Brain morphometric features predict depression symptom phenotypes in late-life depression using a deep learning model
ObjectivesOur objective was to use deep learning models to identify underlying brain regions associated with depression symptom phenotypes in late-life depression (LLD).ParticipantsDiagnosed with LLD (N = 116) and enrolled in a prospective treatment study.DesignCross-sectional.MeasurementsStructural...
Main Authors: | , , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Frontiers Media S.A.
2023-07-01
|
Series: | Frontiers in Neuroscience |
Subjects: | |
Online Access: | https://www.frontiersin.org/articles/10.3389/fnins.2023.1209906/full |
_version_ | 1797777396774993920 |
---|---|
author | Bing Cao Erkun Yang Lihong Wang Zhanhao Mo David C. Steffens Han Zhang Mingxia Liu Guy G. Potter |
author_facet | Bing Cao Erkun Yang Lihong Wang Zhanhao Mo David C. Steffens Han Zhang Mingxia Liu Guy G. Potter |
author_sort | Bing Cao |
collection | DOAJ |
description | ObjectivesOur objective was to use deep learning models to identify underlying brain regions associated with depression symptom phenotypes in late-life depression (LLD).ParticipantsDiagnosed with LLD (N = 116) and enrolled in a prospective treatment study.DesignCross-sectional.MeasurementsStructural magnetic resonance imaging (sMRI) was used to predict five depression symptom phenotypes from the Hamilton and MADRS depression scales previously derived from factor analysis: (1) Anhedonia, (2) Suicidality, (3) Appetite, (4) Sleep Disturbance, and (5) Anxiety. Our deep learning model was deployed to predict each factor score via learning deep feature representations from 3D sMRI patches in 34 a priori regions-of-interests (ROIs). ROI-level prediction accuracy was used to identify the most discriminative brain regions associated with prediction of factor scores representing each of the five symptom phenotypes.ResultsFactor-level results found significant predictive models for Anxiety and Suicidality factors. ROI-level results suggest the most LLD-associated discriminative regions in predicting all five symptom factors were located in the anterior cingulate and orbital frontal cortex.ConclusionsWe validated the effectiveness of using deep learning approaches on sMRI for predicting depression symptom phenotypes in LLD. We were able to identify deep embedded local morphological differences in symptom phenotypes in the brains of those with LLD, which is promising for symptom-targeted treatment of LLD. Future research with machine learning models integrating multimodal imaging and clinical data can provide additional discriminative information. |
first_indexed | 2024-03-12T23:03:26Z |
format | Article |
id | doaj.art-1c9a494cb587478cbdbf4c6283697298 |
institution | Directory Open Access Journal |
issn | 1662-453X |
language | English |
last_indexed | 2024-03-12T23:03:26Z |
publishDate | 2023-07-01 |
publisher | Frontiers Media S.A. |
record_format | Article |
series | Frontiers in Neuroscience |
spelling | doaj.art-1c9a494cb587478cbdbf4c62836972982023-07-19T06:30:20ZengFrontiers Media S.A.Frontiers in Neuroscience1662-453X2023-07-011710.3389/fnins.2023.12099061209906Brain morphometric features predict depression symptom phenotypes in late-life depression using a deep learning modelBing Cao0Erkun Yang1Lihong Wang2Zhanhao Mo3David C. Steffens4Han Zhang5Mingxia Liu6Guy G. Potter7College of Intelligence and Computing, Tianjin University, Tianjin, ChinaDepartment of Radiology and Biomedical Research Imaging Center, University of North Carolina at Chapel Hill, Chapel Hill, NC, United StatesDepartment of Psychiatry, University of Connecticut School of Medicine, University of Connecticut, Farmington, CT, United StatesDepartment of Radiology, China-Japan Union Hospital of Jilin University, Changchun, ChinaDepartment of Psychiatry, University of Connecticut School of Medicine, University of Connecticut, Farmington, CT, United StatesDepartment of Radiology and Biomedical Research Imaging Center, University of North Carolina at Chapel Hill, Chapel Hill, NC, United StatesDepartment of Radiology and Biomedical Research Imaging Center, University of North Carolina at Chapel Hill, Chapel Hill, NC, United StatesDepartment of Psychiatry and Behavioral Sciences, Duke University