Antonio Pepe
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Antonio Pepe has a total of 91 co-authors. Here are the top co-authors by number of shared publications:
- Jan Egger
Institute for Artificial Intelligence in Medicine (IKIM), University Hospital Essen (AöR), Essen, Germany
15 shared publications - Christina Gsaxner
Institute of Computer Graphics and Vision, Graz University of Technology
12 shared publications - Jianning Li
Institute for Artificial Intelligence in Medicine (IKIM) University Hospital Essen (AöR) Girardetstraße 2 45131 Essen Germany
9 shared publications - Dieter Schmalstieg
VISUS, University of Stuttgart, Stuttgart, Germany
6 shared publications Christina Gsaxner
Institute of Computer Graphics and Vision, Graz University of Technology, Graz, Austria
3 shared publications- Dominik Fleischmann
Department of Radiology, Stanford University School of Medicine, Stanford, USA
3 shared publications
Latest Publications
- Open Access
- 27 June 2025
- Published by Association for Computing Machinery (ACM) in ACM Computing Surveys
- Vol. 57 (9) , 1-35
Aortic Vessel Tree Segmentation for Cardiovascular Diseases Treatment: Status Quo
The aortic vessel tree, composed of the aorta and its branches, is crucial for blood supply to the body. Aortic diseases, such as aneurysms and dissections, can lead to life-threatening ruptures, often requiring open surgery. Therefore, patients commonly undergo treatment under constant monitoring, which requires regular inspections of the vessels through medical imaging techniques. Overlapping and comparing aortic vessel tree geometries from consecutive images allows for tracking changes in both the aorta and its branches. Manual reconstruction of the vessel tree is time-consuming and impractical in clinical settings. In contrast, automatic or semiautomatic segmentation algorithms can perform this task much faster, making them suitable for routine clinical use. This article systematically reviews methods for the automatic and semiautomatic segmentation of the aortic vessel tree, concluding with a discussion on their clinical applicability, the current research landscape, and ongoing challenges.
- 27 June 2025
- Published by Association for Computing Machinery (ACM) in ACM Computing Surveys
- Vol. 57 (9) , 1-35
Open AccessAortic Vessel Tree Segmentation for Cardiovascular Diseases Treatment: Status Quo
The aortic vessel tree, composed of the aorta and its branches, is crucial for blood supply to the body. Aortic diseases, such as aneurysms and dissections, can lead to life-threatening ruptures, often requiring open surgery. Therefore, patients commonly undergo treatment under constant monitoring, which requires regular inspections of the vessels through medical imaging techniques. Overlapping and comparing aortic vessel tree geometries from consecutive images allows for tracking changes in both the aorta and its branches. Manual reconstruction of the vessel tree is time-consuming and impractical in clinical settings. In contrast, automatic or semiautomatic segmentation algorithms can perform this task much faster, making them suitable for routine clinical use. This article systematically reviews methods for the automatic and semiautomatic segmentation of the aortic vessel tree, concluding with a discussion on their clinical applicability, the current research landscape, and ongoing challenges.
- 24 June 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Medical Imaging
- Vol. 44 (12) , 4775-4786
Deep Medial Voxels: Learned Medial Axis Approximations for Anatomical Shape Modeling
Shape reconstruction from imaging volumes is a recurring need in medical image analysis. Common workflows start with a segmentation step, followed by careful post-processing and, finally, ad hoc meshing algorithms. As this sequence can be time-consuming, neural networks are trained to reconstruct shapes through template deformation. These networks deliver state-of-the-art results without manual intervention, but, so far, they have primarily been evaluated on anatomical shapes with little topological variety between individuals. In contrast, other works favor learning implicit shape models, which have multiple benefits for meshing and visualization. Our work follows this direction by introducing deep medial voxels, a semi-implicit representation that faithfully approximates the topological skeleton from imaging volumes and eventually leads to shape reconstruction via convolution surfaces. Our reconstruction technique shows potential for both visualization and computer simulations1.
