By: Lisa Stephens
It all started with a complaint from our IT Systems Analyst regarding the ever-growing size of our digital image storage. Our digital pathology department had been scanning the daily control slides, so pathologists could conveniently view them from their desk or from one of the regional hospitals (instead of walking to the histology department and sorting through all of the slides, or waiting until they were on the main campus). We noted that several of the sections of tissue being used for the controls were quite large. A large section of tissue will, clearly, be a larger file to store. One must remember that digital image storage is not just a one-time cost. It is continuous, year after year, unless the file is deleted. CAP guidelines state that these slides (images, implied) must be kept for 10 years, so the storage cost is compounded over that time.
The original reason for having such large sections of tissue was to be sure the tissue contained all of the characteristics needed to represent the staining properties. However, if one knew exactly what those characteristics looked like grossly, the tissue would be absolutely sure to have the qualities needed. Consequently, the size of the tissue could be smaller and more specific, resulting in a reduced file size. In theory, it could work.
So, I dusted-off my grossing skills and went to work creating new controls. The most challenging was the Movat control which needed representative tissue from lung, heart and vessel to demonstrate muscle, elastic, mucin, collagen and fibrin. As there were multiple sections of tissue, they needed to be embedded as closely together as possible to take up the least amount of surface area.
"The theory was proven to be quite effective. By using the smaller control tissue, we reduced the storage size by 74%, which equated to a savings of $3,565 per year."
Additionally, the histology department has remained vigilant in maintaining the smaller size of the control tissue. The poster demonstrating this study was presented at the 2016 NSH Symposium/Convention and was afterwards turned into a brief interview for NSH’s Podcast Series, HistoTalks. Listen to the podcast.
The poster was also presented at the DPA (Digital Pathology Association) Visions 2016 conference where it won “Best Clinical Poster”. Ultimately, the entire paper was published in the Journal of Histotechnology (Volume 40, Issue 3) as “A Small Reduction in Tissue Size Yields a Significant Reduction in Digital Image Storage Cost”. It can be accessed online here.
This year at the 2018 NSH Symposium/Convention in St. Louis, MO, we will be presenting another poster about digital pathology. We are troubleshooting digital slide scanning issues that can cause an image to be unusable for interpretation. Stay tuned!