Skip to content
DraconiticResearch
Search Sources API
210,089 studies31 sourcesUpdated 15 min ago
Peer-reviewedCC BY 4.0Psychiatry and Clinical Neurosciences

Differentiation between bipolar versus unipolar depression by AI‐based multimodal behavioral analysis

  • Hidenori Yamasue1,2
  • Sanskar Rahul Nanegaonkar3
  • Yumi Kiyama4
  • Mamindla Vishwa Raghava Reddy3
  • Kotaro Ando3
  • Shozo Kamiya3
  • Jun Inoue1,5
  • Asuka Sakurai1
  • Tahichi Nakano6
  • Amane Sato6
  • Yosuke Kameno1
  • Tomoyasu Wakuda1
2 more authors
  • Keiichi Odagiri4
  • Tadafumi Kato7
7 affiliations
  1. 1Department of Psychiatry Hamamatsu University School of Medicine Hamamatsu Japan
  2. 2United Graduate School of Child Development Osaka University, Kanazawa University, Hamamatsu University School of Medicine, Chiba University and University of Fukui Suita Japan
  3. 3I'mbesideyou Tokyo Japan
  4. 4Center for Clinical Research Hamamatsu University Hospital Hamamatsu Japan
  5. 5Department of Child and Adolescent Psychiatry Hamamatsu University School of Medicine Hamamatsu Japan
  6. 6Hamamatsu University School of Medicine Hamamatsu Japan
  7. 7Department of Psychiatry and Behavioral Science Juntendo University Graduate School of Medicine Tokyo Japan
Read full textPDFDOICite

Abstract

Background: Given their differing pathophysiology and treatment approaches, the primary aim of the current study was to develop a methodology accurately distinguishing bipolar from unipolar depression.

Methods: Adults with bipolar disorder or major depressive disorder were recruited nationwide in Japan and assessed via recorded interviews. During the interview, diagnoses were established using the Structured Clinical Interview for DSM‐5‐Research Version. Depressive and manic symptom severity was assessed using the Hamilton Depression Rating Scale (HDRS) and Young Mania Rating Scale (YMRS), respectively. Artificial intelligence (AI)‐based integrative, quantitative analyses of behavioral characteristics, including facial expressions, eye gaze, and voice prosody, were conducted on the recorded video.

Results: Between August 2022 and March 2024, 379 participants completed interviews, of whom 366 were analyzed (120 bipolar I, 93 bipolar II, 122 major depressive, and 74 persistent depressive disorder). AI‐based diagnostic differentiation between bipolar and unipolar depression using four‐fold, three‐times repeated cross‐validation achieved an overall accuracy of 78%, with an area under the curve of 0.76. Key features contributing to bipolar diagnosis included greater temporal instabilities in neutral facial expression, eye gaze, sad facial expression, and root mean square amplitude of speech. HDRS could be predicted in unipolar and bipolar depression, with influential predictors of depression including sleep inefficiency, autistic traits, variability in sad facial expression, and comorbid anxiety disorder. YMRS could be predicted with major predictors including comorbid anxiety disorder, sleep inefficiency, and variabilities in sad facial expression and eye movement.

Conclusions: Integrating multimodal behavioral indicators with AI‐based modeling enables clinically meaningful differentiation between bipolar and unipolar depression.

Related

  1. Multimodal predictors of response to intravenous ketamine in treatment-resistant depression: toward personalized treatment in major depressive and bipolar disorders
    Magda Malewska-Kasprzak, Filip Rybakowski, Monika Dmitrzak-Weglarz
    CNS SpectrumsPeer-reviewedCC BY 4.0Full textPharmacology & toxicologyPMC
  2. Device-derived sleep metrics and nocturnal heart rate variability for discriminating bipolar from unipolar depressive episodes in adolescents: a retrospective cross-sectional study.
    Hao Yu, Xueli Zhang, Jiaqi Xiang +8
    BMC psychiatryPeer-reviewedCC BY-NC-NDFull text1 citedPsychiatry & mental healthEurope PMCPMC
  3. Precision stratification of risk for suicidal behavior in people with bipolar depression
    Nina de Lacy, Wai-Yin Lam, Monika Baker +5
    Scientific ReportsPeer-reviewedCC BY-NC-ND 4.0Full textPsychiatry & mental healthCrossref
  4. Inpatient Lurasidone Treatment for Adolescent Bipolar Depression: A Retrospective Study
    Fevzi Tuna Ocakoğlu, Yiğit Özaydın, Merve Mergen +2
    Psychiatry and clinical psychopharmacologyPeer-reviewedCC BY-NCFull textPsychiatry & mental healthEurope PMCPMC
  5. Association Between Eating Disorders and Subsequently Recorded Depression and Bipolar Disorder: A Retrospective Cohort Study
    Vanessza Hadar, Battamir Ulambayar, Attila Csaba Nagy +1
    NutrientsPeer-reviewedCC BY 4.0Full textPsychiatry & mental healthCrossrefEurope PMC

