Daniel L. Rubin MD, Abstract Co-Author: Grant, General Electric Company

Mia Levy MD, Abstract Co-Author: Nothing to Disclose

Alan Snyder, Presenter: Nothing to Disclose

Debra Willrett, Abstract Co-Author: Nothing to Disclose

A challenge to using imaging to evaluate treatment response in cancer is that there are many images, many studies, and no systematic way to identify the tumor burden being tracked nor to consistently capture quantitative aspects of each lesion. Our objective was to develop an open source Web-based tool to automate the quantitative assessment of tumor burden on serial imaging studies.

We developed and evaluated a tool to automatically apply RECIST response criteria directly from image annotations, eliminating the need to perform calculations and apply criteria by hand. This approach may improve the ability of oncologists and radiologists to use quantitative information in images to evaluate treatment response.

Our ePAD tool produced RECIST target lesion response category assessments that agreed with those determined by the oncologist. The advantage of our tool is that all the information needed to derive automated assessment of tumor burden is directly derived from image annotations; all the user needs to do is identify (with linear or circumscribed ROI) each lesion. The tool generates a quantitative imaging report which can facilitate cancer response assessment. In addition, since the tool is Web-based, it can run on any platform for tracking lesions.

We extended our previously-developed tool called ePAD (the electronic Physician Annotation Device), an implementation of the Annotation and Image Markup (AIM) standard in a rich Web client, to enable lesion tracking for cancer imaging interpretation. The ePAD tool extracts the quantitative information from radiologists as they identify target lesions and measure them on images, it classifies them, it calculates quantitative features needed to assess the RECIST criteria, and it saves the image metadata in AIM format. We evaluated the tool by asking a radiologist use it to annotate lesions from a cancer research study. An oncologist validated the treatment response assessments automatically derived from ePAD. This exhibit will show the features and operation of ePAD for tracking cancer lesions automatically, and its potential to streamline the reporting of quantitative imaging studies.

Rubin, D, Levy, M, Snyder, A, Willrett, D, Automated Computational Assessment of Tumor Burden with a Web-based Image Annotation Tool.  Radiological Society of North America 2012 Scientific Assembly and Annual Meeting, November 25 - November 30, 2012 ,Chicago IL. http://archive.rsna.org/2012/12030506.html