João Nunes

2papers

2 Papers

4.8CRMay 5
On Digital Twins in Defence: Overview and Applications

Marco Giberna, Holger Voos, Paulo Tavares et al.

Digital twins have emerged as a transformative technology for modeling and simulation in various industries, including defense. This paper provides a comprehensive review of digital twin applications in defense modeling and simulation, focusing on how digital twins can enhance simulation fidelity, interoperability, and decision support within defense systems. We consolidate existing research into a unified framework that links digital twin concepts, simulation-driven application, and real-world deployment in defense scenarios. We discuss the role of digital twin in applications like planning, training, execution and monitoring, and debriefing. We introduce a standardized digital twin characterization framework suitable for defense application that aligns with industrial modeling and simulation standards, and present a taxonomy of defense specific use cases, highlighting recurring requirements. Additionally, practical evidence is provided from a targeted questionnaire distributed to defense stakeholders and Ministries of Defense, revealing current challenges in digital twin integration and deployment. Finally, we conclude by identifying key gaps in digital twins application for defense modeling and simulation, including interoperability, security, and system integration, and we outline future research directions and development opportunities. This review aims to inform defense modeling and simulation practitioners and researchers, guiding future work on digital twin design, implementation and deployment across defense applications.

IVMar 1, 2022
Colon Nuclei Instance Segmentation using a Probabilistic Two-Stage Detector

Pedro Costa, Yongpan Fu, João Nunes et al.

Cancer is one of the leading causes of death in the developed world. Cancer diagnosis is performed through the microscopic analysis of a sample of suspicious tissue. This process is time consuming and error prone, but Deep Learning models could be helpful for pathologists during cancer diagnosis. We propose to change the CenterNet2 object detection model to also perform instance segmentation, which we call SegCenterNet2. We train SegCenterNet2 in the CoNIC challenge dataset and show that it performs better than Mask R-CNN in the competition metrics.