Farewell to Our Students
Summer break is approaching, and with it, the end of our collaboration with several of our dear students from both the University of Bergen (UIB) and the Western Norway University of Applied Sciences (HVL).
It has been a pleasure to get to know such wonderful people and work with them over the last year. Their efforts have been a significant part of PiV's work and progress. Let’s look one more time at what they were working on and what they achieved here with us at PiV.
Jennifer started at PiV last summer and has just pursued a master's degree in biomedical sciences at the University of Bergen. On June 17th, she defended her thesis “Image Registration for Automated Alignment and Ordering of Non-neoplastic Kidney Biopsy Sections.” Good job, Jennifer!
Her master's project was part of an endeavor to develop a future pipeline to automatically align randomly positioned biopsy sections into an ordered sequence. The software tool has the potential to facilitate/assist nephrologists when evaluating kidney biopsies. The pipeline will consist of two main components: 1) image registration and 2) consecutive section ordering based on when the section was cut using a microtome. Her thesis focused on the first component, namely the image registration phase. Image registration is the process of aligning multiple histological images to enable comparative analysis.
Julie and Kathinka were in their second year of the medical technology (MTEK) program at the University of Bergen when they joined us to do their internship at PiV. They have already completed the internship and left in May.
Julie and Kathinka continued the project started by last year's MTEK students, Thea and Eivind, working on developing a tool for quality control of immunohistochemical stains in digital pathology. They managed to visualize day-to-day variations in immunohistochemical staining quality in whole slide images (WSI). With this tool, it would be possible to screen WSIs and automatically detect aberrant staining, prompting the lab to perform new staining. The tool has the ability to increase the quality of the images and save time for pathologists.
How to integrate an algorithm? Sander, Alina, and Kamil tried to answer this question in their bachelor thesis. They joined PiV in their last year before completing their bachelor’s degrees in computer science at the Western Norway University of Applied Sciences and worked on their bachelor thesis in collaboration with PiV.
Their bachelor thesis focused on digital pathology and whether it is possible to effectively integrate AI algorithms into pathologists' workflows by using the EMPAIA API as an intermediary, or if it is too complicated to be used effectively. They succeeded! They connected a simple algorithm for quantifying interstitial fibrosis in kidney biopsies to the EMPAIA platform, and we could see both the quantitative results and a visualization of detected fibrotic areas.
Last but not least, Radwan and LB submitted their thesis about the quantification of Ki67 in neuroendocrine tumors using the open-access image analysis software QuPath. Now we know all about the possibilities for this quantification because they conducted a comprehensive literature review. Moreover, they established the quantification method with the help of the Stardist tool in QuPath and achieved good results.
Good results were achieved in all these projects, and on behalf of PiV, we would like to say thank you to all and wish you good luck in your future endeavors! Who knows, maybe our paths will cross again in the future! 😊 For now, we say farewell and good luck to the students who are no longer with the PiV project, and we wish a good summer break to those who will continue with PiV after the summer.