Patient-specific vs Multi-Patient Vision Transformer for Markerless Tumor Motion Forecasting

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: de Hertaing, Gauthier Rotsart, Manjah, Dani, Macq, Benoit
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916837235621888
author de Hertaing, Gauthier Rotsart
Manjah, Dani
Macq, Benoit
author_facet de Hertaing, Gauthier Rotsart
Manjah, Dani
Macq, Benoit
contents Background: Accurate forecasting of lung tumor motion is essential for precise dose delivery in proton therapy. While current markerless methods mostly rely on deep learning, transformer-based architectures remain unexplored in this domain, despite their proven performance in trajectory forecasting. Purpose: This work introduces a markerless forecasting approach for lung tumor motion using Vision Transformers (ViT). Two training strategies are evaluated under clinically realistic constraints: a patient-specific (PS) approach that learns individualized motion patterns, and a multi-patient (MP) model designed for generalization. The comparison explicitly accounts for the limited number of images that can be generated between planning and treatment sessions. Methods: Digitally reconstructed radiographs (DRRs) derived from planning 4DCT scans of 31 patients were used to train the MP model; a 32nd patient was held out for evaluation. PS models were trained using only the target patient's planning data. Both models used 16 DRRs per input and predicted tumor motion over a 1-second horizon. Performance was assessed using Average Displacement Error (ADE) and Final Displacement Error (FDE), on both planning (T1) and treatment (T2) data. Results: On T1 data, PS models outperformed MP models across all training set sizes, especially with larger datasets (up to 25,000 DRRs, p < 0.05). However, MP models demonstrated stronger robustness to inter-fractional anatomical variability and achieved comparable performance on T2 data without retraining. Conclusions: This is the first study to apply ViT architectures to markerless tumor motion forecasting. While PS models achieve higher precision, MP models offer robust out-of-the-box performance, well-suited for time-constrained clinical settings.
format Preprint
id arxiv_https___arxiv_org_abs_2507_07811
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Patient-specific vs Multi-Patient Vision Transformer for Markerless Tumor Motion Forecasting
de Hertaing, Gauthier Rotsart
Manjah, Dani
Macq, Benoit
Computer Vision and Pattern Recognition
Background: Accurate forecasting of lung tumor motion is essential for precise dose delivery in proton therapy. While current markerless methods mostly rely on deep learning, transformer-based architectures remain unexplored in this domain, despite their proven performance in trajectory forecasting. Purpose: This work introduces a markerless forecasting approach for lung tumor motion using Vision Transformers (ViT). Two training strategies are evaluated under clinically realistic constraints: a patient-specific (PS) approach that learns individualized motion patterns, and a multi-patient (MP) model designed for generalization. The comparison explicitly accounts for the limited number of images that can be generated between planning and treatment sessions. Methods: Digitally reconstructed radiographs (DRRs) derived from planning 4DCT scans of 31 patients were used to train the MP model; a 32nd patient was held out for evaluation. PS models were trained using only the target patient's planning data. Both models used 16 DRRs per input and predicted tumor motion over a 1-second horizon. Performance was assessed using Average Displacement Error (ADE) and Final Displacement Error (FDE), on both planning (T1) and treatment (T2) data. Results: On T1 data, PS models outperformed MP models across all training set sizes, especially with larger datasets (up to 25,000 DRRs, p < 0.05). However, MP models demonstrated stronger robustness to inter-fractional anatomical variability and achieved comparable performance on T2 data without retraining. Conclusions: This is the first study to apply ViT architectures to markerless tumor motion forecasting. While PS models achieve higher precision, MP models offer robust out-of-the-box performance, well-suited for time-constrained clinical settings.
title Patient-specific vs Multi-Patient Vision Transformer for Markerless Tumor Motion Forecasting
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2507.07811