Full Text

Turn on search term navigation

© 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.

Abstract

Left Ventricle (LV) detection from Cardiac Magnetic Resonance (CMR) imaging is a fundamental step, preliminary to myocardium segmentation and characterization. This paper focuses on the application of a Visual Transformer (ViT), a novel neural network architecture, to automatically detect LV from CMR relaxometry sequences. We implemented an object detector based on the ViT model to identify LV from CMR multi-echo T2* sequences. We evaluated performances differentiated by slice location according to the American Heart Association model using 5-fold cross-validation and on an independent dataset of CMR T2*, T2, and T1 acquisitions. To the best of our knowledge, this is the first attempt to localize LV from relaxometry sequences and the first application of ViT for LV detection. We collected an Intersection over Union (IoU) index of 0.68 and a Correct Identification Rate (CIR) of blood pool centroid of 0.99, comparable with other state-of-the-art methods. IoU and CIR values were significantly lower in apical slices. No significant differences in performances were assessed on independent T2* dataset (IoU = 0.68, p = 0.405; CIR = 0.94, p = 0.066). Performances were significantly worse on the T2 and T1 independent datasets (T2: IoU = 0.62, CIR = 0.95; T1: IoU = 0.67, CIR = 0.98), but still encouraging considering the different types of acquisition. This study confirms the feasibility of the application of ViT architectures in LV detection and defines a benchmark for relaxometry imaging.

Details

Title
Left Ventricle Detection from Cardiac Magnetic Resonance Relaxometry Images Using Visual Transformer
Author
De Santi, Lisa Anita 1   VIAFID ORCID Logo  ; Meloni, Antonella 2   VIAFID ORCID Logo  ; Maria Filomena Santarelli 3   VIAFID ORCID Logo  ; Pistoia, Laura 4   VIAFID ORCID Logo  ; Spasiano, Anna 5 ; Casini, Tommaso 6   VIAFID ORCID Logo  ; Putti, Maria Caterina 7 ; Cuccia, Liana 8 ; Cademartiri, Filippo 4   VIAFID ORCID Logo  ; Positano, Vincenzo 2   VIAFID ORCID Logo 

 Department of Information Engineering, University of Pisa, 56122 Pisa, Italy; [email protected]; U.O.C. Bioingegneria, Fondazione G. Monasterio CNR-Regione Toscana, 56124 Pisa, Italy; [email protected] 
 U.O.C. Bioingegneria, Fondazione G. Monasterio CNR-Regione Toscana, 56124 Pisa, Italy; [email protected]; Department of Radiology, Fondazione G. Monasterio CNR-Regione Toscana, 56124 Pisa, Italy; [email protected] (L.P.); 
 CNR Institute of Clinical Physiology, 56124 Pisa, Italy; [email protected] 
 Department of Radiology, Fondazione G. Monasterio CNR-Regione Toscana, 56124 Pisa, Italy; [email protected] (L.P.); 
 Unità Operativa Semplice Dipartimentale Malattie Rare del Globulo Rosso, Azienda Ospedaliera di Rilievo Nazionale “A. Cardarelli”, 80131 Napoli, Italy 
 Centro Talassemie ed Emoglobinopatie, Ospedale “Meyer”, 50139 Firenze, Italy 
 Clinica di Emato-Oncologia Pediatrica, Dipartimento di Salute della Donna e del Bambino, Azienda Ospedale Università, 35128 Padova, Italy 
 Unità Operativa Complessa Ematologia con Talassemia, ARNAS Civico “Benfratelli-Di Cristina”, 90127 Palermo, Italy 
First page
3321
Publication year
2023
Publication date
2023
Publisher
MDPI AG
e-ISSN
14248220
Source type
Scholarly Journal
Language of publication
English
ProQuest document ID
2791721671
Copyright
© 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.