Advanced search

Simulation Pipeline for Virtual Clinical Trials of Dermatology Images

(2019)
Author
Organization
Abstract
We have designed a pipeline for developing Virtual Clinical Trials (VCTs) for optical imaging of skin lesions. Skin cancer is the most common cancer in the US; one in five Americans will develop it by the age of 70. Early diagnosis offers more favorable treatment options; currently available diagnostics, however, shows large reader-dependence. A simulation approach offers advantages in terms of the known ground truth and flexibility to model clinically observed lesion variability. The pipeline is motivated by our success with VCTs of breast imaging. Our open source pipeline has been used to develop and evaluate novel clinical-grade displays, a new generation of tomosynthesis systems, and search observer models. The proposed pipeline includes modules for the simulation of skin and subcutaneous tissue, insertion of lesion models, and simulated acquisition of optical images, followed by the virtual reading of simulated images and the analysis of task-based system performance. The physical properties of the skin and lesion were selected among clinically plausible values. Skin lesion images were simulated assuming the ambient white light, and a linear camera model. We have utilized the standards for optimal data representation based upon the XML schema, adopted from our breast imaging VCTs. The presentation describes VCT principles used for designing the pipeline. We present proof-of-concept results, and discuss the preliminary visual and quantitative evaluation. Integration of pipeline modules and its validation is ongoing
Keywords
Skin lesion imaging, skin cancer, computer simulation of skin anatomy, virtual clinical trials, Blender

Citation

Please use this url to cite or link to this publication:

Chicago
Vasudev, Varun, Bastian Piepers, Andrew Maidment, Tom Kimpe, Predrag Bakic, Ljiljana Platisa, and Wilfried Philips. 2019. “Simulation Pipeline for Virtual Clinical Trials of Dermatology Images.” In .
APA
Vasudev, V., Piepers, B., Maidment, A., Kimpe, T., Bakic, P., Platisa, L., & Philips, W. (2019). Simulation Pipeline for Virtual Clinical Trials of Dermatology Images. Presented at the Physics of Medical Imaging.
Vancouver
1.
Vasudev V, Piepers B, Maidment A, Kimpe T, Bakic P, Platisa L, et al. Simulation Pipeline for Virtual Clinical Trials of Dermatology Images. 2019.
MLA
Vasudev, Varun, Bastian Piepers, Andrew Maidment, et al. “Simulation Pipeline for Virtual Clinical Trials of Dermatology Images.” 2019. Print.
@inproceedings{8576713,
  abstract     = {We have designed a pipeline for developing Virtual Clinical Trials (VCTs) for optical imaging of skin lesions.  Skin cancer is the most common cancer in the US; one in five Americans will develop it by the age of 70.  Early diagnosis offers more favorable treatment options; currently available diagnostics, however, shows large reader-dependence.  A simulation approach offers advantages in terms of the known ground truth and flexibility to model clinically observed lesion variability. 
The pipeline is motivated by our success with VCTs of breast imaging.  Our open source pipeline has been used to develop and evaluate novel clinical-grade displays, a new generation of tomosynthesis systems, and search observer models. 
The proposed pipeline includes modules for the simulation of skin and subcutaneous tissue, insertion of lesion models, and simulated acquisition of optical images, followed by the virtual reading of simulated images and the analysis of task-based system performance.  The physical properties of the skin and lesion were selected among clinically plausible values.  Skin lesion images were simulated assuming the ambient white light, and a linear camera model.  We have utilized the standards for optimal data representation based upon the XML schema, adopted from our breast imaging VCTs.  
The presentation describes VCT principles used for designing the pipeline.  We present proof-of-concept results, and discuss the preliminary visual and quantitative evaluation.  Integration of pipeline modules and its validation is ongoing},
  author       = {Vasudev, Varun and Piepers, Bastian and Maidment, Andrew and Kimpe, Tom and Bakic, Predrag and Platisa, Ljiljana and Philips, Wilfried},
  keyword      = {Skin lesion imaging,skin cancer,computer simulation of skin anatomy,virtual clinical trials,Blender},
  language     = {eng},
  location     = {San Diego, California},
  title        = {Simulation Pipeline for Virtual Clinical Trials of Dermatology Images},
  year         = {2019},
}