Machine learning can be used to automatically extract and quantify black pigment cells from zebrafish skin images. However, the training data used can significantly impact the results. The author aims to determine the size and distribution of black cells in zebrafish stripes. Differences in lighting conditions and microscope lenses between training and test images can affect the accuracy of cell predictions. Future work will evaluate image processing methods and isolate factors like lighting that influence results.
Original Description:
expo 2021 poster
Original Title
Harita Duggirala Expo Poster 2 Template .Pptx (1) - Harita Duggirala
Machine learning can be used to automatically extract and quantify black pigment cells from zebrafish skin images. However, the training data used can significantly impact the results. The author aims to determine the size and distribution of black cells in zebrafish stripes. Differences in lighting conditions and microscope lenses between training and test images can affect the accuracy of cell predictions. Future work will evaluate image processing methods and isolate factors like lighting that influence results.
Machine learning can be used to automatically extract and quantify black pigment cells from zebrafish skin images. However, the training data used can significantly impact the results. The author aims to determine the size and distribution of black cells in zebrafish stripes. Differences in lighting conditions and microscope lenses between training and test images can affect the accuracy of cell predictions. Future work will evaluate image processing methods and isolate factors like lighting that influence results.
Machine Learning to Identify Cells in Zebrafish-Skin Patterns
Harita Duggirala (Computer Science, 2024)
haritaduggirala2024@u.northwestern.edu
Broad Motivation Challenges Role of Training Set on Results
● Machine learning methods are often a black box deccanherald.com ● Fish pictures are taken under various lighting conditions & microscopic lenses (see below) Zebrafish deccanherald.com
Zebrafish deccanherald.com
Zebrafish (Danio rerio) share a similar genetic
structure to humans. Their black pigment cells Our results show that using different images for training during machine learning has: (melanophores) are related to the originating cells ● Minimal effect on our predictions of cell size for melanoma in humans.The skin patterns of ● Significant effect on the number of cells we extract Objectives zebrafish are comprised of horizontal stripes of different colored pigment cells. Accurately ● Automatically extract black cells from a large set quantifying the cell size of various pigment cells of fish images Future Research ● Quantitatively describe how different machine in relation to their location on the skin is a crucial ● Compare accuracy of different image processing learning methods influence resulting step towards better understanding cell behavior. measurements software to determine optimal method for extracting cells from fish images Method ● Isolate and determine the individual effects of Motivation for My Project different lighting and microscopic-lens conditions Determine the size and distribution of black ● Identify black cells manually in small set of on the accuracy of our cell predictions pigment cells (melanophores) in the stripes. images using Ilastik ● Utilize batch processing to generate predictions of black cells for larger set of images Acknowledgments ● Quantitatively describe effects of using different ● Close-up images of cells courtesy of Hans Georg Frohnhöfer and initial training sets on accuracy of cell prediction Christiane Nüsslein-Volhard (Max Planck Institute for Developmental Biology) ● The study resulting in this presentation was assisted by a grant administered by Northwestern University's Office of Undergraduate Research. However, the conclusions, opinions, Magnified image of zebrafish stripes and other statements in this presentation are the author's and not necessarily those of the sponsoring institution. ● Thank you to Dr. Alexandria Volkening for your mentorship and Approach: Machine learning to extract cells from support for this project. Ilastik Object Classification Ilastik Object Prediction fish images in an automated process