Hugues Favre

Hugues Favre

Auto-generated profile, based on publicly available information

Hugues Favre

Department of Imaging Physics, Delft University of Technology, Delft, the Netherlands

Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
20 March 2023 in Physics in Medicine & Biology
Verified
Journal Article
Closed access
HFHugues Favre
Hugues Favre
MPMathieu Pernot
Mathieu Pernot
MTMickael Tanter
Mickael Tanter
CPClément Papadacci
Clément Papadacci
Objective . Early diagnosis and acute knowledge of cerebral disease require to map the microflows of the whole brain. Recently, ultrasound localization microscopy (ULM) was applied to map and quantify blood microflows in 2D in the brain of adult patients down to the micron scale. Whole brain 3D clinical ULM remains challenging due to the transcranial energy loss which reduces significantly the imaging sensitivity. Approach . Large aperture probes with a large surface can increase both the field of view and sensitivity. However, a large active surface implies thousands of acoustic elements, which limits clinical translation. In a previous simulation study, we developed a new probe concept combining a limited number of elements and a large aperture. It is based on large elements, to increase sensitivity, and a multi-lens diffracting layer to improve the focusing quality. In this study, a 16 elements prototype, driven at 1 MHz frequency, was made and in vitro experiments were performed to validate the imaging capabilities of this new probe concept. Main results . First, pressure fields emitted from a large single transducer element without and with diverging lens were compared. Low directivity was measured for the large element with the diverging lens while maintaining high transmit pressure. The focusing quality of 4 × 3cm matrix arrays of 16 elements without/with lenses were compared. In vitro experiments in a water tank and through a human skull were achieved to localize and track microbubbles in tubes. Significance. ULM was achieved demonstrating the strong potential of multi-lens diffracting layer to enable microcirculation assessment over a large field of view through the bones.

7 April 2022 in Physics in Medicine & Biology
Verified
Journal Article
Closed access
HFHugues Favre
Hugues Favre
MPMathieu Pernot
Mathieu Pernot
MTMickael Tanter
Mickael Tanter
CPClément Papadacci
Clément Papadacci
Mapping blood microflows of the whole brain is crucial for early diagnosis of cerebral diseases. Ultrasound localization microscopy (ULM) was recently applied to map and quantify blood microflows in 2D in the brain of adult patients down to the micron scale. Whole brain 3D clinical ULM remains challenging due to the transcranial energy loss which significantly reduces the imaging sensitivity. Large aperture probes with a large surface can increase both resolution and sensitivity. However, a large active surface implies thousands of acoustic elements, with limited clinical translation. In this study, we investigate via simulations a new high-sensitive 3D imaging approach based on large diverging elements, combined with an adapted beamforming with corrected delay laws, to increase sensitivity. First, pressure fields from single elements with different sizes and shapes were simulated. High directivity was measured for curved element while maintaining high transmit pressure. Matrix arrays of 256 elements with a dimension of 10 × 10 cm with small ( λ /2), large (4 λ ), and curved elements (4 λ ) were compared through point spread functions analysis. A large synthetic microvessel phantom filled with 100 microbubbles per frame was imaged using the matrix arrays in a transcranial configuration. 93% of the bubbles were detected with the proposed approach demonstrating that the multi-lens diffracting layer has a strong potential to enable 3D ULM over a large field of view through the bones.

28 October 2025 in Nature Communications
Verified
Journal Article
Open access
PDF file
NHNabil Haidour
Nabil Haidour
HFHugues Favre
Hugues Favre
PMPhilippe Mateo
Philippe Mateo
JRJuliette Reydet
Juliette Reydet
ABAlain Bizé
Alain Bizé
CPClement Papadacci
Clement Papadacci
Mapping microcirculation at the whole-organ scale in 3D is crucial for understanding vascular pathologies and improving diagnostics. Although 3D ultrasound localization microscopy (ULM) enables microscopic resolution by localizing intravenously injected microbubbles in small animal models, visualizing entire organs in large animals or humans remains challenging due to limited field of view, low sensitivity, and probe technological complexity. Here, we demonstrate how a multi-lens array method overcomes these limitations. Combined with 3D ULM, it maps and quantifies large vascular volumes (up to 120 × 100 × 82 mm³) at high spatial resolution (125–200 µm) with a volumetric acquisition rate of 312 Hz, using low-cost technology. This approach enables deeper insights into hemodynamics from large vessels to pre-capillary arterioles, by providing vast and rich datasets of whole-organ vascularization. It could also facilitate diagnosis of microcirculation disorders and monitoring of small-vessel disease treatments by addressing key limitations of current imaging modalities.

21 July 2025 in IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control
Verified
Journal Article
Open access
HFHugues Favre
Hugues Favre
MBMerijn Berendsen
Merijn Berendsen
RWRick Waasdorp
Rick Waasdorp
DMDavid Maresca
David Maresca
Advances in 4-D ultrasound imaging open new perspectives in biomedical research by reducing the long-standing challenge of operator dependency. Extensive research efforts are focused on developing next generation of 2-D transducer arrays for 4-D imaging. Here, we present a compact 2-D array design based on hexagonal-shaped transducer elements. We demonstrate that 2-D hexagonal arrays provide an optimal compact sampling, resulting in lower grating lobe levels and an improved imaging quality compared to conventional square-shaped transducer element arrays. A prototype array made of hexagonal transducer elements is presented, and its characterization is provided, demonstrating its imaging capabilities.

