{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,19]],"date-time":"2026-06-19T05:54:51Z","timestamp":1781848491451,"version":"3.54.5"},"reference-count":21,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,6,19]],"date-time":"2026-06-19T00:00:00Z","timestamp":1781827200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Psychiatry"],"abstract":"<jats:sec>\n                    <jats:title>Objective<\/jats:title>\n                    <jats:p>To develop a multimodal deep learning\u2013based classification model for adolescent bipolar depression (ABD) with verbal auditory hallucinations (AVHs).<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>A retrospective analysis was conducted on 47 untreated ABD patients within 30 days, between January 2024 and August 2025. Comprehensive clinical data were collected, including sex, age, age at onset, years of education, and the presence of suicidal or self-harming behaviors. Based on the PANSS P3 score and the presence of AVHs, patients were divided into a hallucination group (P3 score &amp;gt; 3, n = 24) and a non-hallucination group (P3 score \u2264 3, n = 23). All participants underwent 1H-MRS scanning of the ventromedial prefrontal cortex (vmPFC). A multimodal deep learning model was constructed using MRS-derived features in combination with clinical parameters.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>The model achieved an optimal classification accuracy of 71.43% on the fixed test set, as obtained by the second-fold model. This best-performing model demonstrated balanced and stable classification performance for both positive and negative samples, with precision, recall, and F1-score all reaching 0.75.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>This study proposes a novel multimodal Transformer-based framework and evaluates its effectiveness in classifying ABD patients experiencing depressive episodes with AVHs. The results suggest that the advanced model architecture, incorporating mechanisms such as bidirectional cross-attention and an Ensemble of Experts classifier, can effectively integrate heterogeneous data and achieve a test accuracy of 71.43% on a small dataset, indicating preliminary technical feasibility.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/fpsyt.2026.1844684","type":"journal-article","created":{"date-parts":[[2026,6,19]],"date-time":"2026-06-19T05:47:33Z","timestamp":1781848053000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Multi-modal deep learning model for bipolar depression adolescents with verbal auditory hallucinations"],"prefix":"10.3389","volume":"17","author":[{"given":"Qinnaer","family":"Bolatijiang","sequence":"first","affiliation":[{"name":"Graduate School, Xinjiang Medical University","place":["\u00dcr\u00fcmqi, Xinjiang, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Shaohong","family":"Zou","sequence":"additional","affiliation":[{"name":"Department of Clinical Psychology, People\u2019s Hospital of Xinjiang Uygur Autonomous Region","place":["Urumqi, Xinjiang, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Cheng","family":"Zhang","sequence":"additional","affiliation":[{"name":"Department of Clinical Psychology, People\u2019s Hospital of Xinjiang Uygur Autonomous Region","place":["Urumqi, Xinjiang, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Chengji","family":"Wang","sequence":"additional","affiliation":[{"name":"Department of Clinical Psychology, People\u2019s Hospital of Xinjiang Uygur Autonomous Region","place":["Urumqi, Xinjiang, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jianliang","family":"Zhang","sequence":"additional","affiliation":[{"name":"Department of Radiology, People\u2019s Hospital of Xinjiang Uygur Autonomous Region","place":["Urumqi, Xinjiang, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,6,19]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","DOI":"10.1016\/S2215-0366(21)00395-3","article-title":"Global, regional, and national burden of 12 mental disorders in 204 countries and territories 1990\u20132019: A systematic analysis for the global burden of disease study 2019","volume":"9","author":"A","year":"2022","journal-title":"Lancet Psychiatry"},{"key":"B2","doi-asserted-by":"publisher","DOI":"10.1111\/bdi.13034","article-title":"Prospectively ascertained mania and hypomania among young adults with child\u2010 and adolescent\u2010onset bipolar disorder","volume":"23","author":"Hafeman","year":"2021","journal-title":"Bipolar Disord"},{"key":"B3","doi-asserted-by":"publisher","DOI":"10.1097\/HRP.0000000000000309","article-title":"Martino D J, Valerio M P. Bipolar depression: A historical perspective of the current concept, with a focus on future research","volume":"29","author":"Martino","year":"2021","journal-title":"Harvard Rev Psychiatry"},{"key":"B4","doi-asserted-by":"publisher","DOI":"10.1111\/Bdi.13367","article-title":"The diagnosis of bipolar disorder in children and adolescents past, present and future","volume":"25","author":"Malhi","year":"2023","journal-title":"Bipolar Disord"},{"key":"B5","doi-asserted-by":"publisher","DOI":"10.2174\/1570159X14666160606210811","article-title":"Depression and mania in bipolar disorder","volume":"15","author":"Tondo","year":"2017","journal-title":"Curr