Rtdsm: An Open-Source Software for Dose-Surface Map Generation and Analysis
Presentations
MO-H345-IePD-F6-2 (Monday, 7/11/2022) 3:45 PM - 4:15 PM [Eastern Time (GMT-4)]
Exhibit Hall | Forum 6
Purpose: To develop an open-source software for dose-surface map (DSM) calculation and analysis with the intent to improve the reproducibility of and the access to DSM-based research in medical physics.
Methods: A literature review was conducted to identify essential functionalities and prevailing calculation approaches to guide development. Based on this review, a modular branching DSM calculation workflow was designed in Python to support the two dominant DSM calculation approaches. Surface meshes are created from DICOM-RT files and the dose is sampled equiangularly according to the user-specified resolution and calculation approach (planar or non-planar). From there, they are unwrapped to produce a 2D representations of dose to the surface of the organs of interest. Additional functions to support the calculation of commonly used DSM features (such as dose cluster size and shape), as well as to statistically compare DSMs were also developed.
Results: Our software, rtdsm, is the first open-source package capable of calculating both common dose-surface map approaches. Calculation times are short (≦ 8 seconds per DSM), making use feasible for large datasets. Additional features for analysis use include equivalent dose conversions, common DSM feature extraction (cluster and ellipse-based), simple DSM accumulation, and significance testing using multiple comparisons permutation testing. These features encompass everything required for routine comparative DSM analysis, making DSM-based research more easily accessible to all.
Conclusion: We have developed and implemented a standalone software package that provides all essential functionalities required to perform a DSM-based study. It has been released open-source on Github as ‘rtdsm’ for collaboration and community use.
Funding Support, Disclosures, and Conflict of Interest: H. Patrick acknowledges funding from the Fonds de recherche du Quebec - Sante in the form of a Doctoral Training Award, as well as partial support by the CREATE Responsible Health and Healthcare Data Science (SDRDS) grant of the Natural Sciences and Engineering Research Council.
Keywords
Computer Software, Surface Dose, Radiation Therapy
Taxonomy
IM/TH- Image Analysis (Single Modality or Multi-Modality): Quantitative imaging
Contact Email