

A pulmonary vein atlas for radiotherapy planning

Technical Note: STRATIS: A Cloud-enabled Software Toolbox for Radiotherapy and Imaging Analysis

Contact Info

Product

Resources

About