Multitask 3D CBCT-to-CT Translation and Organs-at-Risk Segmentation Using Physics-Based Data Augmentation

@article{Dahiya2021Multitask3C,
  title={Multitask 3D CBCT-to-CT Translation and Organs-at-Risk Segmentation Using Physics-Based Data Augmentation},
  author={Navdeep Dahiya and Sadegh R. Alam and Pengpeng Zhang and Si-Yuan Zhang and Anthony J. Yezzi and Saad Nadeem},
  journal={Medical physics},
  year={2021}
}
PURPOSE In current clinical practice, noisy and artifact-ridden weekly cone-beam computed tomography (CBCT) images are only used for patient setup during radiotherapy. Treatment planning is done once at the beginning of the treatment using high-quality planning CT (pCT) images and manual contours for organs-at-risk (OARs) structures. If the quality of the weekly CBCT images can be improved while simultaneously segmenting OAR structures, this can provide critical information for adapting… 

Figures and Tables from this paper

Improving CBCT image quality to the CT level using RegGAN in esophageal cancer adaptive radiotherapy.

  • Hao WangXiao Liu Yongkang Zhou
  • Medicine, Physics
    Strahlentherapie und Onkologie : Organ der Deutschen Rontgengesellschaft ... [et al]
  • 2023
The proposed deep-learning RegGAN model seems promising for generation of high-quality sCT images from stand-alone thoracic CBCT images in an efficient way and thus has the potential to support CBCT-based esophageal cancer adaptive radiotherapy.

RMSim: controlled respiratory motion simulation on static patient scans.

This work presents a novel 3D Seq2Seq deep learning respiratory motion simulator (RMSim) that learns from 4D-CT images and predicts future breathing phases given a static CT image, and provides the ground truth for validating deformable image registration (DIR) algorithms and driving more accurate deep learning based DIR.

Deep learning based direct segmentation assisted by deformable image registration for cone-beam CT based auto-segmentation for adaptive radiotherapy.

A DL-based direct CBCT segmentation model can be improved to outperform DIR-based segmentation models by using deformed pCT contours as pseudo labels and influencer volumes for initial training, and by using a smaller set of true labels for model fine tuning.

Automatic Masseter Muscle Accurate Segmentation from CBCT Using Deep Learning-Based Model

A deep learning-based automatic model could simultaneously and accurately segment the MM structures on CBCT and CT, which can improve clinical efficiency and efficacy, and provide critical information for personalized treatment and long-term follow-up.

A Cascaded Multi-Task Generative Framework for Detecting Aortic Dissection on 3-D Non-Contrast-Enhanced Computed Tomography

A novel cascaded multi-task generative framework for AD detection using NCE-CT volumes obtains superior performance to state-of-the-art models in AD detection and has great potential to reduce the misdiagnosis of AD using N CE-CT in clinical practice.

Deep learning methods for enhancing cone‐beam CT image quality toward adaptive radiation therapy: A systematic review

This review encompasses DL approaches for synthetic CT generation, as well as projection domain methods employed in the CBCT correction literature, and makes recommendations for both clinicians and DL practitioners based on literature trends and the current DL state‐of‐the‐art methods utilized in radiation oncology.

Exploring the combination of deep-learning based direct segmentation and deformable image registration for cone-beam CT based auto-segmentation for adaptive radiotherapy

Utilizing deformed pCT contours as pseudo labels for training and as bounding box for shape and location feature extraction in a DS model is a good way to combine DIR and DS.

Exploring constraints on CycleGAN-based CBCT enhancement for adaptive radiotherapy

Synthetic images generated from the methods are quantitatively and qualitatively investigated and outperform the baseline CycleGAN and other approaches, and no observable artifacts or loss in image quality is found, which is critical for acceptance of these synthetic images.

Feasibility evaluation of kilovoltage cone-beam computed tomography dose calculation following scatter correction: investigations of phantom and representative tumor sites

The results suggest that CBCT has high accuracy in dose calculation via scatter correction and HU-RED calibration and the dose distribution index (DDI) and the gamma index were assessed.

References

SHOWING 1-10 OF 35 REFERENCES

Synthetic CT Generation from CBCT images via Deep Learning.

A deep U-net-based approach that synthesizes CT-like images with accurate numbers from on-treatment CBCT and planning CT, while keeping the same anatomical structure as on- treatment CBCT potentially enables advanced CBCT applications for adaptive treatment planning.

Daily cone-beam CT multi-organ segmentation for prostate adaptive radiotherapy

A synthetic MRI-aided multi-organ segmentation from cone-beam CT for prostate adaptive radiotherapy using a cycle-consistent generative adversarial network to predict the final segmentation of these critical structures.

Generating synthesized computed tomography (CT) from cone-beam computed tomography (CBCT) using CycleGAN for adaptive radiation therapy

A cycle-consistent generative adversarial network framework (CycleGAN) is developed to synthesize CT images from CBCT images that are visually and quantitatively similar to real CT images and demonstrate a higher accuracy than those on CBCT in a 3D gamma index analysis.

Generalizable cone beam CT esophagus segmentation using physics-based data augmentation

A semantic physics-based data augmentation method for segmenting the esophagus in both planning CT (pCT) and cone beam CT (CBCT) using 3D convolutional neural networks and can generalize well across modalities, eventually improving the accuracy of treatment setup and response analysis.

Visual enhancement of Cone-beam CT by use of CycleGAN.

This paper proposes a synthetic approach to translate CBCT images with deep neural networks that produces visually PlanCT-like images fromCBCT images while preserving anatomical structures.

CBCT correction using a cycle-consistent generative adversarial network and unpaired training to enable photon and proton dose calculation

The feasibility of utilizing a cycle-consistent generative adversarial network (cycleGAN) for CBCT correction using unpaired training is demonstrated, achieving high dose calculation accuracy for VMAT.

Cone-beam CT image quality improvement using Cycle-Deblur consistent adversarial networks (Cycle-Deblur GAN) for chest CT imaging in breast cancer patients

The proposed deep learning model, “Cycle-Deblur GAN”, combined with CycleGAN and Deblur-GAN models to improve the image quality and CT-value accuracy and preserved structural details for chest CBCT images.

Low-Dose CT With a Residual Encoder-Decoder Convolutional Neural Network

This work combines the autoencoder, deconvolution network, and shortcut connections into the residual encoder–decoder convolutional neural network (RED-CNN) for low-dose CT imaging and achieves a competitive performance relative to the-state-of-art methods in both simulated and clinical cases.

kV Cone-Beam CT-Based IGRT

CBCT allows for daily pretreatment position verification and online correction of set-up errors which improves the precision of patient repositioning with the possibility of shrinking safety margins, sparing organs at risk, and escalating radiation doses.

Autosegmentation for thoracic radiation treatment planning: A grand challenge at AAPM 2017

The results of the Thoracic Auto-Segmentation Challenge showed that the lungs and heart can be segmented fairly accurately by various algorithms, while deep-learning methods performed better on the esophagus.