- 24 June 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Medical Imaging
- Vol. 44 (12) , 4775-4786
Deep Medial Voxels: Learned Medial Axis Approximations for Anatomical Shape Modeling
Shape reconstruction from imaging volumes is a recurring need in medical image analysis. Common workflows start with a segmentation step, followed by careful post-processing and, finally, ad hoc meshing algorithms. As this sequence can be time-consuming, neural networks are trained to reconstruct shapes through template deformation. These networks deliver state-of-the-art results without manual intervention, but, so far, they have primarily been evaluated on anatomical shapes with little topological variety between individuals. In contrast, other works favor learning implicit shape models, which have multiple benefits for meshing and visualization. Our work follows this direction by introducing deep medial voxels, a semi-implicit representation that faithfully approximates the topological skeleton from imaging volumes and eventually leads to shape reconstruction via convolution surfaces. Our reconstruction technique shows potential for both visualization and computer simulations1.
- 10 February 2024
- Published by Springer Nature
- p. 110-126
Optimizing Aortic Segmentation with an Innovative Quality Assessment: The Role of Global Sensitivity Analysis
- 10 February 2024
- Published by Springer Nature
- p. 110-126
Optimizing Aortic Segmentation with an Innovative Quality Assessment: The Role of Global Sensitivity Analysis
- Open Access
- 18 August 2023
- Published by Elsevier in Computers in Biology and Medicine
- Vol. 165, 107365
Automated cross-sectional view selection in CT angiography of aortic dissections with uncertainty awareness and retrospective clinical annotations
- 18 August 2023
- Published by Elsevier in Computers in Biology and Medicine
- Vol. 165, 107365
Open AccessAutomated cross-sectional view selection in CT angiography of aortic dissections with uncertainty awareness and retrospective clinical annotations
- Open Access
- 16 May 2023
- Published by Springer Nature in Engineering with Computers
- Vol. 40 (2) , 931-951
Higher-order block-structured hex meshing of tubular structures
Numerical simulations of the cardiovascular system are growing in popularity due to the increasing availability of computational power, and their proven contribution to the understanding of pathodynamics and validation of medical devices with in-silico trials as a potential future breakthrough. Such simulations are performed on volumetric meshes reconstructed from patient-specific imaging data. These meshes are most often unstructured, and result in a brutally large amount of elements, significantly increasing the computational complexity of the simulations, whilst potentially adversely affecting their accuracy. To reduce such complexity, we introduce a new approach for fully automatic generation of higher-order, structured hexahedral meshes of tubular structures, with a focus on healthy blood vessels. The structures are modeled as skeleton-based convolution surfaces. From the same skeleton, the topology is captured by a block-structure, and the geometry by a higher-order surface mesh. Grading may be induced to obtain tailored refinement, thus resolving, e.g., boundary layers. The volumetric meshing is then performed via transfinite mappings. The resulting meshes are of arbitrary order, their elements are of good quality, while the spatial resolution may be as coarse as needed, greatly reducing computing time. Their suitability for practical applications is showcased by a simulation of physiological blood flow modelled by a generalised Newtonian fluid in the human aorta.
- 16 May 2023
- Published by Springer Nature in Engineering with Computers
- Vol. 40 (2) , 931-951
Open AccessHigher-order block-structured hex meshing of tubular structures
Numerical simulations of the cardiovascular system are growing in popularity due to the increasing availability of computational power, and their proven contribution to the understanding of pathodynamics and validation of medical devices with in-silico trials as a potential future breakthrough. Such simulations are performed on volumetric meshes reconstructed from patient-specific imaging data. These meshes are most often unstructured, and result in a brutally large amount of elements, significantly increasing the computational complexity of the simulations, whilst potentially adversely affecting their accuracy. To reduce such complexity, we introduce a new approach for fully automatic generation of higher-order, structured hexahedral meshes of tubular structures, with a focus on healthy blood vessels. The structures are modeled as skeleton-based convolution surfaces. From the same skeleton, the topology is captured by a block-structure, and the geometry by a higher-order surface mesh. Grading may be induced to obtain tailored refinement, thus resolving, e.g., boundary layers. The volumetric meshing is then performed via transfinite mappings. The resulting meshes are of arbitrary order, their elements are of good quality, while the spatial resolution may be as coarse as needed, greatly reducing computing time. Their suitability for practical applications is showcased by a simulation of physiological blood flow modelled by a generalised Newtonian fluid in the human aorta.