Neuropharmacol"},{"key":"B6","doi-asserted-by":"publisher","DOI":"10.1590\/1516-4446-2019-0455","article-title":"Psychotic and affective symptoms of early-onset bipolar disorder: An observational study of patients in first manic episode","volume":"42","author":"Fu-I","year":"2020","journal-title":"Braz J Psychiatry"},{"key":"B7","doi-asserted-by":"publisher","DOI":"10.5498\/wjp.v12.i9.1204","article-title":"Psychotic symptoms in bipolar disorder and their impact on the illness: A systematic review","volume":"12","author":"Chakrabarti","year":"2022","journal-title":"WJP"},{"key":"B8","doi-asserted-by":"publisher","DOI":"10.1186\/s40345-025-00369-8","article-title":"Characteristics of people with bipolar disorder I with and without auditory verbal hallucinations","volume":"13","author":"Javier","year":"2025","journal-title":"Int J Bipolar Disord"},{"key":"B9","doi-asserted-by":"publisher","DOI":"10.1192\/bjo.2024.784","article-title":"Associations between early experiences of thought interference and auditory-verbal hallucinations with first-rank symptoms and suicidality in adulthood","volume":"10","author":"Gofton","year":"2024","journal-title":"BJPsych Open"},{"key":"B10","doi-asserted-by":"publisher","DOI":"10.1002\/mp.16831","article-title":"Denoising magnetic resonance spectroscopy (MRS) data using stacked autoencoder for improving signal\u2010to\u2010noise ratio and speed of MRS","volume":"50","author":"Wang","year":"2023","journal-title":"Med Phys"},{"key":"B11","doi-asserted-by":"publisher","DOI":"10.1016\/j.rcl.2021.06.004","article-title":"Basic artificial intelligence techniques","volume":"59","author":"Erickson","year":"2021","journal-title":"Radiologic Clinics North America"},{"key":"B12","doi-asserted-by":"publisher","DOI":"10.1016\/j.crmeth.2021.100003","article-title":"Deep learning neural network tools for proteomics","volume":"1","author":"Meyer","year":"2021","journal-title":"Cell Rep Methods"},{"key":"B13","doi-asserted-by":"publisher","DOI":"10.1109\/EMBC46164.2021.9630693","article-title":"Multi-modal deep learning of functional and structural neuroimaging and genomic data to predict mental illness","volume":"2021","author":"Rahaman","year":"2021","journal-title":"Annu Int Conf IEEE Eng Med Biol Soc IEEE Eng Med Biol Soc Annu Int Conf"},{"key":"B14","doi-asserted-by":"publisher","first-page":"14497","DOI":"10.1038\/s41598-025-99167-6","article-title":"Trade-offs between machine learning and deep learning for mental illness detection on social media","volume":"15","author":"Ding","year":"2025","journal-title":"Sci Rep"},{"key":"B15","doi-asserted-by":"publisher","first-page":"e85540","DOI":"10.7759\/cureus.85540","article-title":"Mental Illness Diagnostic Criteria Can Be Simplified With Higher Symptom Prevalence and Correlations: A Simulation Study","volume":"17","author":"Chao","year":"2025","journal-title":"Cureus"},{"key":"B16","doi-asserted-by":"publisher","first-page":"1101","DOI":"10.1001\/jamapsychiatry.2024.2047","article-title":"Unveiling the structure in mental disorder presentations","volume":"81","author":"Spiller","year":"2024","journal-title":"JAMA Psychiatry"},{"key":"B17","doi-asserted-by":"publisher","first-page":"146","DOI":"10.1001\/jamapsychiatry.2022.4076","article-title":"Deep learning for cross-diagnostic prediction of mental disorder diagnosis and prognosis using Danish nationwide register and genetic data","volume":"80","author":"Alles\u00f8","year":"2023","journal-title":"JAMA Psychiatry"},{"key":"B18","doi-asserted-by":"publisher","DOI":"10.1016\/j.compmedimag.2024.102368","article-title":"Multi modality fusion transformer with spatio-temporal feature aggregation module for psychiatric disorder diagnosis","volume":"114","author":"Wang","year":"2024","journal-title":"Comput Med Imaging Graphics"},{"key":"B19","doi-asserted-by":"publisher","DOI":"10.1109\/JBHI.2021.3127688","article-title":"Cross-modal prostate cancer segmentation via self-attention distillation","volume":"26","author":"Zhang","year":"2022","journal-title":"IEEE J BioMed Health Inf"},{"key":"B20","doi-asserted-by":"publisher","first-page":"25704","DOI":"10.1038\/s41598-024-76234-y","article-title":"Dual-attention transformer-based hybrid network for multi-modal medical image segmentation","volume":"14","author":"Zhang","year":"2024","journal-title":"Sci Rep"},{"key":"B21","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2024.3441012","article-title":"RemixFormer++: A multi-modal transformer model for precision skin tumor differential diagnosis with memory-efficient attention","volume":"44","author":"Xu","year":"2025","journal-title":"IEEE Trans Med Imaging"}],"container-title":["Frontiers in Psychiatry"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fpsyt.2026.1844684\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,19]],"date-time":"2026-06-19T05:47:34Z","timestamp":1781848054000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fpsyt.2026.1844684\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,6,19]]},"references-count":21,"alternative-id":["10.3389\/fpsyt.2026.1844684"],"URL":"https:\/\/doi.org\/10.3389\/fpsyt.2026.1844684","relation":{},"ISSN":["1664-0640"],"issn-type":[{"value":"1664-0640","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,6,19]]},"article-number":"1844684"}}