L
Lucian Mihai Itu
Researcher at Siemens
Publications - 114
Citations - 1720
Lucian Mihai Itu is an academic researcher from Siemens. The author has contributed to research in topics: Fractional flow reserve & Deep learning. The author has an hindex of 19, co-authored 97 publications receiving 1383 citations. Previous affiliations of Lucian Mihai Itu include Transilvania University of Brașov & Princeton University.
Papers
More filters
Patent
Viscoelastic modeling of blood vessels
TL;DR: In this article, a method for modeling a blood vessel includes: (a) modeling a first segment of the blood vessel based on medical imaging data acquired from a subject; (b) computing a first modeling parameter at an interior point of the first segment; and (c) computing another model at a boundary point of first segment using a viscoelastic wall model.
Journal ArticleDOI
Privacy-Preserving and Explainable AI in Industrial Applications
Iulian Ogrezeanu,Anamaria Vizitiu,Costin Florian Ciusdel,Andrei Puiu,Simona Coman,Cristian Boldisor,Alina Itu,R.M. Demeter,Florin Moldoveanu,Constantin Suciu,Lucian Mihai Itu +10 more
TL;DR: This paper focuses on recent advancements related to the challenges mentioned above, discusses the industrial impact of proposed solutions, and identifies challenges for future research, and comments on the interaction between the identified challenges in the conclusions.
Journal ArticleDOI
Privacy-Preserving and Explainable AI for Cardiovascular Imaging
TL;DR: Recent developments related to explainability and interpretability have become core requirements for AI algorithms, to ensure that the rationale behind output inference can be revealed, and the clinical impact of proposed solutions are discussed.
Journal ArticleDOI
Obfuscation Algorithm for Privacy-Preserving Deep Learning-Based Medical Image Analysis
Andreea Bianca Popescu,Ioana Antonia Taca,Anamaria Vizitiu,Cosmin Nita,Constantin Suciu,Lucian Mihai Itu,Alexandru Scafa-Udriste +6 more
TL;DR: An image obfuscation algorithm is proposed that combines a variational autoencoder (VAE) with random non-bijective pixel intensity mapping to protect the content of medical images, which are subsequently employed in the development of DL-based solutions.
Posted Content
A Self-Taught Artificial Agent for Multi-Physics Computational Model Personalization
Dominik Neumann,Dominik Neumann,Tommaso Mansi,Lucian Mihai Itu,Lucian Mihai Itu,Bogdan Georgescu,Elham Kayvanpour,Farbod Sedaghat-Hamedani,Ali Amr,Jan Haas,Hugo A. Katus,Benjamin Meder,Stefan Steidl,Joachim Hornegger,Dorin Comaniciu +14 more
TL;DR: In this article, a self-taught artificial agent, Vito, learns a representative decision process model through exploration of the computational model: it learns how the model behaves under change of parameters.