Abstract
BACKGROUND: Schizophrenia (SZ) and bipolar disorder (BD) share substantial neurodevelopmental components affecting brain maturation and architecture. This necessitates a dynamic lifespan perspective in which brain aberrations are inferred from deviations from expected lifespan trajectories. We applied machine learning to diffusion tensor imaging (DTI) indices of white matter structure and organization to estimate and compare brain age between patients with SZ, patients with BD, and healthy control (HC) subjects across 10 cohorts.
METHODS: We trained 6 cross-validated models using different combinations of DTI data from 927 HC subjects (18-94 years of age) and applied the models to the test sets including 648 patients with SZ (18-66 years of age), 185 patients with BD (18-64 years of age), and 990 HC subjects (17-68 years of age), estimating the brain age for each participant. Group differences were assessed using linear models, accounting for age, sex, and scanner. A meta-analytic framework was applied to assess the heterogeneity and generalizability of the results.
RESULTS: Tenfold cross-validation revealed high accuracy for all models. Compared with HC subjects, the model including all feature sets significantly overestimated the age of patients with SZ (Cohen's d = -20.29) and patients with BD (Cohen's d = 0.18), with similar effects for the other models. The meta-analysis converged on the same findings. Fractional anisotropy-based models showed larger group differences than the models based on other DTI-derived metrics.
CONCLUSIONS: Brain age prediction based on DTI provides informative and robust proxies for brain white matter integrity. Our results further suggest that white matter aberrations in SZ and BD primarily consist of anatomically distributed deviations from expected lifespan trajectories that generalize across cohorts and scanners.
Original language | English |
---|---|
Pages (from-to) | 1095-1103 |
Number of pages | 9 |
Journal | Biological Psychiatry: Cognitive Neuroscience and Neuroimaging |
Volume | 5 |
Issue number | 12 |
DOIs | |
Publication status | Published - Dec 2020 |
Keywords
- ABNORMALITIES
- CORPUS-CALLOSUM
- DEFICITS
- GRAY
- INTEGRITY
- METAANALYSIS
- MICROSTRUCTURE
- MOVEMENT
- MYELINATION
- PATTERNS
- ONSET
Access to Document
- 10.1016/j.bpsc.2020.06.014Licence: CC BY
Cite this
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver
}
In: Biological Psychiatry: Cognitive Neuroscience and Neuroimaging, Vol. 5, No. 12, 12.2020, p. 1095-1103.
Research output: Contribution to journal › Article › Academic › peer-review
TY - JOUR
T1 - Brain Age Prediction Reveals Aberrant Brain White Matter in Schizophrenia and Bipolar Disorder
T2 - A Multisample Diffusion Tensor Imaging Study
AU - Tonnesen, Siren
AU - Kaufmann, Tobias
AU - de Lange, Ann-Marie G.
AU - Richard, Genevieve
AU - Nhat Trung Doan, null
AU - Alnaes, Dag
AU - van der Meer, Dennis
AU - Rokicki, Jaroslav
AU - Moberget, Torgeir
AU - Maximov, Ivan I.
AU - Agartz, Ingrid
AU - Aminoff, Sofie R.
AU - Beck, Dani
AU - Barch, Deanna M.
AU - Beresniewicz, Justyna
AU - Cervenka, Simon
AU - Fatouros-Bergman, Helena
AU - Craven, Alexander R.
AU - Flyckt, Lena
AU - Gurholt, Tiril P.
AU - Haukvik, Unn K.
AU - Hugdahl, Kenneth
AU - Johnsen, Erik
AU - Jonsson, Erik G.
AU - Kolskar, Knut K.
AU - Kroken, Rune Andreas
AU - Lagerberg, Trine
AU - Loberg, Else-Marie
AU - Nordvik, Jan Egil
AU - Sanders, Anne-Marthe
AU - Ulrichsen, Kristine
AU - Andreassen, Ole A.
AU - Westlye, Lars T.
