Estimating Gender and Age from Brain Structural MRI of Children and Adolescents: A 3D Convolutional Neural Network Multitask Learning Model

Sergio Leonardo Mendes; Walter Hugo Lopez Pinaya; Pedro Pan; João Ricardo Sato

doi:10.1155/2021/5550914

Estimating Gender and Age from Brain Structural MRI of Children and Adolescents: A 3D Convolutional Neural Network Multitask Learning Model

Sergio Leonardo Mendes, Walter Hugo Lopez Pinaya, Pedro Pan, João Ricardo Sato^*

^*Corresponding author for this work

Biomedical Engineering Department

Research output: Contribution to journal › Article › peer-review

7 Citations (Scopus)

Abstract

Despite recent advances, assessing biological measurements for neuropsychiatric disorders is still a challenge, where confounding variables such as gender and age (as a proxy for neurodevelopment) play an important role. This study explores brain structural magnetic resonance imaging (sMRI) from two public data sets (ABIDE-II and ADHD-200) with healthy control (HC, N = 894), autism spectrum disorder (ASD, N = 251), and attention deficit hyperactivity disorder (ADHD, N = 357) individuals. We used gray and white matter preprocessed via voxel-based morphometry (VBM) to train a 3D convolutional neural network with a multitask learning strategy to estimate gender, age, and mental health status from structural brain differences. Gradient-based methods were employed to generate attention maps, providing clinically relevant identification of most representative brain regions for models' decision-making. This approach resulted in satisfactory predictions for gender and age. ADHD-200-trained models, evaluated in 10-fold cross-validation procedures on test set, obtained a mean absolute error (MAE) of 1.43 years (±0.22 SD) for age prediction and an area under the curve (AUC) of 0.85 (±0.04 SD) for gender classification. In out-of-sample validation, the best-performing ADHD-200 models satisfactorily predicted age (MAE = 1.57 years) and gender (AUC = 0.89) in the ABIDE-II data set. The models' accuracy was in line with the current state-of-the-art machine learning applications in neuroimaging. Key regions for models' accuracy were presented as a meaningful graphical output. New implementations, such as the use of VBM along with a 3D convolutional neural network multitask learning model and a brain imaging graphical output, reinforce the relevance of the proposed workflow.

Original language	English
Article number	5550914
Journal	Computational Intelligence and Neuroscience
Volume	2021
DOIs	https://doi.org/10.1155/2021/5550914
Publication status	Published - 2021

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1155/2021/5550914

Cite this

@article{56d961d7f1ee4451be8579992dd8e788,

title = "Estimating Gender and Age from Brain Structural MRI of Children and Adolescents: A 3D Convolutional Neural Network Multitask Learning Model",

abstract = "Despite recent advances, assessing biological measurements for neuropsychiatric disorders is still a challenge, where confounding variables such as gender and age (as a proxy for neurodevelopment) play an important role. This study explores brain structural magnetic resonance imaging (sMRI) from two public data sets (ABIDE-II and ADHD-200) with healthy control (HC, N = 894), autism spectrum disorder (ASD, N = 251), and attention deficit hyperactivity disorder (ADHD, N = 357) individuals. We used gray and white matter preprocessed via voxel-based morphometry (VBM) to train a 3D convolutional neural network with a multitask learning strategy to estimate gender, age, and mental health status from structural brain differences. Gradient-based methods were employed to generate attention maps, providing clinically relevant identification of most representative brain regions for models' decision-making. This approach resulted in satisfactory predictions for gender and age. ADHD-200-trained models, evaluated in 10-fold cross-validation procedures on test set, obtained a mean absolute error (MAE) of 1.43 years (±0.22 SD) for age prediction and an area under the curve (AUC) of 0.85 (±0.04 SD) for gender classification. In out-of-sample validation, the best-performing ADHD-200 models satisfactorily predicted age (MAE = 1.57 years) and gender (AUC = 0.89) in the ABIDE-II data set. The models' accuracy was in line with the current state-of-the-art machine learning applications in neuroimaging. Key regions for models' accuracy were presented as a meaningful graphical output. New implementations, such as the use of VBM along with a 3D convolutional neural network multitask learning model and a brain imaging graphical output, reinforce the relevance of the proposed workflow. ",

author = "Mendes, {Sergio Leonardo} and Pinaya, {Walter Hugo Lopez} and Pedro Pan and Sato, {Jo{\~a}o Ricardo}",

year = "2021",

doi = "10.1155/2021/5550914",

language = "English",

volume = "2021",

journal = "Computational Intelligence and Neuroscience",

issn = "1687-5265",

publisher = "Hindawi Publishing Corporation",

}

TY - JOUR

T1 - Estimating Gender and Age from Brain Structural MRI of Children and Adolescents

