Dosimetrically triggered adaptive radiotherapy for head and neck cancer: Considerations for the implementation of clinical protocols

Montero, Aldemar; Ooteghem, G. Van; Dumont, Damien; Rivas, Sara Teruel; Sterpin, Edmond; Geets, Xavier

doi:10.1002/acm2.14095

Cited by 2 publications

(5 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Barragàn-Montera et al developed a protocol for automatically and dosimetrically triggered ART. 88 The authors emphasize the complexity of designing universal flagging conditions to suit clinical requirements. Guidelines formulated by expert groups to establish tolerance levels are necessary to standardize flagging criteria and minimalize the existing inter-center variability.…”

Section: Referencesmentioning

confidence: 99%

“…102 In addition, the automation within these software systems introduces certain "black-box steps" that require meticulous attention. After using and testing the ETHOS technology, 88,98 Barragàn-Montero et al stated that, ideally, an experienced RO should perform a visual inspection of all contours. 88 These limitations may explain why in the existing trials, summarized in Table 2, off-line ART approaches have shown more consistent dose benefits.…”

Section: Practical Challengesmentioning

confidence: 99%

“…After using and testing the ETHOS technology, 88,98 Barragàn-Montero et al stated that, ideally, an experienced RO should perform a visual inspection of all contours. 88 These limitations may explain why in the existing trials, summarized in Table 2, off-line ART approaches have shown more consistent dose benefits. Lastly, in the current version of ETHOS, the user can only choose at the planning stage to use an adaptive workflow, with no possibility to change that decision during RT.…”

Section: Practical Challengesmentioning

confidence: 99%

“…Barragàn‐Montera et al. developed a protocol for automatically and dosimetrically triggered ART 88 . The authors emphasize the complexity of designing universal flagging conditions to suit clinical requirements.…”

Section: Pitfalls For Clinical Implementationmentioning

confidence: 99%

“… 102 In addition, the automation within these software systems introduces certain “black‐box steps” that require meticulous attention. After using and testing the ETHOS technology, 88 , 98 Barragàn‐Montero et al. stated that, ideally, an experienced RO should perform a visual inspection of all contours.…”

Section: Pitfalls For Clinical Implementationmentioning

confidence: 99%

See 4 more Smart Citations

Adaptive radiotherapy for head and neck cancer: Pitfalls and possibilities from the radiation oncologist's point of view

Nuyts,

Bollen,

Eisbruch

et al. 2024

Cancer Medicine

View full text Add to dashboard Cite

BackgroundPatients with head and neck cancer (HNC) may experience substantial anatomical changes during the course of radiotherapy treatment. The implementation of adaptive radiotherapy (ART) proves effective in managing the consequent impact on the planned dose distribution.MethodsThis narrative literature review comprehensively discusses the diverse strategies of ART in HNC and the documented dosimetric and clinical advantages associated with these approaches, while also addressing the current challenges for integration of ART into clinical practice.Results and ConclusionAlthough based on mainly non‐randomized and retrospective trials, there is accumulating evidence that ART has the potential to reduce toxicity and improve quality of life and tumor control in HNC patients treated with RT. However, several questions remain regarding accurate patient selection, the ideal frequency and timing of replanning, and the appropriate way for image registration and dose calculation. Well‐designed randomized prospective trials, with a predetermined protocol for both image registration and dose summation, are urgently needed to further investigate the dosimetric and clinical benefits of ART.

show abstract

Section: Referencesmentioning

confidence: 99%

Section: Practical Challengesmentioning

confidence: 99%

Section: Practical Challengesmentioning

confidence: 99%

Section: Pitfalls For Clinical Implementationmentioning

confidence: 99%

Section: Pitfalls For Clinical Implementationmentioning

confidence: 99%

See 3 more Smart Citations

Adaptive radiotherapy for head and neck cancer: Pitfalls and possibilities from the radiation oncologist's point of view

Nuyts,

Bollen,

Eisbruch

et al. 2024

Cancer Medicine

View full text Add to dashboard Cite

show abstract

Can input reconstruction be used to directly estimate uncertainty of a dose prediction U‐Net model?

Huet‐Dastarac,

Nguyen,

Longton

et al. 2024

Medical Physics

View full text Add to dashboard Cite

BackgroundThe reliable and efficient estimation of uncertainty in artificial intelligence (AI) models poses an ongoing challenge in many fields such as radiation therapy. AI models are intended to automate manual steps involved in the treatment planning workflow. We focus in this study on dose prediction models that predict an optimal dose trade‐off for each new patient for a specific treatment modality. They can guide physicians in the optimization, be part of automatic treatment plan generation or support decision in treatment indication.Most common uncertainty estimation methods are based on Bayesian approximations, like Monte Carlo dropout (MCDO) or Deep ensembling (DE). These two techniques, however, have a high inference time (i.e., require multiple inference passes) and might not work for detecting out‐of‐distribution (OOD) data (i.e., overlapping uncertainty estimate for in‐distribution (ID) and OOD).PurposeIn this study, we present a direct uncertainty estimation method and apply it for a dose prediction U‐Net architecture. It can be used to flag OOD data and give information on the quality of the dose prediction.MethodsOur method consists in the addition of a branch decoding from the bottleneck which reconstructs the CT scan given as input. The input reconstruction error can be used as a surrogate of the model uncertainty. For the proof‐of‐concept, our method is applied to proton therapy dose prediction in head and neck cancer patients. A dataset of 60 oropharyngeal patients was used to train the network using a nested cross‐validation approach with 11 folds (training: 50 patients, validation: 5 patients, test: 5 patients). For the OOD experiment, we used 10 extra patients with a different head and neck sub‐location. Accuracy, time‐gain, and OOD detection are analyzed for our method in this particular application and compared with the popular MCDO and DE.ResultsThe additional branch did not reduce the accuracy of the dose prediction model. The median absolute error is close to zero for the target volumes and less than 1% of the dose prescription for organs at risk. Our input reconstruction method showed a higher Pearson correlation coefficient with the prediction error (0.620) than DE (0.447) and MCDO (between 0.599 and 0.612). Moreover, our method allows an easier identification of OOD (no overlap for ID and OOD data and a Z‐score of 34.05). The uncertainty is estimated simultaneously to the regression task, therefore requires less time and computational resources.ConclusionsThis study shows that the error in the CT scan reconstruction can be used as a surrogate of the uncertainty of the model. The Pearson correlation coefficient with the dose prediction error is slightly higher than state‐of‐the‐art techniques. OOD data can be more easily detected and the uncertainty metric is computed simultaneously to the regression task, therefore faster than MCDO or DE. The code and pretrained model are available on the gitlab repository: https://gitlab.com/ai4miro/ct‐reconstruction‐for‐uncertainty‐quatification‐of‐hdunet

show abstract

Dosimetrically triggered adaptive radiotherapy for head and neck cancer: Considerations for the implementation of clinical protocols

Cited by 2 publications

References 44 publications

Adaptive radiotherapy for head and neck cancer: Pitfalls and possibilities from the radiation oncologist's point of view

Adaptive radiotherapy for head and neck cancer: Pitfalls and possibilities from the radiation oncologist's point of view

Can input reconstruction be used to directly estimate uncertainty of a dose prediction U‐Net model?

Contact Info

Product

Resources

About