open access publication

Article, 2024

Using brain structural neuroimaging measures to predict psychosis onset for individuals at clinical high-risk

Molecular Psychiatry, ISSN 1359-4184, 1476-5578, Volume 29, 5, Pages 1465-1477, 10.1038/s41380-024-02426-7

Contributors

Zhu, Yinghan [1] Maikusa, Norihide 0000-0003-0943-4684 [1] Radua, Joaquim 0000-0003-1240-5438 [2] Sämann, Philipp Gero [3] Fusar-Poli, Paolo 0000-0003-3582-6788 [4] [5] Agartz, Ingrid A 0000-0002-9839-5391 [6] [7] [8] Andreassen, Ole Andreas 0000-0002-4461-3568 [8] Bachman, Peter [9] Baeza, Inmaculada [2] Chen, Xiaogang [10] [11] [12] Choi, Sunah 0000-0001-5666-3485 [13] Corcoran, Cheryl Mary 0000-0002-8902-4353 [14] [15] Ebdrup, Bjørn Hylsebeck 0000-0002-2590-5055 [16] [17] Fortea, Adriana 0000-0003-3681-0841 [18] [19] Garani, Ranjini R [20] Glenthøj, Birte Yding 0000-0003-3056-7262 [16] [17] Glenthøj, Louise Birkedal 0000-0003-3621-8450 [16] [21] Haas, Shalaila Siobhan 0000-0003-1385-1050 [15] Hamilton, Holly K 0000-0002-4058-0949 [22] [23] Hayes, Rebecca A. [9] He, Ying [12] Heekeren, Karsten 0000-0001-5105-1922 [24] [25] [26] Kasai, Kiyoto 0000-0002-4443-4535 [1] Katagiri, Naoyuki 0000-0001-7631-061X [27] Kim, Minah 0000-0001-8668-0817 [13] [28] Kristensen, Tina Dam 0000-0001-9616-762X [17] Kwon, Jun Soo 0000-0002-1060-1462 [13] [28] Lawrie, Stephen M 0000-0002-2444-5675 [29] Lebedeva, Ирина Сергеевна 0000-0002-0649-6663 [30] Lee, Jimmy [31] [32] Loewy, Rachel L [22] Mathalon, Daniel H 0000-0001-6090-4974 [22] [23] McGuire, Philip [33] Mizrahi, Romina [20] Mizuno, Masafumi [34] Møller, Paul 0000-0002-5669-0122 [35] Nemoto, Takahiro [27] Nordholm, Dorte 0000-0002-6383-6169 [16] [21] Omelchenko, Maria A 0000-0001-8343-168X [30] Raghava, Jayachandra Mitta 0000-0001-9008-6757 [16] [17] Røssberg, Jan Ivar 0000-0003-4115-9133 [8] Rössler, Wulf 0000-0003-0049-4533 [25] [26] [36] Salisbury, Dean F 0000-0002-8533-0599 [37] Sasabayashi, Daiki 0000-0001-8512-7438 [38] Smigielski, Lukasz 0000-0002-7428-7644 [25] [26] Sugranyes, Gisela 0000-0002-2545-7862 [2] Takahashi, Tsutomu [38] Tamnes, Christian Krog 0000-0002-9191-6764 [7] [8] Tang, Jinsong 0000-0003-3796-1377 [39] [40] Theodoridou, Anastasia 0000-0003-4792-385X [25] [26] Tomyshev, Alexander S [30] Uhlhaas, Peter J 0000-0002-0892-2224 [36] [41] Værnes, Tor Gunnar [8] [42] Van Amelsvoort, Therese A M J 0000-0003-1135-5133 [43] Waltz, James A. [44] Westlye, Lars Tjelta 0000-0001-8644-956X [8] Zhou, Juan Helen 0000-0002-0180-8648 [45] Thompson, Paul 0000-0002-4720-8867 [46] Hernaus, Dennis 0000-0002-8370-5756 [43] Jalbrzikowski, Maria E [9] [47] Koike, Shinsuke 0000-0002-3375-236X (Corresponding author) [1]

