Title :

Deep Learning on Bone Scintigraphy to Detect Abnormal Cardiac Uptake at Risk of Cardiac Amyloidosis

Author(s) :

Delbarre, Marc-Antoine [Auteur]
CHU Amiens-Picardie
Girardon, François [Auteur]
CODOC SAS [Paris] [CS]
Roquette, Lucien [Auteur]
CODOC SAS [Paris] [CS]
Blanc-Durand, Paul [Auteur]
Hôpital Henri Mondor
Hubaut, Marc-Antoine [Auteur]
Hôpital Roger Salengro [Lille]
Hachulla, Éric [Auteur]
Institute for Translational Research in Inflammation - U 1286 [INFINITE (Ex-Liric)]
Semah, Franck [Auteur]
Troubles cognitifs vasculaires et dégénératifs
Huglo, Damien [Auteur]
Thérapies Laser Assistées par l'Image pour l'Oncologie - U 1189 [ONCO-THAI]
Garcelon, Nicolas [Auteur]
Health data- and model- driven Knowledge Acquisition [HeKA]
Centre de Recherche des Cordeliers [CRC (UMR_S_1138 / U1138)]
Marchal, Etienne [Auteur]
CHU Amiens-Picardie
El Esper, Isabelle [Auteur]
CHU Amiens-Picardie
Tribouilloy, Christophe [Auteur]
CHU Amiens-Picardie
Mécanismes physiopathologiques et conséquences des calcifications vasculaires - UR UPJV 7517 [MP3CV]
Lamblin, Nicolas [Auteur]
Facteurs de Risque et Déterminants Moléculaires des Maladies liées au Vieillissement - U 1167 [RID-AGE]
Duhaut, Pierre [Auteur]
Mécanismes physiopathologiques et conséquences des calcifications vasculaires - UR UPJV 7517 [MP3CV]
CHU Amiens-Picardie
Schmidt, Jean [Auteur]
Mécanismes physiopathologiques et conséquences des calcifications vasculaires - UR UPJV 7517 [MP3CV]
CHU Amiens-Picardie
Itti, Emmanuel [Auteur]
CHU Henri Mondor [Créteil]
Damy, Thibaud [Auteur]
CHU Henri Mondor [Créteil]

Journal title :

JACC: Cardiovascular Imaging

Publisher :

Elsevier/American College of Cardiology

Publication date :

2023-04

ISSN :

1936-878X

HAL domain(s) :

Sciences du Vivant [q-bio]/Médecine humaine et pathologie

English abstract : [en]

BackgroundCardiac uptake on technetium-99m whole-body scintigraphy (WBS) is almost pathognomonic of transthyretin cardiac amyloidosis. The rare false positives are often related to light-chain cardiac amyloidosis. However, ...
Show more >BackgroundCardiac uptake on technetium-99m whole-body scintigraphy (WBS) is almost pathognomonic of transthyretin cardiac amyloidosis. The rare false positives are often related to light-chain cardiac amyloidosis. However, this scintigraphic feature remains largely unknown, leading to misdiagnosis despite characteristic images. A retrospective review of all WBSs in a hospital database to detect those with cardiac uptake may allow the identification of undiagnosed patients.ObjectivesThe authors sought to develop and validate the first deep learning–based model that automatically detects significant cardiac uptake (≥Perugini grade 2) on WBS from large hospital databases in order to retrieve patients at risk of cardiac amyloidosis.MethodsThe model is based on a convolutional neural network with image-level labels. The performance evaluation was performed with C-statistics using a 5-fold cross-validation scheme stratified so that the proportion of positive and negative WBSs remained constant across folds and using an external validation data set.ResultsThe training data set consisted of 3,048 images: 281 positives (≥Perugini 2) and 2,767 negatives. The external validation data set consisted of 1,633 images: 102 positives and 1,531 negatives. The performance of the 5-fold cross-validation and external validation was as follows: 98.9% (± 1.0) and 96.1% for sensitivity, 99.5% (± 0.4) and 99.5% for specificity, and 0.999 (SD = 0.000) and 0.999 for the area under the curve of the receiver-operating characteristic curves. Sex, age <90 years, body mass index, injection-acquisition delay, radionuclides, and the indication of WBS only slightly affected performances.ConclusionsThe authors’ detection model is effective at identifying patients with cardiac uptake ≥Perugini 2 on WBS and may help in the diagnosis of patients with cardiac amyloidosis.Show less >

Language :

Anglais

Peer reviewed article :

Oui

Audience :

Internationale

Popular science :

Non

Collections :

• Thérapies Lasers Assistées par l'Image pour l'Oncologie (ONCO-THAI) - U1189

Source :

Harvested from HAL