Voxel-Wise Feature Selection Method for CNN Binary Classification of Neuroimaging Data
Copyright © 2021 Messina, Borrelli, Russo, Salvatore and Aiello..
Voxel-wise group analysis is presented as a novel feature selection (FS) technique for a deep learning (DL) approach to brain imaging data classification. The method, based on a voxel-wise two-sample t-test and denoted as t-masking, is integrated into the learning procedure as a data-driven FS strategy. t-Masking has been introduced in a convolutional neural network (CNN) for the test bench of binary classification of very-mild Alzheimer's disease vs. normal control, using a structural magnetic resonance imaging dataset of 180 subjects. To better characterize the t-masking impact on CNN classification performance, six different experimental configurations were designed. Moreover, the performances of the presented FS method were compared to those of similar machine learning (ML) models that relied on different FS approaches. Overall, our results show an enhancement of about 6% in performance when t-masking was applied. Moreover, the reported performance enhancement was higher with respect to similar FS-based ML models. In addition, evaluation of the impact of t-masking on various selection rates has been provided, serving as a useful characterization for future insights. The proposed approach is also highly generalizable to other DL architectures, neuroimaging modalities, and brain pathologies.
| Medienart: |
E-Artikel |
|---|
| Erscheinungsjahr: |
2021 |
|---|---|
| Erschienen: |
2021 |
| Enthalten in: |
Zur Gesamtaufnahme - volume:15 |
|---|---|
| Enthalten in: |
Frontiers in neuroscience - 15(2021) vom: 18., Seite 630747 |
| Sprache: |
Englisch |
|---|
| Beteiligte Personen: |
Messina, Domenico [VerfasserIn] |
|---|
| Links: |
|---|
| Themen: |
Alzheimer’s disease |
|---|
| Anmerkungen: |
Date Revised 13.05.2021 published: Electronic-eCollection Citation Status PubMed-not-MEDLINE |
|---|
| doi: |
10.3389/fnins.2021.630747 |
|---|
| funding: |
|
|---|---|
| Förderinstitution / Projekttitel: |
|
| PPN (Katalog-ID): |
NLM325108374 |
|---|
| LEADER | 01000naa a22002652 4500 | ||
|---|---|---|---|
| 001 | NLM325108374 | ||
| 003 | DE-627 | ||
| 005 | 20231225191734.0 | ||
| 007 | cr uuu---uuuuu | ||
| 008 | 231225s2021 xx |||||o 00| ||eng c | ||
| 024 | 7 | |a 10.3389/fnins.2021.630747 |2 doi | |
| 028 | 5 | 2 | |a pubmed24n1083.xml |
| 035 | |a (DE-627)NLM325108374 | ||
| 035 | |a (NLM)33958980 | ||
| 040 | |a DE-627 |b ger |c DE-627 |e rakwb | ||
| 041 | |a eng | ||
| 100 | 1 | |a Messina, Domenico |e verfasserin |4 aut | |
| 245 | 1 | 0 | |a Voxel-Wise Feature Selection Method for CNN Binary Classification of Neuroimaging Data |
| 264 | 1 | |c 2021 | |
| 336 | |a Text |b txt |2 rdacontent | ||
| 337 | |a ƒaComputermedien |b c |2 rdamedia | ||
| 338 | |a ƒa Online-Ressource |b cr |2 rdacarrier | ||
| 500 | |a Date Revised 13.05.2021 | ||
| 500 | |a published: Electronic-eCollection | ||
| 500 | |a Citation Status PubMed-not-MEDLINE | ||
| 520 | |a Copyright © 2021 Messina, Borrelli, Russo, Salvatore and Aiello. | ||
| 520 | |a Voxel-wise group analysis is presented as a novel feature selection (FS) technique for a deep learning (DL) approach to brain imaging data classification. The method, based on a voxel-wise two-sample t-test and denoted as t-masking, is integrated into the learning procedure as a data-driven FS strategy. t-Masking has been introduced in a convolutional neural network (CNN) for the test bench of binary classification of very-mild Alzheimer's disease vs. normal control, using a structural magnetic resonance imaging dataset of 180 subjects. To better characterize the t-masking impact on CNN classification performance, six different experimental configurations were designed. Moreover, the performances of the presented FS method were compared to those of similar machine learning (ML) models that relied on different FS approaches. Overall, our results show an enhancement of about 6% in performance when t-masking was applied. Moreover, the reported performance enhancement was higher with respect to similar FS-based ML models. In addition, evaluation of the impact of t-masking on various selection rates has been provided, serving as a useful characterization for future insights. The proposed approach is also highly generalizable to other DL architectures, neuroimaging modalities, and brain pathologies | ||
| 650 | 4 | |a Journal Article | |
| 650 | 4 | |a Alzheimer’s disease | |
| 650 | 4 | |a brain disorders | |
| 650 | 4 | |a deep learning | |
| 650 | 4 | |a feature selection | |
| 650 | 4 | |a magnetic resonance imaging | |
| 650 | 4 | |a neuroimaging | |
| 650 | 4 | |a statistical parametric mapping | |
| 650 | 4 | |a t-masking | |
| 700 | 1 | |a Borrelli, Pasquale |e verfasserin |4 aut | |
| 700 | 1 | |a Russo, Paolo |e verfasserin |4 aut | |
| 700 | 1 | |a Salvatore, Marco |e verfasserin |4 aut | |
| 700 | 1 | |a Aiello, Marco |e verfasserin |4 aut | |
| 773 | 0 | 8 | |i Enthalten in |t Frontiers in neuroscience |d 2007 |g 15(2021) vom: 18., Seite 630747 |w (DE-627)NLM184194563 |x 1662-4548 |7 nnns |
| 773 | 1 | 8 | |g volume:15 |g year:2021 |g day:18 |g pages:630747 |
| 856 | 4 | 0 | |u http://dx.doi.org/10.3389/fnins.2021.630747 |3 Volltext |
| 912 | |a GBV_USEFLAG_A | ||
| 912 | |a GBV_NLM | ||
| 951 | |a AR | ||
| 952 | |d 15 |j 2021 |b 18 |h 630747 | ||