Voice-based machine learning for rapid screening of bipolar disorder and major depressive disorder in children and adolescents: a robust and low-complexity diagnostic model - Takeaways - MDSpire

Voice-based machine learning for rapid screening of bipolar disorder and major depressive disorder in children and adolescents: a robust and low-complexity diagnostic model

  • By

  • Zhaojun Li

  • Xushan Li

  • Zhuo Wang

  • Jie Gao

  • Jie Luo

  • Fan He

  • Yi Zheng

  • Lihui Feng

  • Jihua Lu

  • January 30, 2026

  • 0 min

Share

  • 1

    Bipolar disorder (BD) and major depressive disorder (MDD) are prevalent psychiatric disorders affecting millions globally.

  • 2

    Misdiagnosis of BD as MDD occurs in 40% to 67% of cases, complicating treatment and prolonging the correct diagnosis.

  • 3

    Voice features provide a promising avenue for diagnosing BD and MDD due to their non-intrusive nature and sensitivity to physiological changes.

  • 4

    The study aims to distinguish between MDD, BD, and healthy controls in youth using voice features and advanced classification techniques.

  • 5

    A total of 150 subjects were recruited for the study, with diagnoses confirmed by expert consensus based on DSM-5 criteria.

Original Source(s)

Voice-based machine learning for rapid screening of bipolar disorder and major depressive disorder in children and adolescents: a robust and low-complexity diagnostic model

  • by Zhaojun Li, Xushan Li, Zhuo Wang, Jie Gao, Jie Luo, Fan He, Yi Zheng, Lihui Feng, Jihua Lu

  • January 30, 2026

Related Content

Premenstrual Disorders Linked to Broad Psychiatric Risk

Swedish study finds two-way associations between premenstrual disorders and psychiatric conditions, with strongest links involving depression, anxiety, attention-deficit/hyperactivity disorder, bipolar disorder, and personality disorders.

Development and validation of a machine learning–based risk prediction model for non-suicidal self-injury in adolescents

Aggression and emotional distress in adolescents: a cross-sectional chain mediation model of internet addiction and somatization