Identificador persistente para citar o vincular este elemento: http://hdl.handle.net/10553/120766
Título: AI-based segmentation of intraoperative glioblastoma hyperspectral images
Autores/as: La Salvia, Marco
Torti, Emanuele
Gazzoni, Marco
Marenzi, Elisa
León, Raquel 
Ortega, Samuel 
Fabelo, Himar 
Callicó, Gustavo M. 
Leporati, Francesco
Clasificación UNESCO: 3314 Tecnología médica
Palabras clave: Deep Learning
Glioblastoma
Hyperspectral Imaging
Intraoperative Computer-Aided Assistance
Fecha de publicación: 2023
Publicación seriada: Proceedings of SPIE - The International Society for Optical Engineering 
Conferencia: Hyperspectral Imaging and Applications II 2022
Resumen: Glioblastoma surgical resection is a problematic mission for neurosurgeons. Tumour complete resection improves patients healing chances and prognosis, whilst excessive resection could lead to neurological deficits. Nevertheless, surgeons' sight hardly traces the tumour's extent and boundaries. Indeed, most surgical processes result in subtotal resections. Histopathological testing might enable complete tumour elimination, though it is not feasible due to the time required for tissue investigation. Several studies reported tumour cells having unique molecular signatures and properties. Hyperspectral imaging (HSI) is an emerging, non-contact, non-ionizing, label-free and minimally invasive optical imaging technique able to extract information concerning the observed tissue at the molecular level. Here, we exploited extensive data augmentation, transfer learning, the U-Net++ and the DeepLab-V3+ architectures to perform the automatic end-toend segmentation of intraoperative glioblastoma hyperspectral images meeting competitive processing times and segmentation results concerning the gold-standard procedure. Based on ground truths provided by the HELICoiD framework, we dramatically improved HSIs processing times, enabling the end-to-end segmentation of glioblastomas targeting the real-time processing to be employed during open craniotomy in surgery, thus improving the gold-standard ML pipeline. We measured competitive inference times concerning the standard CUDA environment offered by MatLab 2020a. The HELICoiD fastest parallel version took 1.68 s to elaborate the most prominent image of the database, whilst our methodology performs segmentation inference in 0.29 ± 0.17 s, hence being real-time compliant concerning the 21 seconds constraint imposed on processing. Furthermore, we evaluated our segmentation results qualitatively and quantitatively regarding the ground truth produced by HELICoiD.
URI: http://hdl.handle.net/10553/120766
ISBN: 9781510657489
ISSN: 0277-786X
DOI: 10.1117/12.2646782
Fuente: Proceedings of SPIE - The International Society for Optical Engineering [ISSN 0277-786X], v. 12338, (Enero 2023)
Colección:Actas de congresos
Adobe PDF (395,93 kB)
Vista completa

Google ScholarTM

Verifica

Altmetric


Comparte



Exporta metadatos



Los elementos en ULPGC accedaCRIS están protegidos por derechos de autor con todos los derechos reservados, a menos que se indique lo contrario.