T2 - A 3D Convolutional Neural Network Multitask Learning Model

AU - Mendes, Sergio Leonardo

AU - Pinaya, Walter Hugo Lopez

AU - Pan, Pedro

AU - Sato, João Ricardo

PY - 2021

Y1 - 2021

N2 - Despite recent advances, assessing biological measurements for neuropsychiatric disorders is still a challenge, where confounding variables such as gender and age (as a proxy for neurodevelopment) play an important role. This study explores brain structural magnetic resonance imaging (sMRI) from two public data sets (ABIDE-II and ADHD-200) with healthy control (HC, N = 894), autism spectrum disorder (ASD, N = 251), and attention deficit hyperactivity disorder (ADHD, N = 357) individuals. We used gray and white matter preprocessed via voxel-based morphometry (VBM) to train a 3D convolutional neural network with a multitask learning strategy to estimate gender, age, and mental health status from structural brain differences. Gradient-based methods were employed to generate attention maps, providing clinically relevant identification of most representative brain regions for models' decision-making. This approach resulted in satisfactory predictions for gender and age. ADHD-200-trained models, evaluated in 10-fold cross-validation procedures on test set, obtained a mean absolute error (MAE) of 1.43 years (±0.22 SD) for age prediction and an area under the curve (AUC) of 0.85 (±0.04 SD) for gender classification. In out-of-sample validation, the best-performing ADHD-200 models satisfactorily predicted age (MAE = 1.57 years) and gender (AUC = 0.89) in the ABIDE-II data set. The models' accuracy was in line with the current state-of-the-art machine learning applications in neuroimaging. Key regions for models' accuracy were presented as a meaningful graphical output. New implementations, such as the use of VBM along with a 3D convolutional neural network multitask learning model and a brain imaging graphical output, reinforce the relevance of the proposed workflow.

AB - Despite recent advances, assessing biological measurements for neuropsychiatric disorders is still a challenge, where confounding variables such as gender and age (as a proxy for neurodevelopment) play an important role. This study explores brain structural magnetic resonance imaging (sMRI) from two public data sets (ABIDE-II and ADHD-200) with healthy control (HC, N = 894), autism spectrum disorder (ASD, N = 251), and attention deficit hyperactivity disorder (ADHD, N = 357) individuals. We used gray and white matter preprocessed via voxel-based morphometry (VBM) to train a 3D convolutional neural network with a multitask learning strategy to estimate gender, age, and mental health status from structural brain differences. Gradient-based methods were employed to generate attention maps, providing clinically relevant identification of most representative brain regions for models' decision-making. This approach resulted in satisfactory predictions for gender and age. ADHD-200-trained models, evaluated in 10-fold cross-validation procedures on test set, obtained a mean absolute error (MAE) of 1.43 years (±0.22 SD) for age prediction and an area under the curve (AUC) of 0.85 (±0.04 SD) for gender classification. In out-of-sample validation, the best-performing ADHD-200 models satisfactorily predicted age (MAE = 1.57 years) and gender (AUC = 0.89) in the ABIDE-II data set. The models' accuracy was in line with the current state-of-the-art machine learning applications in neuroimaging. Key regions for models' accuracy were presented as a meaningful graphical output. New implementations, such as the use of VBM along with a 3D convolutional neural network multitask learning model and a brain imaging graphical output, reinforce the relevance of the proposed workflow.

UR - http://www.scopus.com/inward/record.url?scp=85107684771&partnerID=8YFLogxK

U2 - 10.1155/2021/5550914

DO - 10.1155/2021/5550914

M3 - Article

AN - SCOPUS:85107684771

SN - 1687-5265

VL - 2021

JO - Computational Intelligence and Neuroscience

JF - Computational Intelligence and Neuroscience

M1 - 5550914

ER -

Estimating Gender and Age from Brain Structural MRI of Children and Adolescents: A 3D Convolutional Neural Network Multitask Learning Model

Abstract

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this