2006 Scientific Session Abstracts
Scientific Session Presentations & Schedule
E-Poster Presentations
Submission Compliance and Guidelines
Copyright Transfer Form (PDF)
Abstract Archive
Software-assisted Visual Analysis and Review of Tissue
Microarray (TMA) Images Facilitated by XML
http://www.acsr.mid-region.org/
David G. Nohle, MS (nohle.1@osu.edu); Barbara A. Hackman, MS, MPH; Leona W. Ayers, MD; The Mid-Region AIDS and Cancer Specimen Resource (ACSR), Department of Pathology , The Ohio State University, Columbus, Ohio
Context: The AIDS and Cancer Specimen Resource (ACSR) is a tissue bank consortium sponsored by the National Cancer Institute’s Division of Cancer Treatment and Diagnosis. The ACSR offers tissue micro-array (TMA) sections with de-identified clinical data. Organized approaches to grading tissue cores and correctly associating data with cores along with visual quality review are necessary. API’s XML TMA Data Exchange Specification (TMADES) proposed format is used for exported TMA data and provides a common format for development of tissue core screen displays to ease visual review.
Technology: Internet Explorer (Microsoft Corp., Redmond, WA), HTML, XML, Extensible Stylesheet Language Transformation (XSLT) style sheets, JavaScripts based on EdX XML Document Editor (Microsoft), Word (Microsoft) mail merge, Excel (Microsoft), T2 ScanScope (Aperio, Vista, CA) with 20X objective at 0.4667 mm/pixel, TMALab with nuclear, positive pixel count and membrane algorithms (Aperio), ImageScope (Aperio).
Design: ScanScope digitizes stained TMA sections. TMALab gridded each core image and produced a separate image file for each core. TMALab/ImageScope applies an algorithm appropriate to stain with parameters determined by experimentation. Staining intensity and positive cell density were mapped. Microsoft Word Mail Merge and XSLT scripts created initial TMADES files containing clinical data. XSLT scripts integrated TMALab/ImageScope exports into TMADES files.
AnalyzeSlide, AnalyzeStain and AnalyzeCase, JavaScript programs based on EdX XML Editor (Microsoft Corp., Redmond, WA, USA), allow HTML applications and web pages to display images and edit data in TMADES files. AnalyzeSlide displays core section images with algorithmically derived grades in TMA organization and allows pathologist quality inspection and grade editing. AnalyzeStain is similar but sorts by human and algorithm grades. AnalyzeCase displays core section images relating to a selected patient from core repetitions and stains/slides with clinical data, algorithm outputs and pathologist grades and allows pathologist review of case and case-wide information entry. Each program provides a reference grading scale for each stain with example of each grade. Controls allow individual core image enlargement and invoking ImageScope with image.
Results: AnalyzeSlide, AnalyzeStain, and AnalyzeCase collate cores and display of algorithmic core grading while adjusting or entering intensity and density grades where appropriate. Resulting TMADES files work with BrowseTMA, described previously, posting images and data in web formats.
Conclusions: Software-assisted display of core images that are sorted by human or algorithmic grades allows for an efficient and accurate visual quality review of TMA.