EUVIMEDEUVIMED entdecken
Multizentrische StudieOpen Access

An Efficient Contrastive Deep Learning Model for Identifying Schizophrenia-Specific Neuroanatomical Variations

Lin Du, Biying Peng, Yuqing Sun, Chaoyue Ding, Xiaohan Tian, Wenkun Lei, Qi Wang, Xinghui Zhao, Yingjie Peng, Jing Lou, Junxing Xian, Xinyi Dong und 4 weitere

Schizophrenia Bulletin · 07.03.2026 · 52 (2)

DOI: 10.1093/schbul/sbaf261 ↗

Kurz gefasst

Multizentrische Studie in der Zeitschrift Schizophrenia Bulletin, online erschienen am 21.03.2026. Verfasst von Lin Du und 15 weiteren Autor:innen. Beteiligte Einrichtung: State Key Laboratory of Cognitive Neuroscience and Learning, Beijing Normal University, Beijing 100875, China. Hauptthemen laut MEDLINE: Schizophrenia, Autoencoder, Convolutional Neural Networks, Gray Matter. Die Arbeit nennt 90 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar. Gefördert von: Natural Science Foundation of China, STI 2030-Major Projects, Beijing Nova Program.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background and Hypothesis Schizophrenia (SZ) is a debilitating mental disorder characterized by heterogeneous clinical manifestations and widespread brain structural abnormalities. However, the relationship between individual neuroanatomical abnormalities and clinical symptoms remains inconsistent. We hypothesize that isolating SZ-specific neuroanatomical variations could deepen our understanding of its pathophysiology and yield more reliable biomarkers for its clinical heterogeneity. Study Design To investigate this, we developed DECODE-SZ (Dual Encoder Contrastive Decoding for Schizophrenia), a novel model combining contrastive learning, 3D convolutional neural networks, and variational autoencoders (VAE) to isolate SZ-specific neuroanatomical features. We applied this model to structural MRI data from 641 patients diagnosed with SZ and 609 healthy controls across 8 independent sites in China, employing a leave-one-site-out cross-validation approach to ensure robust and unbiased results. Our analysis focused on examining the relationship between SZ-specific gray matter alterations and clinical symptoms (measured by PANSS scores), while also considering non-clinical variables such as age, sex, and education. Study Results The DECODE-SZ model successfully extracted SZ-specific gray matter features, revealing that these features, rather than common variations with controls, were more strongly associated with PANSS scores. Consistent brain regions exhibiting these alterations were identified across multiple sites, supporting the reliability of the findings. A control experiment using a traditional VAE model demonstrated the superior performance of DECODE-SZ in isolating meaningful SZ-specific neuroanatomical variations. Conclusions Our findings highlight the potential of SZ-specific neuroanatomical alterations as key biomarkers for clinical outcomes in SZ. DECODE-SZ offers a promising tool for advancing the understanding of SZ and may inform future diagnostic and therapeutic strategies.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

SchizophreniaAutoencoderConvolutional Neural NetworksGray MatterImage Interpretation, Computer-AssistedCase-Control StudiesSoftware ValidationHumansMaleFemaleYoung AdultAdultAnatomic Variation

Stichwörter der Autor:innen

contrastive learningdeep learningneuroanatomyschizophrenia

Autor:innen und Institutionen

  • State Key Laboratory of Cognitive Neuroscience and Learning, Beijing Normal University, Beijing 100875, China.
  • IDG/McGovern Institute for Brain Research, Beijing Normal University, Beijing 100875, China.
  • Brainnetome Center and National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences, Beijing 100190, China.
  • State Key Lab of Brain and Cognitive Science, Institute of Biophysics, Chinese Academy of Sciences, Beijing 100101, China.
  • Chinese Institute for Brain Research, Beijing 102206, China.

ORCID-Profile: Bing Liu

Förderung

  • Natural Science Foundation of China
  • STI 2030-Major Projects
  • Beijing Nova Program
  • STI 2030—Major Projects

Zitieren

APA

Du, L., Peng, B., Sun, Y., Ding, C., Tian, X., Lei, W., Wang, Q., Zhao, X., Peng, Y., Lou, J., Xian, J., Dong, X., Hong, Y., Yang, R., Wang, M., & Liu, B. (2026). An Efficient Contrastive Deep Learning Model for Identifying Schizophrenia-Specific Neuroanatomical Variations. Schizophrenia Bulletin, 52(2). https://doi.org/10.1093/schbul/sbaf261

Vancouver

Du L, Peng B, Sun Y, Ding C, Tian X, Lei W, et al. An Efficient Contrastive Deep Learning Model for Identifying Schizophrenia-Specific Neuroanatomical Variations. Schizophrenia Bulletin. 2026;52(2). doi:10.1093/schbul/sbaf261

BibTeX
@article{Du2026,
  author = {Du, Lin and Peng, Biying and Sun, Yuqing and Ding, Chaoyue and Tian, Xiaohan and Lei, Wenkun and Wang, Qi and Zhao, Xinghui and Peng, Yingjie and Lou, Jing and Xian, Junxing and Dong, Xinyi and Hong, Yuxuan and Yang, Ruoxin and Wang, Meng and Liu, Bing},
  title = {An Efficient Contrastive Deep Learning Model for Identifying Schizophrenia-Specific Neuroanatomical Variations},
  journal = {Schizophrenia Bulletin},
  year = {2026},
  volume = {52},
  number = {2},
  doi = {10.1093/schbul/sbaf261}
}
Kostenlos starten →