• Corpus ID: 237532332

Improving Reproducibility and Performance of Radiomics in Low Dose CT using Cycle GANs

  title={Improving Reproducibility and Performance of Radiomics in Low Dose CT using Cycle GANs},
  author={Junhua Chen and Leonard Wee and Andre Dekker and I{\~n}igo Bermejo},
1 Improving Reproducibility and Performance of Radiomics in Low Dose CT using Cycle GANs Running Title: Improving Radiomics Using Cycle GANs Junhua Chen , MS; Leonard Wee , PhD; Andre Dekker, PhD; Inigo Bermejo, PhD . Department of Radiation Oncology (MAASTRO), GROW School for Oncology and Developmental Biology, Maastricht University Medical Centre+, Maastricht, 6229 ET, Netherlands a Corresponding author, Tel.: +31 0684 6149 49 E-mail address: j.chen@maastrichtuniversity.nl 

Figures and Tables from this paper


Reproducibility of CT Radiomic Features within the Same Patient: Influence of Radiation Dose and CT Reconstruction Settings.
Most radiomic features are highly affected by CT acquisition and reconstruction settings, to the point of being nonreproducible, and selecting reproducible radiomic Features with dedicated correction factors offers improved clustering reproducibility.
Generative Models Improve Radiomics Performance in Different Tasks and Different Datasets: An Experimental Study
The Generative Model improves Radiomics Performance in Oncology and Developmental Biology by improving the predictive power of the explicit versus the implicit response of the immune system.
Stability of radiomic features in CT perfusion maps.
This study aimed to identify a set of stable radiomic parameters in CT perfusion (CTP) maps with respect to CTP calculation factors and image discretization, as an input for future prognostic models for local tumor response to chemo-radiotherapy, to build a reliable predictive model based on CTP radiomics analysis.
Vulnerabilities of radiomic signature development: The need for safeguards.
  • M. Welch, C. McIntosh, +8 authors D. Jaffray
  • Medicine
    Radiotherapy and oncology : journal of the European Society for Therapeutic Radiology and Oncology
  • 2019
Vulnerabilities in radiomic signature development processes are revealed and safeguards that can be used to refine methodologies, and ensure productive radiomic development using objective and independent features are suggested.
Radiomics of CT Features May Be Nonreproducible and Redundant: Influence of CT Acquisition Parameters.
Many RFs were redundant and nonreproducible, and if all the CT parameters are fixed except field of view, tube voltage, and milliamperage, then the information provided by the analyzed RFs can be summarized in only 10 RFs because of redundancy.
Radiogenomics: bridging imaging and genomics
The field of radiogenomics originates from image processing techniques developed decades ago; however, many technical and clinical challenges still need to be addressed.
Computational Radiomics System to Decode the Radiographic Phenotype.
PyRadiomics, a flexible open-source platform capable of extracting a large panel of engineered features from medical images, is developed and its application in characterizing lung lesions is demonstrated.
Decoding tumour phenotype by noninvasive imaging using a quantitative radiomics approach
The data suggest that radiomics identifies a general prognostic phenotype existing in both lung and head-and-neck cancer, which may have a clinical impact as imaging is routinely used in clinical practice, providing an unprecedented opportunity to improve decision-support in cancer treatment at low cost.
Evaluating variability in tumor measurements from same-day repeat CT scans of patients with non-small cell lung cancer.
Chest CT scans are well reproducible and changes in unidimensional lesion size of 8% or greater exceed the measurement variability of the computer method and can be considered significant when estimating the outcome of therapy in a patient.
Reproducibility of Tumor Uptake Heterogeneity Characterization Through Textural Feature Analysis in 18F-FDG PET
The reproducibility results suggest that these 18F-FDG PET–derived parameters, which have already been shown to have predictive and prognostic value in certain cancer models, may be used to monitor therapy response and predict patient outcome.