- Open Access
Stochastic Modeling of Inhomogeneities in the Aortic Wall and Uncertainty Quantification using a Bayesian Encoder-Decoder Surrogate
Inhomogeneities in the aortic wall can lead to localized stress accumulations, possibly initiating dissection. In many cases, a dissection results from pathological changes such as fragmentation or loss of elastic fibers. But it has been shown that even the healthy aortic wall has an inherent heterogeneous microstructure. Some parts of the aorta are particularly susceptible to the development of inhomogeneities due to pathological changes, however, the distribution in the aortic wall and the spatial extent, such as size, shape, and type, are difficult to predict. Motivated by this observation, we describe the heterogeneous distribution of elastic fiber degradation in the dissected aortic wall using a stochastic constitutive model. For this purpose, random field realizations, which model the stochastic distribution of degraded elastic fibers, are generated over a non-equidistant grid. The random field then serves as input for a uni-axial extension test of the pathological aortic wall, solved with the finite-element (FE) method. To include the microstructure of the dissected aortic wall, a constitutive model developed in a previous study is applied, which also includes an approach to model the degradation of inter-lamellar elastic fibers. Then to assess the uncertainty in the output stress distribution due to this stochastic constitutive model, a convolutional neural network, specifically a Bayesian encoder-decoder, was used as a surrogate model that maps the random input fields to the output stress distribution obtained from the FE analysis. The results show that the neural network is able to predict the stress distribution of the FE analysis while significantly reducing the computational time. In addition, it provides the probability for exceeding critical stresses within the aortic wall, which could allow for the prediction of delamination or fatal rupture.
Open AccessStochastic Modeling of Inhomogeneities in the Aortic Wall and Uncertainty Quantification using a Bayesian Encoder-Decoder Surrogate
Inhomogeneities in the aortic wall can lead to localized stress accumulations, possibly initiating dissection. In many cases, a dissection results from pathological changes such as fragmentation or loss of elastic fibers. But it has been shown that even the healthy aortic wall has an inherent heterogeneous microstructure. Some parts of the aorta are particularly susceptible to the development of inhomogeneities due to pathological changes, however, the distribution in the aortic wall and the spatial extent, such as size, shape, and type, are difficult to predict. Motivated by this observation, we describe the heterogeneous distribution of elastic fiber degradation in the dissected aortic wall using a stochastic constitutive model. For this purpose, random field realizations, which model the stochastic distribution of degraded elastic fibers, are generated over a non-equidistant grid. The random field then serves as input for a uni-axial extension test of the pathological aortic wall, solved with the finite-element (FE) method. To include the microstructure of the dissected aortic wall, a constitutive model developed in a previous study is applied, which also includes an approach to model the degradation of inter-lamellar elastic fibers. Then to assess the uncertainty in the output stress distribution due to this stochastic constitutive model, a convolutional neural network, specifically a Bayesian encoder-decoder, was used as a surrogate model that maps the random input fields to the output stress distribution obtained from the FE analysis. The results show that the neural network is able to predict the stress distribution of the FE analysis while significantly reducing the computational time. In addition, it provides the probability for exceeding critical stresses within the aortic wall, which could allow for the prediction of delamination or fatal rupture.
- Top Cited PapersOpen Access
- 6 January 2022
- Published by Elsevier in Data in Brief
- Vol. 40, 107801
AVT: Multicenter aortic vessel tree CTA dataset collection with ground truth segmentation masks
- 6 January 2022
- Published by Elsevier in Data in Brief
- Vol. 40, 107801
Top Cited PapersOpen AccessAVT: Multicenter aortic vessel tree CTA dataset collection with ground truth segmentation masks
- Open Access
- 17 November 2021
- Published by PeerJ in PeerJ Computer Science
- Vol. 7, e773
Deep learning—a first meta-survey of selected reviews across scientific disciplines, their commonalities, challenges and research impact
Deep learning belongs to the field of artificial intelligence, where machines perform tasks that typically require some kind of human intelligence. Deep learning tries to achieve this by drawing inspiration from the learning of a human brain. Similar to the basic structure of a brain, which consists of (billions of) neurons and connections between them, a deep learning algorithm consists of an artificial neural network, which resembles the biological brain structure. Mimicking the learning process of humans with their senses, deep learning networks are fed with (sensory) data, like texts, images, videos or sounds. These networks outperform the state-of-the-art methods in different tasks and, because of this, the whole field saw an exponential growth during the last years. This growth resulted in way over 10,000 publications per year in the last years. For example, the search engine PubMed alone, which covers only a sub-set of all publications in the medical field, provides already over 11,000 results in Q3 2020 for the search term ‘deep learning’, and around 90% of these results are from the last three years. Consequently, a complete overview over the field of deep learning is already impossible to obtain and, in the near future, it will potentially become difficult to obtain an overview over a subfield. However, there are several review articles about deep learning, which are focused on specific scientific fields or applications, for example deep learning advances in computer vision or in specific tasks like object detection. With these surveys as a foundation, the aim of this contribution is to provide a first high-level, categorized meta-survey of selected reviews on deep learning across different scientific disciplines and outline the research impact that they already have during a short period of time. The categories (computer vision, language processing, medical informatics and additional works) have been chosen according to the underlying data sources (image, language, medical, mixed). In addition, we review the common architectures, methods, pros, cons, evaluations, challenges and future directions for every sub-category.
