Format

Send to

Choose Destination
Int J Neural Syst. 2019 Mar 3:1950010. doi: 10.1142/S0129065719500102. [Epub ahead of print]

Scaled Subprofile Modeling and Convolutional Neural Networks for the Identification of Parkinson's Disease in 3D Nuclear Imaging Data.

Author information

1
1 Department of Neurology, University Medical Center Groningen, University of Groningen, Hanzeplein 1, 9713 GZ, Groningen, The Netherlands.
2
2 Faculty of Medical Sciences, University Medical Center Groningen, University of Groningen, A. Deusinglaan 1, Groningen, The Netherlands.
3
3 Institute of Cognitive Sciences and Technologies, Consiglio Nazionale delle Ricerche, via S. Martino della Battaglia, 44-00185 Rome, Italy.
4
4 Department of Nuclear Medicine, Karolinska University Hospital, Huddinge, SE-141 86, Stockholm, Sweden.
5
5 Department of Nuclear Medicine, University Medical Center Groningen, University of Groningen, Hanzeplein 1,9713 GZ Groningen, The Netherlands.
6
6 Department of Neuroscience, Rehabilitation, Opthalmology, Genetics and Maternal and Child Science (DINOGMI), University of Genoa Largo Paolo Daneo 3, 16132 Genoa, Italy.
7
7 IRCCS AOU San Martino - IST, Largo R. Benzi 10, 16132 Genoa, Italy.
8
8 CINAC, HM Puerta del Sur, Avda. de Carlos V 70, 28938 Móstoles (Madrid), Spain.
9
9 CEU Universidad San Pablo, C/Julián Romea 18, 28003 Madrid, Spain.
10
10 CIBERNED, Instituto Carlos III, C/Valderrebollo 5, 28031 Madrid, Spain.
11
11 Department of Neurosciences, Biodonostia Health Research Institute, Begiristain Doktorea Pasealekua, 20014 Donostia-San Sebastián, Guipúzcoa, Spain.
12
12 Nuclear Medicine Unit, Department of Health Sciences (DISSAL), University of Genoa via A. Pastore 1, 16132 Genoa, Italy.
13
13 Department of Neurology, University Medical Center Groningen, University of Groningen, Hanzeplein 1, 9713 GZ Groningen, The Netherlands.

Abstract

Over the last years convolutional neural networks (CNNs) have shown remarkable results in different image classification tasks, including medical imaging. One area that has been less explored with CNNs is Positron Emission Tomography (PET). Fluorodeoxyglucose Positron Emission Tomography (FDG-PET) is a PET technique employed to obtain a representation of brain metabolic function. In this study we employed 3D CNNs in FDG-PET brain images with the purpose of discriminating patients diagnosed with Parkinson's disease (PD) from controls. We employed Scaled Subprofile Modeling using Principal Component Analysis as a preprocessing step to focus on specific brain regions and limit the number of voxels that are used as input for the CNNs, thereby increasing the signal-to-noise ratio in our data. We performed hyperparameter optimization on three CNN architectures to estimate the classification accuracy of the networks on new data. The best performance that we obtained was accuracy = 0.86 and area under the receiver operating characteristic curve (AUC ROC) = 0.94 on the test set. We believe that, with larger datasets, PD patients could be reliably distinguished from controls by FDG-PET scans alone and that this technique could be applied to more clinically challenging tasks, like the differential diagnosis of neurological disorders with similar symptoms, such as PD, Progressive Supranuclear Palsy (PSP) and Multiple System Atrophy (MSA).

KEYWORDS:

Parkinson’s disease; Positron Emission Tomography; convolutional neural networks; principal component analysis

PMID:
31046514
DOI:
10.1142/S0129065719500102

Supplemental Content

Full text links

Icon for Atypon
Loading ...
Support Center