Cite

BibTeXRIS
BibTeX
@article{yamasue2026differentiation,
  title = {{Differentiation between bipolar versus unipolar depression by AI‐based multimodal behavioral analysis}},
  author = {Hidenori Yamasue and Sanskar Rahul Nanegaonkar and Yumi Kiyama and Mamindla Vishwa Raghava Reddy and Kotaro Ando and Shozo Kamiya and Jun Inoue and Asuka Sakurai and Tahichi Nakano and Amane Sato and Yosuke Kameno and Tomoyasu Wakuda and Keiichi Odagiri and Tadafumi Kato},
  journal = {Psychiatry and Clinical Neurosciences},
  year = {2026},
  month = sep,
  doi = {10.1111/pcn.70124},
  url = {https://doi.org/10.1111/pcn.70124},
  note = {CC BY 4.0},
}
RIS
TY  - JOUR
TI  - Differentiation between bipolar versus unipolar depression by AI‐based multimodal behavioral analysis
AU  - Hidenori Yamasue
AU  - Sanskar Rahul Nanegaonkar
AU  - Yumi Kiyama
AU  - Mamindla Vishwa Raghava Reddy
AU  - Kotaro Ando
AU  - Shozo Kamiya
AU  - Jun Inoue
AU  - Asuka Sakurai
AU  - Tahichi Nakano
AU  - Amane Sato
AU  - Yosuke Kameno
AU  - Tomoyasu Wakuda
AU  - Keiichi Odagiri
AU  - Tadafumi Kato
PY  - 2026
DA  - 2026/09/11
JO  - Psychiatry and Clinical Neurosciences
DO  - 10.1111/pcn.70124
UR  - https://doi.org/10.1111/pcn.70124
AB  - Background: Given their differing pathophysiology and treatment approaches, the primary aim of the current study was to develop a methodology accurately distinguishing bipolar from unipolar depression.  Methods: Adults with bipolar disorder or major depressive disorder were recruited nationwide in Japan and assessed via recorded interviews. During the interview, diagnoses were established using the Structured Clinical Interview for DSM‐5‐Research Version. Depressive and manic symptom severity was assessed using the Hamilton Depression Rating Scale (HDRS) and Young Mania Rating Scale (YMRS), respectively. Artificial intelligence (AI)‐based integrative, quantitative analyses of behavioral characteristics, including facial expressions, eye gaze, and voice prosody, were conducted on the recorded video.  Results: Between August 2022 and March 2024, 379 participants completed interviews, of whom 366 were analyzed (120 bipolar I, 93 bipolar II, 122 major depressive, and 74 persistent depressive disorder). AI‐based diagnostic differentiation between bipolar and unipolar depression using four‐fold, three‐times repeated cross‐validation achieved an overall accuracy of 78%, with an area under the curve of 0.76. Key features contributing to bipolar diagnosis included greater temporal instabilities in neutral facial expression, eye gaze, sad facial expression, and root mean square amplitude of speech. HDRS could be predicted in unipolar and bipolar depression, with influential predictors of depression including sleep inefficiency, autistic traits, variability in sad facial expression, and comorbid anxiety disorder. YMRS could be predicted with major predictors including comorbid anxiety disorder, sleep inefficiency, and variabilities in sad facial expression and eye movement.  Conclusions: Integrating multimodal behavioral indicators with AI‐based modeling enables clinically meaningful differentiation between bipolar and unipolar depression.
SN  - 1323-1316
SN  - 1440-1819
ER  -