AU - Karolinska Schizophrenia Project Consortium
N1 - Funding Information: This work was supported by South-Eastern Norway Regional Health Authority Grant Nos. 2014097 , 2015073 , 2016083 , 2016044 , 2017112, 2019107, 2020086, and 2019101 ; Western Norway Regional Health Authority Grant Nos. 911820 and 911679 ; Research Council of Norway Grant Nos. 213700 , 204966 , 249795 , 223273 , 213727 , 286838, 298646 , and 300768 ; KG Jebsen Stiftelsen ; European Commission’s 7th Framework Programme Grant No. 602450 (IMAGEMEND); and the European Research Council under European Union Horizon 2020 Research and Innovation Program Grant No. ERC StG 802998 . The Karolinska Schizophrenia Project was supported by Swedish Medical Research Council Grant Nos. SE: 2009-7053 , 2013-2838 , and SC: 523-2014-3467 ; the Swedish Brain Foundation ; Åhlén-siftelsen ; Svenska Läkaresällskapet ; Petrus och Augusta Hedlunds Stiftelse ; Torsten Söderbergs Stiftelse ; the AstraZeneca-Karolinska Institutet Joint Research Program in Translational Science; So¨derbergs Ko¨nigska Stiftelse; Professor Bror Gadelius; Knut och Alice Wallenbergs stiftelse; Stockholm County Council (ALF and PPG grants); the Centre for Psychiatry Research ; and KID funding from the Karolinska Institutet. Data collection and sharing for this project was provided by the Cambridge Centre for Ageing and Neuroscience. Cambridge Centre for Ageing and Neuroscience funding was provided by the UK Biotechnology and Biological Sciences Research Council Grant No. BB/H008217/1 ), together with support from the UK Medical Research Council and University of Cambridge , United Kingdom. The Conte Centers for the Neuroscience of Mental Disorders were supported through National Institutes of Health Grants P50 MH071616 and R01 MH56584 . Consortium for Neuropsychiatric Phenomics (CNP) was supported by the Consortium for Neuropsychiatric Phenomics (National Institutes of Health Roadmap for Medical Research Grant Nos. UL1-DE019580, RL1MH083268, RL1MH083269, RL1DA024853, RL1MH083270, RL1LM009833, PL1MH083271, and PL1NS062410). The HUBIN (Human Brain Informatics) project was supported by the Swedish Research Council Grant Nos. 2006-2992 , 2006-986 , K2007-62X-15077-04-1 , 2008-2167 , K2008-62P-20597-01-3 , K2010-62X-15078-07-2 , K2012-61X-15078-09-3 , 2017-00949 , and K2015-62X-15077-12-3 ; the regional agreement on medical training and clinical research between Stockholm County Council and the Karolinska Institutet; and the Knut and Alice Wallenberg Foundation. StrokeMRI was supported by the Research Council of Norway Grant Nos. 249795 and 248238 ; South-Eastern Norway Regional Health Authority Grant Nos. 2014097 , 2015044 , 2015073 , and 2016083 ; and Norwegian ExtraFoundation for Health and Rehabilitation Grant No. 2015/FO5146 . Data collection and sharing for the NMorphCH (Neuromorphometry by Computer Algorithm Chicago) project was funded by National Institute of Mental Health Grant No. A R01 MH056584 . The BergenPsykose project was supported by the European Research Council Grant No. ERC AdG 693124 and Western Norway Health-Authorities Grant No. 912045 . Funding Information: This work was supported by South-Eastern Norway Regional Health Authority Grant Nos. 2014097, 2015073, 2016083, 2016044, 2017112, 2019107, 2020086, and 2019101; Western Norway Regional Health Authority Grant Nos. 911820 and 911679; Research Council of Norway Grant Nos. 213700, 204966, 249795, 223273, 213727, 286838, 298646, and 300768; KG Jebsen Stiftelsen; European Commission's 7th Framework Programme Grant No. 602450 (IMAGEMEND); and the European Research Council under European Union Horizon 2020 Research and Innovation Program Grant No. ERC StG 802998. The Karolinska Schizophrenia Project was supported by Swedish Medical Research Council Grant Nos. SE: 2009-7053, 2013-2838, and SC: 523-2014-3467; the Swedish Brain Foundation; Åhlén-siftelsen; Svenska Läkaresällskapet; Petrus och Augusta Hedlunds Stiftelse; Torsten Söderbergs Stiftelse; the AstraZeneca-Karolinska Institutet Joint Research Program in Translational Science; So¨derbergs Ko¨nigska Stiftelse; Professor Bror Gadelius; Knut och Alice Wallenbergs stiftelse; Stockholm County Council (ALF and PPG grants); the Centre for Psychiatry Research; and KID funding from the Karolinska Institutet. Data collection and sharing for this project was provided by the Cambridge Centre for Ageing and Neuroscience. Cambridge Centre for Ageing and Neuroscience funding was provided by the UK Biotechnology and Biological Sciences Research Council Grant No. BB/H008217/1), together with support from the UK Medical Research Council and University of Cambridge, United Kingdom. The Conte Centers for the Neuroscience of Mental Disorders were supported through National Institutes of Health Grants P50 MH071616 and R01 MH56584. Consortium for Neuropsychiatric Phenomics (CNP) was supported by the Consortium for Neuropsychiatric Phenomics (National Institutes of Health Roadmap for Medical Research Grant Nos. UL1-DE019580, RL1MH083268, RL1MH083269, RL1DA024853, RL1MH083270, RL1LM009833, PL1MH083271, and PL1NS062410). The HUBIN (Human Brain Informatics) project was supported by the Swedish Research Council Grant Nos. 2006-2992, 2006-986, K2007-62X-15077-04-1, 2008-2167, K2008-62P-20597-01-3, K2010-62X-15078-07-2, K2012-61X-15078-09-3, 2017-00949, and K2015-62X-15077-12-3; the