Affiliations

  1. [1] The University of Tokyo
    2. [NORA names: Japan; Asia, East; OECD];
  2. [2] August Pi i Sunyer Biomedical Research Institute
    3. [NORA names: Spain; Europe, EU; OECD];
  3. [3] Max Planck Institute of Psychiatry
    4. [NORA names: Germany; Europe, EU; OECD];
  4. [4] King's College London
    5. [NORA names: United Kingdom; Europe, Non-EU; OECD];
  5. [5] University of Pavia
    6. [NORA names: Italy; Europe, EU; OECD];

Abstract

Machine learning approaches using structural magnetic resonance imaging (sMRI) can be informative for disease classification, although their ability to predict psychosis is largely unknown. We created a model with individuals at CHR who developed psychosis later (CHR-PS+) from healthy controls (HCs) that can differentiate each other. We also evaluated whether we could distinguish CHR-PS+ individuals from those who did not develop psychosis later (CHR-PS-) and those with uncertain follow-up status (CHR-UNK). T1-weighted structural brain MRI scans from 1165 individuals at CHR (CHR-PS+, n = 144; CHR-PS-, n = 793; and CHR-UNK, n = 228), and 1029 HCs, were obtained from 21 sites. We used ComBat to harmonize measures of subcortical volume, cortical thickness and surface area data and corrected for non-linear effects of age and sex using a general additive model. CHR-PS+ (n = 120) and HC (n = 799) data from 20 sites served as a training dataset, which we used to build a classifier. The remaining samples were used external validation datasets to evaluate classifier performance (test, independent confirmatory, and independent group [CHR-PS- and CHR-UNK] datasets). The accuracy of the classifier on the training and independent confirmatory datasets was 85% and 73% respectively. Regional cortical surface area measures-including those from the right superior frontal, right superior temporal, and bilateral insular cortices strongly contributed to classifying CHR-PS+ from HC. CHR-PS- and CHR-UNK individuals were more likely to be classified as HC compared to CHR-PS+ (classification rate to HC: CHR-PS+, 30%; CHR-PS-, 73%; CHR-UNK, 80%). We used multisite sMRI to train a classifier to predict psychosis onset in CHR individuals, and it showed promise predicting CHR-PS+ in an independent sample. The results suggest that when considering adolescent brain development, baseline MRI scans for CHR individuals may be helpful to identify their prognosis. Future prospective studies are required about whether the classifier could be actually helpful in the clinical settings.

Keywords

CHR individuals, Chr, MRI scans, accuracy, adolescent brain development, age, approach, area data, baseline, baseline MRI scans, bilateral insular cortex, brain, brain MRI scans, brain development, classification, classifier, classifier performance, clinical high risk, clinical setting, combat, confirmatory dataset, control, cortex, cortical thickness, data, dataset, development, disease, disease classification, effects of age, evaluate classifier performance, follow-up status, healthy controls, high risk, images, independent samples, individuals, insular cortex, learning approach, machine, machine learning approach, magnetic resonance imaging, measurements, measures of subcortical volumes, model, neuroimaging measures, non-linear effects, non-linear effects of age, onset, performance, prognosis, prospective study, psychosis, psychosis onset, resonance imaging, results, samples, scanning, sets, sex, sites, status, structural brain MRI scans, structural magnetic resonance imaging, structural neuroimaging measures, study, subcortical volumes, surface, surface area data, thickness, training, training dataset, validation dataset, validity, volume

Funders

  • Department of Health and Social Care
  • Japan Society for the Promotion of Science
  • Takeda Science Foundation
  • Japan Agency for Medical Research and Development
  • Medical Research Foundation
  • SENSHIN Medical Research Foundation
  • Japan Science and Technology Agency

Data Provider: Digital Science

LINKS
-

Matching Records in NORA

SUBJECTS
+

METRICS
+