Using brain structural neuroimaging measures to predict psychosis onset for individuals at clinical high-risk
Yinghan Zhu,Norihide Maikusa,Joaquim Radua,Philipp G. Sämann,Paolo Fusar-Poli,Ingrid Agartz,Ole A. Andreassen,Peter Bachman,Inmaculada Baeza,Xiaogang Chen,Sunah Choi,Cheryl M. Corcoran,Bjørn H. Ebdrup,Adriana Fortea,Ranjini RG. Garani,Birte Yding Glenthøj,Louise Birkedal Glenthøj,Shalaila S. Haas,Holly K. Hamilton,Rebecca A. Hayes,Ying He,Karsten Heekeren,Kiyoto Kasai,Naoyuki Katagiri,Minah Kim,Tina D. Kristensen,Jun Soo Kwon,Stephen M. Lawrie,Irina Lebedeva,Jimmy Lee,Rachel L. Loewy,Daniel H. Mathalon,Philip McGuire,Romina Mizrahi,Masafumi Mizuno,Paul Møller,Takahiro Nemoto,Dorte Nordholm,Maria A. Omelchenko,Jayachandra M. Raghava,Jan I. Røssberg,Wulf Rössler,Dean F. Salisbury,Daiki Sasabayashi,Lukasz Smigielski,Gisela Sugranyes,Tsutomu Takahashi,Christian K. Tamnes,Jinsong Tang,Anastasia Theodoridou,Alexander S. Tomyshev,Peter J. Uhlhaas,Tor G. Værnes,Therese A. M. J. van Amelsvoort,James A. Waltz,Lars T. Westlye,Juan H. Zhou,Paul M. Thompson,Dennis Hernaus,Maria Jalbrzikowski,Shinsuke Koike,Paul Allen,Helen Baldwin,Sabrina Catalano,Michael W. L. Chee,Kang Ik K. Cho,Lieuwe de Haan,Leslie E. Horton,Mallory J. Klaunig,Yoo Bin Kwak,Xiaoqian Ma,Merete Nordentoft,Lijun Ouyang,Jose C. Pariente,Franz Resch,Jason Schiffman,Mikkel E. Sørensen,Michio Suzuki,Sophia Vinogradov,Christina Wenneberg,Hidenori Yamasue,Liu Yuan
DOI: https://doi.org/10.1038/s41380-024-02426-7
IF: 11
2024-02-09
Molecular Psychiatry
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.
biochemistry & molecular biology,neurosciences,psychiatry