- 17 November 2021
- Published by PeerJ in PeerJ Computer Science
- Vol. 7, e773
Open AccessDeep learning—a first meta-survey of selected reviews across scientific disciplines, their commonalities, challenges and research impact
Deep learning belongs to the field of artificial intelligence, where machines perform tasks that typically require some kind of human intelligence. Deep learning tries to achieve this by drawing inspiration from the learning of a human brain. Similar to the basic structure of a brain, which consists of (billions of) neurons and connections between them, a deep learning algorithm consists of an artificial neural network, which resembles the biological brain structure. Mimicking the learning process of humans with their senses, deep learning networks are fed with (sensory) data, like texts, images, videos or sounds. These networks outperform the state-of-the-art methods in different tasks and, because of this, the whole field saw an exponential growth during the last years. This growth resulted in way over 10,000 publications per year in the last years. For example, the search engine PubMed alone, which covers only a sub-set of all publications in the medical field, provides already over 11,000 results in Q3 2020 for the search term ‘deep learning’, and around 90% of these results are from the last three years. Consequently, a complete overview over the field of deep learning is already impossible to obtain and, in the near future, it will potentially become difficult to obtain an overview over a subfield. However, there are several review articles about deep learning, which are focused on specific scientific fields or applications, for example deep learning advances in computer vision or in specific tasks like object detection. With these surveys as a foundation, the aim of this contribution is to provide a first high-level, categorized meta-survey of selected reviews on deep learning across different scientific disciplines and outline the research impact that they already have during a short period of time. The categories (computer vision, language processing, medical informatics and additional works) have been chosen according to the underlying data sources (image, language, medical, mixed). In addition, we review the common architectures, methods, pros, cons, evaluations, challenges and future directions for every sub-category.
- Open Access
- 4 November 2021
- Published by Elsevier in Data in Brief
- Vol. 39, 107524
MUG500+: Database of 500 high-resolution healthy human skulls and 29 craniotomy skulls and implants
- 4 November 2021
- Published by Elsevier in Data in Brief
- Vol. 39, 107524
Open AccessMUG500+: Database of 500 high-resolution healthy human skulls and 29 craniotomy skulls and implants
- 3 May 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Medical Imaging
- Vol. 40 (9) , 2329-2342
AutoImplant 2020-First MICCAI Challenge on Automatic Cranial Implant Design
The aim of this paper is to provide a comprehensive overview of the MICCAI 2020 AutoImplant Challenge. The approaches and publications submitted and accepted within the challenge will be summarized and reported, highlighting common algorithmic trends and algorithmic diversity. Furthermore, the evaluation results will be presented, compared and discussed in regard to the challenge aim: seeking for low cost, fast and fully automated solutions for cranial implant design. Based on feedback from collaborating neurosurgeons, this paper concludes by stating open issues and post-challenge requirements for intra-operative use. The codes can be found at https://github.com/Jianningli/tmi.
- 3 May 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Medical Imaging
- Vol. 40 (9) , 2329-2342
AutoImplant 2020-First MICCAI Challenge on Automatic Cranial Implant Design
The aim of this paper is to provide a comprehensive overview of the MICCAI 2020 AutoImplant Challenge. The approaches and publications submitted and accepted within the challenge will be summarized and reported, highlighting common algorithmic trends and algorithmic diversity. Furthermore, the evaluation results will be presented, compared and discussed in regard to the challenge aim: seeking for low cost, fast and fully automated solutions for cranial implant design. Based on feedback from collaborating neurosurgeons, this paper concludes by stating open issues and post-challenge requirements for intra-operative use. The codes can be found at https://github.com/Jianningli/tmi.