Robustness of Transformer-Based Fluence Map Prediction Under Clinically Realistic Perturbations

• URL: http://arxiv.org/abs/2605.00904v1
• Date: Tue, 28 Apr 2026 21:32:47 GMT
• Title: Robustness of Transformer-Based Fluence Map Prediction Under Clinically Realistic Perturbations
• Abstract summary: We study a two-stage transformer pipeline that maps anatomy (CT and contours) to dose and then to beamlet fluence maps.<n>We compare fluence-stage transformer backbones with hierarchical, global, and hybrid attention, trained with a physics-informed loss enforcing energy consistency.<n>Results show smooth degradation under moderate perturbations but sharp failures under severe rotations and noise.
• Score: 4.066732323672965
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Learning-based fluence map prediction offers a fast alternative to iterative inverse planning in intensity-modulated radiation therapy (IMRT), but its robustness under realistic distribution shifts remains unclear. We study a two-stage transformer pipeline that maps anatomy (CT and contours) to dose and then to beamlet fluence maps. We compare fluence-stage transformer backbones with hierarchical, global, and hybrid attention, trained with a physics-informed loss enforcing energy consistency. Robustness is evaluated under geometric perturbations, radiometric noise, reduced training data, and domain shifts using a prostate IMRT dataset, with additional evaluation of the dose stage on public datasets. Results show smooth degradation under moderate perturbations but sharp failures under severe rotations and noise. Hierarchical transformers (e.g., SwinUNETR) exhibit slower growth in upper-quartile energy error, indicating improved robustness. We further show that SSIM alone fails to capture clinically relevant errors, highlighting the need for physics-informed evaluation.

