A conditional registration network for continuous 4D respiratory motion synthesis

Sang, Yudi; Ruan, Dan

doi:10.1002/mp.16226

Cited by 4 publications

(3 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We evaluated the robustness of our results at multiple scales considering global image fidelity, global lung motion, tumor motion, and also motion of neighboring OARs. Unlike previous studies that only considered global reconstruction metrics (Lee et al 2023, Sang andRuan 2023), covering more aspects of the model's performance across different metrics and datasets allowed for a more comprehensive evaluation of the properties of the model. To remove dependency on object size, we calculated in addition to volumetric overlaps the Euclidean distance between the CoM of tumors as a volume-independent metric.…”

Section: Discussionmentioning

confidence: 99%

“…where V A and V B are the lung volumes of the EOE phase and the target phase respectively. As already mentioned in the introduction, alongside global lung motion assessment, it is also crucial for RT treatment planning to evaluate the model accuracy at the tumor level, a problem somewhat overlooked by previous 4DCT synthesis reports (Lee et al 2023, Sang andRuan 2023). To this end, we used the dice similarity coefficient (DSC) between the GTVs delineated on *…”

Section: Validation Setupmentioning

confidence: 99%

“…4D-Lung dataset We used a public dataset from The Cancer Imaging Archive (TCIA), hereafter referred to as 4D-Lung (Hugo et al 2017) and reported in other 4DCT synthesis studies (Huang et al 2022, Sang and Ruan 2023). It is composed of 20 images of locally advanced, non-small cell lung cancer patients and includes GTV segmentation masks.…”

Section: Datasetsmentioning

confidence: 99%

See 2 more Smart Citations

CT respiratory motion synthesis using joint supervised and adversarial learning

Cao,

Bourbonne,

Lucia

et al. 2024

Phys. Med. Biol.

View full text Add to dashboard Cite

Objective: Four-dimensional computed tomography (4DCT) imaging consists in reconstructing a CT acquisition into multiple phases to track internal organ and tumor motion. It is commonly used in radiotherapy treatment planning to establish planning target volumes. However, 4DCT increases protocol complexity, may not align with patient breathing during treatment, and lead to higher radiation delivery. Approach: In this study, we propose a deep synthesis method to generate pseudo respiratory CT phases from static images for motion-aware treatment planning. The model produces patient-specific deformation vector fields (DVFs) by conditioning synthesis on external patient surface-based estimation, mimicking respiratory monitoring devices. A key methodological contribution is to encourage DVF realism through supervised DVF training while using an adversarial term jointly not only on the warped image but also on the magnitude of the DVF itself. This way, we avoid excessive smoothness typically obtained through deep unsupervised learning, and encourage correlations with the respiratory amplitude.Main results: Performance is evaluated using real 4DCT acquisitions with smaller tumor volumes than previously reported. Results demonstrate for the first time that the generated pseudo-respiratory CT phases can capture organ and tumor motion with similar accuracy to repeated 4DCT scans of the same patient. Mean inter-scans tumor center-of-mass distances and Dice similarity coefficients were $1.97$mm and $0.63$, respectively, for real 4DCT phases and $2.35$mm and $0.71$ for synthetic phases, and compares favorably to a state-of-the-art technique (RMSim). Significance: This study presents a deep image synthesis method that addresses the limitations of conventional 4DCT by generating pseudo-respiratory CT phases from static images. Although further studies are needed to assess the dosimetric impact of the proposed method, this approach has the potential to reduce radiation exposure in radiotherapy treatment planning while maintaining accurate motion representation. Our training and testing code can be found at https://github.com/cyiheng/Dynagan.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Validation Setupmentioning

confidence: 99%