ePAD: An Image Annotation and Analysis Platform for Quantitative Imaging

Abstract

Medical imaging is critical for assessing the response of patients to new cancer therapies. Quantitative lesion assessment on images is time-consuming, and adopting new promising quantitative imaging biomarkers of response in clinical trials is challenging. The electronic Physician Annotation Device (ePAD) is a freely available web-based zero-footprint software application for viewing, annotation, and quantitative analysis of radiology images designed to meet the challenges of quantitative evaluation of cancer lesions. For imaging researchers, ePAD calculates a variety of quantitative imaging biomarkers that they can analyze and compare in ePAD to identify potential candidates as surrogate endpoints in clinical trials. For clinicians, ePAD provides clinical decision support tools for evaluating cancer response through reports summarizing changes in tumor burden based on different imaging biomarkers. As a workflow management and study oversight tool, ePAD lets clinical trial project administrators create worklists for users and oversee the progress of annotations created by research groups. To support interoperability of image annotations, ePAD writes all image annotations and results of quantitative imaging analyses in standardized file formats, and it supports migration of annotations from various propriety formats. ePAD also provides a plugin architecture supporting MATLAB server-side modules in addition to client-side plugins, permitting the community to extend the ePAD platform in various ways for new cancer use cases. We present an overview of ePAD as a platform for medical image annotation and quantitative analysis. We also discuss use cases and collaborations with different groups in the Quantitative Imaging Network and future directions.

Keywords: AIM (Annotation and Image Markup); DICOM SR (DICOM Structure Report); biomarker evaluation; feature extraction; medical image annotation.

Figure 1.

Architecture of the ePAD platform, which comprises an image database (dcm4chee PACS as a cache for images), a database of image annotations (AIM XML database), the ePAD Viewer (a web application), ePAD Web Services that communicate data between the image and annotation databases and the ePAD Viewer, and back-end (server-side) and front-end (client-side) plugins enabling the community to extend the ePAD platform.

Figure 2.

ePAD viewer and annotation window. Images are displayed in the ePAD web viewer, and the user records image annotations in using drawing tools (eg, to create an ROI, shown on the left) and an annotation window (to record qualitative image features, shown on right).

Figure 3.

ADLA histogram on a line annotation on a cancer lesion created and visualized in ePAD.

Figure 4.

T1 perfusion map generated by ePAD plugin derived from the algorithm in Jarrett et al.'s study (112), with the map overlaid on magnetic resonance (MR) image using a color lookup table.

Figure 5.

QF Explore Plugin Suite: gray-level co-occurrence matrix feature extraction and comparison chart. The user can compare the feature values for various regions of interest (ROIs). GLCM contrast and correlation is higher for vascular ROIs (85).

Figure 6.

Progress view of ePAD visualizing a particular project (“Liver”) that contains 5 patients. The status column shows the overall status for that series/study or patient, and the user statuses column shows the status of annotations that have been created by each ePAD user associated with that project.

Figure 7.

A tumor burden report (using linear measurement as the imaging biomarker and RECIST response criteria) and a longitudinal annotation report of a patient having 4 time points and 3 lesions. This report is automatically generated from ePAD's image annotations and enables clinicians to determine image-based treatment response in the patient.

Figure 8.

Waterfall report plot based on linear measurement as the imaging biomarker of response and RECIST as the response criteria, showing the best response score for each patient in the study cohort. This plot enables researchers to assess the effectiveness of cancer treatment in the cohort, and a variety of these plots can be generated using different imaging biomarkers of response (upper left corner).

Figure 9.

Cumulative ePAD statistics collected from ePAD instances between 2015 and 2018.