Medical Center, Durham, NC, United StatesObjectivesOur objective was to use deep learning models to identify underlying brain regions associated with depression symptom phenotypes in late-life depression (LLD).ParticipantsDiagnosed with LLD (N = 116) and enrolled in a prospective treatment study.DesignCross-sectional.MeasurementsStructural magnetic resonance imaging (sMRI) was used to predict five depression symptom phenotypes from the Hamilton and MADRS depression scales previously derived from factor analysis: (1) Anhedonia, (2) Suicidality, (3) Appetite, (4) Sleep Disturbance, and (5) Anxiety. Our deep learning model was deployed to predict each factor score via learning deep feature representations from 3D sMRI patches in 34 a priori regions-of-interests (ROIs). ROI-level prediction accuracy was used to identify the most discriminative brain regions associated with prediction of factor scores representing each of the five symptom phenotypes.ResultsFactor-level results found significant predictive models for Anxiety and Suicidality factors. ROI-level results suggest the most LLD-associated discriminative regions in predicting all five symptom factors were located in the anterior cingulate and orbital frontal cortex.ConclusionsWe validated the effectiveness of using deep learning approaches on sMRI for predicting depression symptom phenotypes in LLD. We were able to identify deep embedded local morphological differences in symptom phenotypes in the brains of those with LLD, which is promising for symptom-targeted treatment of LLD. Future research with machine learning models integrating multimodal imaging and clinical data can provide additional discriminative information.https://www.frontiersin.org/articles/10.3389/fnins.2023.1209906/fullcross-sectional late-life depressiondeep learningfactor score predictioncognitive impairmentAlzheimer's diseasestructural MRI Frontiers |
spellingShingle | Bing Cao Erkun Yang Lihong Wang Zhanhao Mo David C. Steffens Han Zhang Mingxia Liu Guy G. Potter Brain morphometric features predict depression symptom phenotypes in late-life depression using a deep learning model Frontiers in Neuroscience cross-sectional late-life depression deep learning factor score prediction cognitive impairment Alzheimer's disease structural MRI Frontiers |
title | Brain morphometric features predict depression symptom phenotypes in late-life depression using a deep learning model |
title_full | Brain morphometric features predict depression symptom phenotypes in late-life depression using a deep learning model |
title_fullStr | Brain morphometric features predict depression symptom phenotypes in late-life depression using a deep learning model |
title_full_unstemmed | Brain morphometric features predict depression symptom phenotypes in late-life depression using a deep learning model |
title_short | Brain morphometric features predict depression symptom phenotypes in late-life depression using a deep learning model |
title_sort | brain morphometric features predict depression symptom phenotypes in late life depression using a deep learning model |
topic | cross-sectional late-life depression deep learning factor score prediction cognitive impairment Alzheimer's disease structural MRI Frontiers |
url | https://www.frontiersin.org/articles/10.3389/fnins.2023.1209906/full |
work_keys_str_mv | AT bingcao brainmorphometricfeaturespredictdepressionsymptomphenotypesinlatelifedepressionusingadeeplearningmodel AT erkunyang brainmorphometricfeaturespredictdepressionsymptomphenotypesinlatelifedepressionusingadeeplearningmodel AT lihongwang brainmorphometricfeaturespredictdepressionsymptomphenotypesinlatelifedepressionusingadeeplearningmodel AT zhanhaomo brainmorphometricfeaturespredictdepressionsymptomphenotypesinlatelifedepressionusingadeeplearningmodel AT davidcsteffens brainmorphometricfeaturespredictdepressionsymptomphenotypesinlatelifedepressionusingadeeplearningmodel AT hanzhang brainmorphometricfeaturespredictdepressionsymptomphenotypesinlatelifedepressionusingadeeplearningmodel AT mingxialiu brainmorphometricfeaturespredictdepressionsymptomphenotypesinlatelifedepressionusingadeeplearningmodel AT guygpotter brainmorphometricfeaturespredictdepressionsymptomphenotypesinlatelifedepressionusingadeeplearningmodel |