How to Cite
Abstract
The escalating rate of adolescent suicide represents a profound global public health crisis that demands immediate and innovative intervention strategies. As traditional clinical assessments often fail to capture the dynamic and multifaceted nature of suicidal ideation, computational methods have emerged as a vital supplementary tool. This paper proposes a comprehensive, multimodal machine learning framework designed to identify suicide risk in adolescents by integrating clinical health records, acoustic speech features, and linguistic markers from digital platforms. By synthesizing diverse data modalities and applying semi-supervised learning techniques, this approach aims to overcome the limitations of isolated datasets and provide a more robust, real-time assessment of patient vulnerability. Ultimately, the integration of advanced predictive models into psychiatric care holds the potential to facilitate timely interventions, reduce the cognitive burden on crisis responders, and save young lives
References
Bhat, Harish S., & Goldman-Mellor, Sidra J. (2017). Predicting Adolescent Suicide Attempts with Neural Networks. https://arxiv.org/pdf/1711.10057v2 https://arxiv.org/pdf/1711.10057v2
Marie, Ambre, Maoudj, Ilias, Dardenne, Guillaume, & Quellec, Gwenolé (2025). Suicide Risk Assessment Using Multimodal Speech Features: A Study on the SW1 Challenge Dataset. https://arxiv.org/pdf/2505.13069v1 https://arxiv.org/pdf/2505.13069v1
Lovitt, Max, Ma, Haotian, Wang, Song, & Peng, Yifan (2024). Suicide Risk Assessment on Social Media with Semi-Supervised Learning. https://arxiv.org/pdf/2411.12767v2 https://arxiv.org/pdf/2411.12767v2
Wei, Zhiyuan, & Mukherjee, Sayanti (2020). Health-behaviors associated with the growing risk of adolescent suicide attempts: A data-driven cross-sectional study. https://arxiv.org/pdf/2009.03966v1 https://arxiv.org/pdf/2009.03966v1
Amiriparian, Shahin, Gerczuk, Maurice, Lutz, Justina, Strube, Wolfgang, Papazova, Irina, Hasan, Alkomiet, Kathan, Alexander, & Schuller, Björn W. (2024). Non-Invasive Suicide Risk Prediction Through Speech Analysis. https://arxiv.org/pdf/2404.12132v3 https://arxiv.org/pdf/2404.12132v3
Gerczuk, Maurice, Amiriparian, Shahin, Lutz, Justina, Strube, Wolfgang, Papazova, Irina, Hasan, Alkomiet, & Schuller, Björn W. (2024). Exploring Gender-Specific Speech Patterns in Automatic Suicide Risk Assessment. https://arxiv.org/pdf/2407.11012v1 https://arxiv.org/pdf/2407.11012v1
Yang, Chenghao, Zhang, Yudong, & Muresan, Smaranda (2021). Weakly-Supervised Methods for Suicide Risk Assessment: Role of Related Domains. https://arxiv.org/pdf/2106.02792v2 https://arxiv.org/pdf/2106.02792v2
Obadinma, Stephen, Lachana, Alia, Norman, Maia, Rankin, Jocelyn, Yu, Joanna, Zhu, Xiaodan, Mastropaolo, Darren, Pandya, Deval, Sultan, Roxana, & Dolatabadi, Elham (2024). FAIIR: Building Toward A Conversational AI Agent Assistant for Youth Mental Health Service Provision. https://arxiv.org/pdf/2405.18553v4 https://arxiv.org/pdf/2405.18553v4
Hamilton, Matthew P, Gao, Caroline X, Wiesner, Glen, Filia, Kate M, Menssink, Jana M, Plencnerova, Petra, Baker, David G, McGorry, Patrick D, Parker, Alexandra, Karnon, Jonathan, Cotton, Sue M, & Mihalopoulos, Cathrine (2023). A prototype software framework for transferable computational health economic models and its early application in youth mental health. PharmacoEconomics (2024). https://doi.org/10.1007/s40273-024-01378-8 https://doi.org/10.1007/s40273-024-01378-8
Poulsen, Adam, Hickie, Ian B., Gorban, Carla, Haan, Zsofi de, Capon, William, Eyeson-Annan, Ebenezer, Radwan, Jalal, Scott, Elizabeth M., Iorfino, Frank, & LaMonica, Haley M. (2026). Young people's perceptions and recommendations for conversational generative artificial intelligence in youth mental health. https://arxiv.org/pdf/2604.13381v1 https://arxiv.org/pdf/2604.13381v1
This work is licensed under a Creative Commons Attribution 4.0 International License.