Related papers

• ResLRP: The Role of Residual Cancellation in Attribution Instability in Vision Transformers [58.411900827911346]
  We show that the missing ingredient is the treatment of residual connections.<n>We introduce Residual-aware Layer-wise Relevance Propagation (ResLRP)<n>ResLRP is a simple extension of LRP whose propagation rules explicitly account for cancellations in residual branches.
  arXiv  Detail & Related papers  (2026-09-15T13:17:30Z)
• TTSD-FAR: Test-Time Self-Distillation with Fisher-Anchored Restoration for Missing-Modality Emotion Recognition in LVLMs [83.78708832901194]
  Large video-language models (LVLMs) have shown remarkable performance on multimodal tasks like multimodal emotion recognition (ER) in the wild.<n>However, real-world deployment remains challenging: modalities may be missing or noisy at test time.<n>We propose a solution in which a frozen teacher guides an adaptive low-rank student via self-distillation.<n>Experiments on MELD, DFEW, and BAH under 0%-50% missing modalities show that this unified adaptation-restoration design consistently outperforms entropy-based adaptation, RAG, and perplexity-based generation.
  arXiv  Detail & Related papers  (2026-08-18T23:40:28Z)
• Test-time Adaptation of Pelvic Bone Segmentation Models via Dynamic Reliability-Guided [62.10276009853225]
  Reliable pelvic bone segmentation from CT is essential for robot-assisted pelvic trauma surgery.<n>Existing methods show limited effectiveness for PBS.<n>We propose a novel closed-loop dynamic Reliability-Guided framework (ReGA) for PBS.
  arXiv  Detail & Related papers  (2026-08-01T08:13:20Z)
• Self-Auditing Residual Drifting for Pathology-Preserving Accelerated Knee MRI [5.844079330191764]
  SA-RDM-DC is a Self-Auditing Residual generative Drifting Model with Data Consistency for accelerated knee MRI.<n>It predicts image- and missing-k-space residual corrections, enforces data consistency with acquired k-space, uses frequency-aware and residual drifting supervision to recover fine detail.<n>It produces dense error maps and slice-level risk scores in the same inference pass.
  arXiv  Detail & Related papers  (2026-07-02T16:59:03Z)
• Tumor-aware augmentation with task-guided attention analysis improves rectal cancer segmentation from magnetic resonance images [1.3958880590777554]
  We show two interacting failure modes in CT-to-MRI transfer.<n>Inefficient token usage caused by zero-padding to match pretrained input dimensions and ineffective feature adaptation.<n>Fine-tuning on identical rectal MRI datasets improved detection rates to 224/247 (90.7%) for SMIT and 219/247 (88.7%) for Swin UNETR.
  arXiv  Detail & Related papers  (2026-05-06T23:46:46Z)
• CMR: Contractive Mapping Embeddings for Robust Humanoid Locomotion on Unstructured Terrains [30.85514738551041]
  We present Contractive Mapping for Robustness (CMR) framework that maps high-dimensional, disturbance-prone observations into a latent space.<n>CMR potently outperforms other locomotion algorithms under increased noise.
  arXiv  Detail & Related papers  (2026-02-03T13:30:18Z)
• FluenceFormer: Transformer-Driven Multi-Beam Fluence Map Regression for Radiotherapy Planning [4.066732323672965]
  We introduce textbfFluenceFormer, a backbone-agnostic transformer framework for direct, geometry-aware fluence regression.<n>FluenceFormer with Swin UNETR achieves the strongest performance among the evaluated models.
  arXiv  Detail & Related papers  (2025-12-27T01:12:15Z)
• An efficient approach with theoretical guarantees to simultaneously reconstruct activity and attenuation sinogram for TOF-PET [6.491602790776125]
  In positron emission tomography (PET), it is indispensable to perform attenuation correction in order to obtain the quantitatively accurate activity map (tracer distribution) in the body.<n>We propose a new mathematical model for simultaneously reconstructing the activity and attenuation sinogram from the time-of-flight (TOF)-PET emission data only.
  arXiv  Detail & Related papers  (2025-10-15T14:01:03Z)
• Heatmap Guided Query Transformers for Robust Astrocyte Detection across Immunostains and Resolutions [1.0405048273969084]
  We propose a hybrid CNN Transformer detector that combines local feature extraction with global contextual reasoning.<n>A heatmap guided query mechanism generates spatially grounded anchors for small and faint astrocytes.<n>The model consistently outperformed Faster R-CNN, YOLOv11 and DETR, achieving higher sensitivity with fewer false positives.
  arXiv  Detail & Related papers  (2025-09-03T14:01:04Z)
• CathFlow: Self-Supervised Segmentation of Catheters in Interventional Ultrasound Using Optical Flow and Transformers [66.15847237150909]
  We introduce a self-supervised deep learning architecture to segment catheters in longitudinal ultrasound images. The network architecture builds upon AiAReSeg, a segmentation transformer built with the Attention in Attention mechanism. We validated our model on a test dataset, consisting of unseen synthetic data and images collected from silicon aorta phantoms.
  arXiv  Detail & Related papers  (2024-03-21T15:13:36Z)
• Rotational Augmented Noise2Inverse for Low-dose Computed Tomography Reconstruction [83.73429628413773]
  Supervised deep learning methods have shown the ability to remove noise in images but require accurate ground truth. We propose a novel self-supervised framework for LDCT, in which ground truth is not required for training the convolutional neural network (CNN) Numerical and experimental results show that the reconstruction accuracy of N2I with sparse views is degrading while the proposed rotational augmented Noise2Inverse (RAN2I) method keeps better image quality over a different range of sampling angles.
  arXiv  Detail & Related papers  (2023-12-19T22:40:51Z)
• Lightweight Convolution Transformer for Cross-patient Seizure Detection in Multi-channel EEG Signals [0.0]
  This study proposes a novel deep learning architecture based lightweight convolution transformer (LCT) The transformer is able to learn spatial and temporal correlated information simultaneously from the multi-channel electroencephalogram (EEG) signal to detect seizures at smaller segment lengths.
  arXiv  Detail & Related papers  (2023-05-07T16:43:52Z)
• Negligible effect of brain MRI data preprocessing for tumor segmentation [36.89606202543839]
  We conduct experiments on three publicly available datasets and evaluate the effect of different preprocessing steps in deep neural networks. Our results demonstrate that most popular standardization steps add no value to the network performance. We suggest that image intensity normalization approaches do not contribute to model accuracy because of the reduction of signal variance with image standardization.
  arXiv  Detail & Related papers  (2022-04-11T17:29:36Z)
• Treatment Learning Causal Transformer for Noisy Image Classification [62.639851972495094]
  In this work, we incorporate this binary information of "existence of noise" as treatment into image classification tasks to improve prediction accuracy. Motivated from causal variational inference, we propose a transformer-based architecture, that uses a latent generative model to estimate robust feature representations for noise image classification. We also create new noisy image datasets incorporating a wide range of noise factors for performance benchmarking.
  arXiv  Detail & Related papers  (2022-03-29T13:07:53Z)
• DuDoTrans: Dual-Domain Transformer Provides More Attention for Sinogram Restoration in Sparse-View CT Reconstruction [13.358197688568463]
  iodine radiation in the imaging process induces irreversible injury. Iterative models are proposed to alleviate the appeared artifacts in sparse-view CT images, but the cost is too expensive. We propose textbfDual-textbfDomain textbfDuDoTrans to reconstruct CT image with both the enhanced and raw sinograms.
  arXiv  Detail & Related papers  (2021-11-21T10:41:07Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.