Vi tilrår at du alltid nyttar siste versjon av nettlesaren din.

New medical technology students Kathinka and Julie join PiV

Hello, our names are Kathinka and Julie, and we are currently in our second year of the medical technology (MTEK) program at University of Bergen. We’re lucky to be doing our internship here at PiV, where we will be until June.

Publisert 13.03.2024
Et par kvinner som sitter ved et bord med en projektorskjerm bak seg
MTEK students Kathinka Tobiassen and Julie Andreassen Misje

We will continue the project started by last year's MTEK students, working on developing a tool for quality control of immunohistochemical stains in digital pathology. The students from last year made a significant progress in this task, by making a classifier that detects and classifies cells into plasma cells (positive), other cells (negative) and other/unidentified structures. By separating into positive and negative cells, we can analyze the color- profiles of each, where both cell types have blue nuclei, but only positive cells show an additional brown color, indicating the presences of immunoglobulins. This is our foundation for developing an automatic method for analysis, which can detect day-to day variation of immunohistochemical staining quality in whole slide images (WSI). 

Today, every WSI is reviewed for quality issues by the pathologist at the diagnostic workstation. This process is time consuming and can delay the pathology diagnosis, particularly if immunohistochemical stains must be repeated. This manual review seems unnecessary when a computer could easily identify and filter out outliers.

Our project aims to address this issue. If our project is succeeded, our tool will screen WSIs just after staining before they are sent to the pathology workstation. It will automatically detect aberrant staining prompting the lab to perform new staining. As a result, fewer images will need quality control by a pathologist, and instances of poor staining will be reduced. By developing an automatic method for analysis, we can increase the quality in the images in less time, which will help the pathologists to conduct more precise analysis and diagnoses. 

When it comes to the working progress, it can be challenging to handle a project with no solution, as well as dealing with numerous new programs. Luckily, we collaborate well, and are surrounded by experienced people who are willing to help. Our weekly supervision meetings with Sabine and Hrafn are very helpful. There are also many other students who has faced many of the same problems as we do. 

En gruppe mennesker som poserer for et bilde
From the left: Sabine, Kathinka, Julie and Hrafn

Last, but not least, we enjoy being a part of PiV, and look forward to the rest of our time here. An easy-going group, that works hard to achieve goals together.  

Greetings :-)