This study addresses the challenges of data noise and model interpretability in depression diagnosis by proposing an intelligent diagnostic framework based on real-world medical scenarios. Utilizing a labeled dataset of 11,188 Chinese online consultation records, we developed a dual-channel architecture integrating BERT/RoBERTa pre-trained models with the SHAP interpretability framework for depression severity classification. Experimental results demonstrated that the BERT model achieved 92% overall accuracy, with 93% accuracy specifically for severe depression detection. SHAP analysis revealed the model’s focus on clinically relevant features like suicidal tendencies and low mood, showing significant alignment with DSM-5 diagnostic criteria. The study confirms pre-trained models’ capability in extracting pathological semantics from medical texts, while the “prediction-interpretation” framework provides a technical prototype for overcoming clinical application barriers of black-box models.
References
[1]
Institute of Health Metrics and Evaluation (2023) Global Health Data Exchange (GHDx). https://vizhub.healthdata.org/gbd-results/
[2]
COVID-19 Mental Disorders Collaborators (2021) Global Prevalence and Burden of Depressive and Anxiety Disorders in 204 Countries and Territories in 2020 Due to the COVID-19 Pandemic. The Lancet, 398, 1700-1712.
[3]
Byeon, H. (2023) Advances in Machine Learning and Explainable Artificial Intelligence for Depression Prediction. International Journal of Advanced Computer Science and Applications, 14, 520-526. https://doi.org/10.14569/ijacsa.2023.0140656
[4]
Eichstaedt, J.C., Smith, R.J., Merchant, R.M., Ungar, L.H., Crutchley, P., Preoţiuc-Pietro, D., et al. (2018) Facebook Language Predicts Depression in Medical Records. Proceedings of the National Academy of Sciences, 115, 11203-11208. https://doi.org/10.1073/pnas.1802331115
[5]
Shrestha, A., Serra, E. and Spezzano, F. (2020) Multi-Modal Social and Psycho-Linguistic Embedding via Recurrent Neural Networks to Identify Depressed Users in Online Forums. Network Modeling Analysis in Health Informatics and Bioinformatics, 9, Article No. 22. https://doi.org/10.1007/s13721-020-0226-0
[6]
Tadesse, M.M., Lin, H., Xu, B. and Yang, L. (2019) Detection of Suicide Ideation in Social Media Forums Using Deep Learning. Algorithms, 13, Article 7. https://doi.org/10.3390/a13010007
[7]
Parapar, J., Martín-Rodilla, P., Losada, D.E. and Crestani, F. (2021) Overview of eRisk 2021: Early Risk Prediction on the Internet. In: Candan, K.S., Ionescu, B., Goeuriot, L., et al., Eds., Experimental IR Meets Multilinguality, Multimodality, and Interaction. CLEF 2021. Lecture Notes in Computer Science, Springer, 324-344. https://doi.org/10.1007/978-3-030-85251-1_22
[8]
Yang, T.T., Li, F., Ji, D.H., Liang, X.H., Xie, T., Tian, S.W., et al. (2021) Fine-Grained Depression Analysis Based on Chinese Micro-Blog Reviews. Information Processing & Management, 58, Article 102681. https://doi.org/10.1016/j.ipm.2021.102681
[9]
Burdisso, S.G., Errecalde, M. and Montes-y-Gómez, M. (2021) Using Text Classification to Estimate the Depression Level of Reddit Users. Journal of Computer Science and Technology, 21, e1. https://doi.org/10.24215/16666038.21.e1
[10]
Abed-Esfahani, P., Howard, D., Maslej, M., et al. (2019) Transfer Learning for De-Pression: Early Detection and Severity Prediction from Social Media Postings. CLEF (Working Notes), 1-6. https://ceur-ws.org/Vol-2380/paper_102.pdf
[11]
Mi, J., Li, A. and Zhou, L. (2020) Review Study of Interpretation Methods for Future Interpretable Machine Learning. IEEE Access, 8, 191969-191985. https://doi.org/10.1109/access.2020.3032756
[12]
Devlin, J., Chang, M.W., Lee, K., et al. (2019) BERT: Pre-Training of Deep Bidirectional Transformers for Language Understanding. Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Lin-Guistics: Human Language Technologies, Minnesota, 2-7 June 2019, 4171-4186.