regional agreement on medical training and clinical research between Stockholm County Council and the Karolinska Institutet; and the Knut and Alice Wallenberg Foundation. StrokeMRI was supported by the Research Council of Norway Grant Nos. 249795 and 248238; South-Eastern Norway Regional Health Authority Grant Nos. 2014097, 2015044, 2015073, and 2016083; and Norwegian ExtraFoundation for Health and Rehabilitation Grant No. 2015/FO5146. Data collection and sharing for the NMorphCH (Neuromorphometry by Computer Algorithm Chicago) project was funded by National Institute of Mental Health Grant No. A R01 MH056584. The BergenPsykose project was supported by the European Research Council Grant No. ERC AdG 693124 and Western Norway Health-Authorities Grant No. 912045. This article was published as a preprint on bioRxiv: doi: https://www.biorxiv.org/content/10.1101/607754v1. KH and ARC own shares in NordicNeuroLab, Inc. which produced add-on hardware for acquisition of data at the Bergen site. All other authors report no biomedical financial interests or potential conflicts of interest. Publisher Copyright: © 2020 Society of Biological Psychiatry
PY - 2020/12
Y1 - 2020/12
N2 - BACKGROUND: Schizophrenia (SZ) and bipolar disorder (BD) share substantial neurodevelopmental components affecting brain maturation and architecture. This necessitates a dynamic lifespan perspective in which brain aberrations are inferred from deviations from expected lifespan trajectories. We applied machine learning to diffusion tensor imaging (DTI) indices of white matter structure and organization to estimate and compare brain age between patients with SZ, patients with BD, and healthy control (HC) subjects across 10 cohorts.METHODS: We trained 6 cross-validated models using different combinations of DTI data from 927 HC subjects (18-94 years of age) and applied the models to the test sets including 648 patients with SZ (18-66 years of age), 185 patients with BD (18-64 years of age), and 990 HC subjects (17-68 years of age), estimating the brain age for each participant. Group differences were assessed using linear models, accounting for age, sex, and scanner. A meta-analytic framework was applied to assess the heterogeneity and generalizability of the results.RESULTS: Tenfold cross-validation revealed high accuracy for all models. Compared with HC subjects, the model including all feature sets significantly overestimated the age of patients with SZ (Cohen's d = -20.29) and patients with BD (Cohen's d = 0.18), with similar effects for the other models. The meta-analysis converged on the same findings. Fractional anisotropy-based models showed larger group differences than the models based on other DTI-derived metrics.CONCLUSIONS: Brain age prediction based on DTI provides informative and robust proxies for brain white matter integrity. Our results further suggest that white matter aberrations in SZ and BD primarily consist of anatomically distributed deviations from expected lifespan trajectories that generalize across cohorts and scanners.
AB - BACKGROUND: Schizophrenia (SZ) and bipolar disorder (BD) share substantial neurodevelopmental components affecting brain maturation and architecture. This necessitates a dynamic lifespan perspective in which brain aberrations are inferred from deviations from expected lifespan trajectories. We applied machine learning to diffusion tensor imaging (DTI) indices of white matter structure and organization to estimate and compare brain age between patients with SZ, patients with BD, and healthy control (HC) subjects across 10 cohorts.METHODS: We trained 6 cross-validated models using different combinations of DTI data from 927 HC subjects (18-94 years of age) and applied the models to the test sets including 648 patients with SZ (18-66 years of age), 185 patients with BD (18-64 years of age), and 990 HC subjects (17-68 years of age), estimating the brain age for each participant. Group differences were assessed using linear models, accounting for age, sex, and scanner. A meta-analytic framework was applied to assess the heterogeneity and generalizability of the results.RESULTS: Tenfold cross-validation revealed high accuracy for all models. Compared with HC subjects, the model including all feature sets significantly overestimated the age of patients with SZ (Cohen's d = -20.29) and patients with BD (Cohen's d = 0.18), with similar effects for the other models. The meta-analysis converged on the same findings. Fractional anisotropy-based models showed larger group differences than the models based on other DTI-derived metrics.CONCLUSIONS: Brain age prediction based on DTI provides informative and robust proxies for brain white matter integrity. Our results further suggest that white matter aberrations in SZ and BD primarily consist of anatomically distributed deviations from expected lifespan trajectories that generalize across cohorts and scanners.
KW - ABNORMALITIES
KW - CORPUS-CALLOSUM
KW - DEFICITS
KW - GRAY
KW - INTEGRITY
KW - METAANALYSIS
KW - MICROSTRUCTURE
KW - MOVEMENT
KW - MYELINATION
KW - PATTERNS
KW - ONSET
U2 - 10.1016/j.bpsc.2020.06.014
DO - 10.1016/j.bpsc.2020.06.014
M3 - Article
C2 - 32859549
SN - 2451-9022
VL - 5
SP - 1095
EP - 1103
JO - Biological Psychiatry: Cognitive Neuroscience and Neuroimaging
JF - Biological Psychiatry: Cognitive Neuroscience and Neuroimaging
IS - 12
ER -