User Interface for Refined Dose-Toxicity Analysis in Radiotherapy
Last updated: Tue May 1 13:30
Summary
The goal of this project is to develop a web-based user interface for radio dose toxicity analysis.
Students: Alaleh Azhir, Santiago Appiani, William Franceschi
Mentor(s): Dr. Todd McNutt, Pranav Lakshminarayanan
Completed UI with yellow representing the DVH and segmentation pipeline, and blue representing auxiliary features and download optionsaua:
Background, Specific Aims, and Significance
Physicians lack a simple method to quantitatively analyze dose distributions within organs before, during and after treatment. A user interface with 3D rendering would allow physicians to better analyze dose distributions before treatment, and make necessary adjustments to the treatment plan. Currently, there exist no simple tools for clinicians or researchers for extracting and analyzing dosage features of treatment plans.
Specific Aims:
Compatible with existing online SQL database for obtaining the medically related data
Create interactive 3D visualizations of objects using JavaScript libraries such as D3.js
Allow physician to easily use segmenting tools on organs, run analysis on new regions
Display the results using interactive DVH histograms on the website
Allow new feature analysis scripts to be easily added to the existing user interface
Enable the users to Download the results of the analyses from the web
Deliverables
Minimum: (completed)
A UI for visualizing organs in 3D, calculating DVH curves, and running python analysis scripts from the javascript layer
Documentation
Expected: (completed)
A UI for segmenting and analyzing organs in 3D using a list of segmentation options.
Dose-volume data analysis scripts are integrated and can be performed on segments of organs.
Results of the analysis can be exported.
Maximum: (ongoing)
An interactive UI for segmenting and analyzing organs in 3D with a flexible segmentation as indicated by the user in addition to the existing ones. (additional segmentation feature ongoing)
Create and export DVH data for batch of patients (completed)
Extract specific points on DVH given list of doses (completed)
User friendly - help direct user step by step & notify user with loading screen for long processes (completed)
Technical Approach
1. Set up Python Web Framework on the back-end
2. Set up front-end visualizations:
Choose Library for data selection features
D3 Library for DBH curves and object 3D visualization
Include options, and later text boxes to help user segment data
3. Enabling communication between back and front end using XHR requests
4. Enabling user to download the results of the analysis as a .csv file
Dependencies
- Dependency : Access to Database
- Dependency : Access to DVH analysis Code
- Dependency : Access to Pranav's Segmentation Code
Milestones and Status
Milestone name: Basic Website Framework
Planned Date: 2/27
Expected Date: 2/27
Status: Complete
Milestone name: Choosing Js Libraries
Planned Date: 2/27
Expected Date: 2/27
Status: Complete
Milestone name: Basic UI with DVH
Planned Date: 3/15
Expected Date: 3/15
Status: Complete
Milestone name: Exporting Data
Planned Date: 3/27
Expected Date: 3/27
Status: Complete
Milestone name: Interact with 3D Visual Representation of ROIs
Planned Date: 4/1
Expected Date: 4/1
Status: Complete
Milestone name: Integrate segmentation tools
Planned Date: 4/10
Expected Date: 4/15
Status: Complete
Milestone name: Finialize Documentation
Planned Date: 4/15
Expected Date: 4/15
Status: Complete
Milestone name: Add customizable Segmentation Tools
Planned Date: 4/25
Expected Date: 4/25
Status: Ongoing
Reports and presentations
Project Plan
Project Background Reading
Paper Seminar Presentations
Santiago:
-
-
Willie:
-
-
Alaleh:
-
-
Checkpoint Presentation
Project Final Presentation
Project Final Report
Project Bibliography
* Lakshminarayanan, P. (2017). Radio-morphology: Parametric Shape-Based Features in Radiotherapy (Unpublished master's thesis). Johns Hopkins University.
* McNutt, T., PhD., & Lakshminarayanan, P. (2018, February 6). User Interface to Extract radio-morphologic features for refined dose-toxicity analysis in radiotherapy. Lecture presented at CIS II Lecture in Hackerman B17, Baltimore, MD.
* Chen R, Gabriel P, Kavanagh B, McNutt T, “How will big data impact clinical decision making and precision medicine in radiation therapy?” Int’l J. of Radiation Oncology, Biology, Physics. Published online: November 27 2015
Other Resources and Project Files