1 code implementation • 5 Feb 2024 • José Morano, Guilherme Aresta, Hrvoje Bogunović
The framework consists of a fully convolutional neural network that recursively refines semantic segmentation maps, correcting manifest classification errors and thus improving topological consistency.
Ranked #1 on Classification on LES-AV
1 code implementation • 2 Feb 2024 • José Morano, Guilherme Aresta, Christoph Grechenig, Ursula Schmidt-Erfurth, Hrvoje Bogunović
The use of multimodal imaging has led to significant improvements in the diagnosis and treatment of many diseases.
no code implementations • 18 Aug 2023 • Botond Fazekas, José Morano, Dmitrii Lachinov, Guilherme Aresta, Hrvoje Bogunović
The Segment Anything Model (SAM) has gained significant attention in the field of image segmentation due to its impressive capabilities and prompt-based interface.
1 code implementation • 6 Jul 2023 • José Morano, Guilherme Aresta, Dmitrii Lachinov, Julia Mai, Ursula Schmidt-Erfurth, Hrvoje Bogunović
Moreover, the proposed SSL method allows further improvement of this performance by up to 23%, and we show that the SSL is beneficial regardless of the network architecture.
no code implementations • 1 Dec 2022 • José Morano, Álvaro S. Hervella, José Rouco, Jorge Novo, José I. Fernández-Vigo, Marcos Ortega
This is possible thanks to the use of a CNN with a custom setting that links the lesions and the diagnosis.
no code implementations • 20 Sep 2022 • José Morano, Álvaro S. Hervella, Jorge Novo, José Rouco
The proposed multi-segmentation method allows to detect more vessels and better segment the different structures, while achieving a competitive classification performance.
no code implementations • 22 May 2022 • José Morano, Álvaro S. Hervella, José Rouco, Jorge Novo, José I. Fernández-Vigo, Marcos Ortega
To overcome these issues, several works have proposed automatic methods for the detection of AMD in retinography images, the most widely used modality for the screening of the disease.
no code implementations • 18 Dec 2020 • José Morano, Álvaro S. Hervella, Noelia Barreira, Jorge Novo, José Rouco
These two issues become specially relevant when applying FCNs to medical image segmentation as, first, the existent models are usually adjusted from broad domain applications over photographic images, and second, the amount of annotated data is usually scarcer.