Work

X-Balloon: A cloud-based platform for annotation and reinforced deep learning in digital pathology images

Year: 2025

Type: software-paper

Abstract: This paper introduces X-Balloon, a cloud-based platform designed to streamline annotation workflows and reinforce deep learning models specifically for biopsy or digital pathology image analysis. The platform integrates a modular architecture comprising a Backend, Annotation, and AI Processing module to address inefficiencies in traditional pathology workflows. By leveraging Mask R-CNN for automated segmentation, X-Balloon achieves high precision in identifying pathological features, thereby enhancing diagnostic accuracy. Its browser-based interface enables seamless collaboration among pathologists while reducing annotation effort through a combination of automation and manual refinement. X-Balloon’s open-source availability and customizable architecture make it a valuable tool for advancing the integration of AI into digital pathology.

Source: SoftwareX

DOI: 10.1016/j.softx.2025.102287

Authors: Odysseas Tsakai, Andreas Miltiadous, Panagiotis N. Smyrlis, Alexandros T. Tzallas, Markos G. Tsipouras, Nikolaos Giannakeas

Institutions: University of Ioannina, University of Western Macedonia

Language: English

Topic: AI in cancer detection

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: gold

Estimated APC paid: $1,200

Funders: European Commission, University of Ioannina