7 April 2026
Verified
Thesis
Closed access
HFHugues Favre
Hugues Favre
Le réseau vasculaire humain est un système complexe multi-échelles, allant de plusieurs millimètres pour les plus grosses artères à quelques micromètres pour les plus petits vaisseaux. Ce réseau fournit les nutriments et le dioxygène aux organes. Il est donc fortement lié aux fonctions des organes et à leur état de santé. Dans le cas du cerveau, les pathologies neuro-vasculaires sont généralement diagnostiquées à des stades ultérieurs à des évènements cérébraux et à l'apparition de symptômes aigus. Ces limitations cliniques sont principalement imputées aux systèmes d’imagerie de la vascularisation cérébrale, majoritairement réalisée par tomodensitométrie et par imagerie par résonance magnétique. Ces techniques d'imagerie sont difficiles d’accès à cause de leurs coûts importants et elles sont limitées à une résolution millimétrique qui ne permet pas d’accéder à la microcirculation.Pour permettre un diagnostic précoce, la capacité à imager en 3D sur de larges champs de vue, la morphologie et la dynamique du réseau vasculaire à une résolution spatiale micrométrique est essentielle. L’imagerie par ultrason permettrait de répondre à un besoin clinique important pour l’imagerie du réseau vasculaire cérébrale. Son accessibilité et sa capacité à imager le flux sanguin (grâce à sa très bonne résolution temporelle) en font un excellent candidat. Néanmoins, à ce jour, aucun système d’imagerie ultrasonore ne permet de faire une imagerie transcrânienne de l’ensemble du cerveau. Plusieurs barrières technologiques restent à lever dont la sensibilité des transducteurs ultrasonores, la capacité à imager des volumes suffisamment larges et la complexité des systèmes d’imagerie ultrarapide 3D.L’objectif de cette thèse est de concevoir un nouveau concept de sonde échographique de haute sensibilité pour l’imagerie super-résolue transcrânienne 3D à large champ de vue. Cette thèse s’appuie sur l’utilisation des techniques de microscopie par localisation ultrasonore (ULM) basée sur l’injection de microbulles dans le réseau vasculaire pour reconstruire des images super-résolues. Nous avons proposé un nouveau concept de sonde multi-lentilles à larges éléments. L’utilisation de larges éléments multi-lentilles nous permet de bénéficier d’une large surface de sonde (10x10cm2) à nombre réduit d’éléments (256), et de pouvoir imager de larges volumes 3D (10x10x10cm3). Les larges éléments permettent également l’augmentation de la sensibilité grâce aux larges surfaces actives des éléments individuels. L’utilisation de multi-lentilles divergentes permet de diminuer la forte directivité des larges éléments pour maintenir une bonne capacité de focalisation, nécessaire à la reconstruction d’image.Par simulation, nous avons étudié et déterminé les différents paramètres d’une sonde multi-lentilles à larges éléments pilotée à basse fréquence (1MHz). Un cadre de simulation a été proposé pour simuler un fantôme dynamique pour l’imagerie ULM 3D. Nous avons démontré les capacités et les performances d’imagerie ULM 3D de notre nouveau concept de sonde dans un milieu fortement atténuant. Un prototype de sonde multi-lentilles à larges éléments a été construit et tester pour valider les résultats de simulation. Des expériences in vitro et ex vivo ont été entreprises. Un tube sub-longueur d’onde a été imagé à travers d’un crâne humain avec succès.Les résultats de cette thèse ont montré le fort potentiel du concept de sonde multi-lentilles à larges éléments pour cartographier la microcirculation d’organe entier tel que le cerveau, à des résolutions micrométriques. Cette nouvelle approche pourrait ouvrir de nouvelles perspectives cliniques pour le dépistage et le diagnostic précoces, ou pour le suivi des pathologies vasculaires grâce à son cout réduit et son caractère non-ionisant, à son large champ de vue et à sa bonne résolution spatiale.

24 September 2025 in IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control
Verified
Journal Article
Open access
HFHugues Favre
Hugues Favre
DMDavid Maresca
David Maresca
The introduction of genetically encoded gas vesicles (GVs), protein nanostructures with the ability to scatter sound, has created the possibility for deep tissue cellular imaging. GVs establish a platform for biomolecular engineering and were successfully repurposed into acoustic reporter genes and acoustic biosensors. Alongside molecular engineering developments, a method called cross amplitude modulation (xAM) has emerged as the gold standard for non-destructive ultrasound imaging of GVs thanks to its sensitivity and specificity in living biological tissues. Here, we present latest xAM theory and imaging principles. Specifically, we report 1) analytical expressions for the X-wave beam width and primary-to-secondary lobes distance; 2) experimental observations of nondiffractive xAM beams; 3) a method to modulate the secondary lobe level of xAM beams; 4) a demonstration of the incoherent nature of the xAM image noise that can be leverage to increase sensitivity through frame averaging, 5) a beamforming formalism to enhance xAM contrast-to-noise ratio without reducing framerate. Ultimately, the rise of the field of Biomolecular Ultrasound will rest on the co-development of genetically encoded probes and dedicated imaging methods such as xAM and its 3D extension, nonlinear sound-sheet microscopy.