ZebRA, an open-source software tool for analyzing single cancer cells injected in Zebrafish models: segmentation and mask visualisation.

Vahidian Qazvini, Shervin (2026) ZebRA, an open-source software tool for analyzing single cancer cells injected in Zebrafish models: segmentation and mask visualisation. [Laurea magistrale], Università di Bologna, Corso di Studio in Physics [LM-DM270], Documento full-text non disponibile
Il full-text non è disponibile per scelta dell'autore. (Contatta l'autore)

Abstract

This project presents the development of the single-cell segmentation and mask visualization modules of ZebRA (Zebrafish Radiomic Analysis), an open-source MATLAB platform for quantitative analysis of cancer cells in zebrafish xenograft models. The workflow integrates a classical Otsu-based method and a Cellpose-based deep-learning approach within a unified graphical interface. The software supports image-channel selection, volume-of-interest definition, magnification-dependent Cellpose parameters, automatic segmentation, slice-by-slice manual correction, and reconstruction of full-size labelled and binary masks. Cellpose uses expected cell diameters of 8 pixels for 10x images and 14 pixels for 20x images. Segmented regions are linked across adjacent z-slices to reconstruct 3D biological objects for downstream analysis. Users can also add missed objects, remove false detections, undo edits, and inspect segmentation overlays. Validation was performed on a 10x fluorescence microscopy stack of 66 z-slices using an independently prepared manual reference containing approximately 2361 labelled 2D segments. A less restrictive Cellpose configuration increased automatic detections from about 576 to 1406. The optimized setup achieved average object-wise Jaccard indices of 0.59 and 0.35, corresponding to Dice similarities of 0.74 and 0.52. Overall, the workflow provides a practical and extensible framework for reproducible segmentation and mask visualization in zebrafish microscopy, with potential application to other 3D models such as spheroids and organoids.

Abstract
Tipologia del documento
Tesi di laurea (Laurea magistrale)
Autore della tesi
Vahidian Qazvini, Shervin
Relatore della tesi
Correlatore della tesi
Scuola
Corso di studio
Indirizzo
Applied Physics
Ordinamento Cds
DM270
Parole chiave
Zebrafish,Microscopy,Image Processing,Segmentation,3D rendering
Data di discussione della Tesi
29 Ottobre 2026
URI

Altri metadati

Gestione del documento: Visualizza il documento

^