[13]
Liu, Y., Ott, M., Goyal, N., et al. (2019) RoBERTa: A Robustly Optimized BERT Pretraining Approach.
[14]
Lundberg, S.M., Erion, G., Chen, H., DeGrave, A., Prutkin, J.M., Nair, B., et al. (2020) From Local Explanations to Global Understanding with Explainable AI for Trees. Nature Machine Intelligence, 2, 56-67. https://doi.org/10.1038/s42256-019-0138-9
[15]
Hueniken, K., Somé, N.H., Abdelhack, M., Taylor, G., Elton Marshall, T., Wickens, C.M., et al. (2021) Machine Learning-Based Predictive Modeling of Anxiety and Depressive Symptoms during 8 Months of the COVID-19 Global Pandemic: Repeated Cross-Sectional Survey Study. JMIR Mental Health, 8, e32876. https://doi.org/10.2196/32876
[16]
Nemesure, M.D., Heinz, M.V., Huang, R. and Jacobson, N.C. (2021) Predictive Modeling of Depression and Anxiety Using Electronic Health Records and a Novel Machine Learning Approach with Artificial Intelligence. Scientific Reports, 11, Article No. 1980. https://doi.org/10.1038/s41598-021-81368-4
[17]
Nguyen, H.V. and Byeon, H. (2022) Explainable Deep-Learning-Based Depression Modeling of Elderly Community after COVID-19 Pandemic. Mathematics, 10, Article 4408. https://doi.org/10.3390/math10234408
[18]
Nordin, N., Zainol, Z., Mohd Noor, M.H. and Chan, L.F. (2023) An Explainable Predictive Model for Suicide Attempt Risk Using an Ensemble Learning and Shapley Additive Explanations (SHAP) Approach. Asian Journal of Psychiatry, 79, Article 103316. https://doi.org/10.1016/j.ajp.2022.103316
[19]
Do, H.P., Baker, P.R.A., Van Vo, T., Murray, A., Murray, L., Valdebenito, S., et al. (2021) Intergenerational Effects of Violence on Women’s Perinatal Wellbeing and Infant Health Outcomes: Evidence from a Birth Cohort Study in Central Vietnam. BMC Pregnancy and Childbirth, 21, Article No. 648. https://doi.org/10.1186/s12884-021-04097-6
[20]
Yates, A., Cohan, A. and Goharian, N. (2017) Depression and Self-Harm Risk Assessment in Online Forums. In: Martha, P. and Rebecca, H. and Sebastian, R. Eds., Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing, Association for Computational Linguistics, 2968-2978. https://doi.org/10.18653/v1/d17-1322
[21]
Bucur, A., Cosma, A. and Dinu, P.L. (2021) Early Risk Detection of Pathological Gambling, Self-Harm and Depression Using BERT. http://dx.doi.org/10.13140/RG.2.2.25060.50567
[22]
Sun, C., Qiu, X.P., Xu, Y. and Huang, X.J. (2019) How to Fine-Tune BERT for Text Classification? In: Lecture Notes in Computer Science, Springer International Publishing, 194-206. https://doi.org/10.1007/978-3-030-32381-3_16
[23]
Shapley, L.S. (1953) 17. A Value for N-Person Games. In: Kuhn, H.W. and Tucker, A.W., Eds., Contributions to the Theory of Games (AM-28), Volume II, Princeton University Press, 307-318. https://doi.org/10.1515/9781400881970-018
[24]
Adadi, A. and Berrada, M. (2018) Peeking Inside the Black-Box: A Survey on Explainable Artificial Intelligence (XAI). IEEE Access, 6, 52138-52160. https://doi.org/10.1109/access.2018.2870052
[25]
Azpiazu, C., Bosch, J., Bortolotti, L., Medrzycki, P., Teper, D., Molowny-Horas, R., et al. (2021) Toxicity of the Insecticide Sulfoxaflor Alone and in Combination with the Fungicide Fluxapyroxad in Three Bee Species. Scientific Reports, 11, Article No. 6821. https://doi.org/10.1038/s41598-021-86036-1
