Adopting transfer learning for neuroimaging : a comparative analysis with a custom 3D convolution neural network model
Soliman, Amira (Halmstad University)
Chang, Jose R. (National Cheng Kung University in Tainan)
Etminani, Kobra (Halmstad University)
Byttner, Stefan (Halmstad University)
Davidsson, Anette (Institution of Medicine and Health Sciences)
Martínez-Sanchis, Begoña (Hospital Universitari i Politècnic La Fe (València))
Camacho, Valle (Hospital de la Santa Creu i Sant Pau (Barcelona, Catalunya))
Bauckneht, Matteo (IRCCS Ospedale Policlinico San Martino)
Stegeran, Roxana (Linköping University Hospital)
Ressner, Marcus (Linköping University Hospital)
Agudelo-Cifuentes, Marc (Hospital Universitari i Politècnic La Fe (València))
Chincarini, Andrea (National Institute of Nuclear Physics)
Brendel, Matthias (University Hospital)
Rominger, Axel (University Hospital Bern)
Bruffaerts, Rose (University of Antwerp)
Vandenberghe, Rik (University Hospitals Leuven (Bèlgica))
Kramberger, Milica G. (University Medical Centre)
Trost, Maja (Univerza V Ljubljani)
Nicastro, Nicolas (Geneva University Hospitals (Suïssa))
Frisoni, Giovanni B. (University Hospitals)
Lemstra, Afina W. (VU Medical Center Alzheimer Center)
Berckel, Bart N. M. van (Vrije Universiteit Amsterdam)
Pilotto, Andrea (University of Brescia)
Padovani, Alessandro (IRCCS Ospedale Policlinico San Martino)
Morbelli, Silvia (Stavanger University Hospital)
Aarsland, Dag (King's College London)
Nobili, Flavio (University of Genoa)
Garibotto, Valentina (Geneva University)
Ochoa-Figueroa, Miguel (Linköping University)
Universitat Autònoma de Barcelona

Fecha:	2022
Resumen:	In recent years, neuroimaging with deep learning (DL) algorithms have made remarkable advances in the diagnosis of neurodegenerative disorders. However, applying DL in different medical domains is usually challenged by lack of labeled data. To address this challenge, transfer learning (TL) has been applied to use state-of-the-art convolution neural networks pre-trained on natural images. Yet, there are differences in characteristics between medical and natural images, also image classification and targeted medical diagnosis tasks. The purpose of this study is to investigate the performance of specialized and TL in the classification of neurodegenerative disorders using 3D volumes of 18F-FDG-PET brain scans. Results show that TL models are suboptimal for classification of neurodegenerative disorders, especially when the objective is to separate more than two disorders. Additionally, specialized CNN model provides better interpretations of predicted diagnosis. TL can indeed lead to superior performance on binary classification in timely and data efficient manner, yet for detecting more than a single disorder, TL models do not perform well. Additionally, custom 3D model performs comparably to TL models for binary classification, and interestingly perform better for diagnosis of multiple disorders. The results confirm the superiority of the custom 3D-CNN in providing better explainable model compared to TL adopted ones.
Derechos:	Aquest document està subjecte a una llicència d'ús Creative Commons. Es permet la reproducció total o parcial, la distribució, la comunicació pública de l'obra i la creació d'obres derivades, fins i tot amb finalitats comercials, sempre i quan es reconegui l'autoria de l'obra original.
Lengua:	Anglès
Documento:	Article ; recerca ; Versió publicada
Materia:	Brain Neurodegenerative Disorders ; Convolution Neural Networks ; Medical Image Classification ; Transfer Learning
Publicado en:	BMC Medical Informatics and Decision Making, Vol. 22 (december 2022) , ISSN 1472-6947

DOI: 10.1186/s12911-022-02054-7
PMID: 36476613

15 p, 2.9 MB

El registro aparece en las colecciones:
Artículos > Artículos de investigación
Artículos > Artículos publicados