pith. sign in

arxiv: 2508.12022 · v2 · submitted 2025-08-16 · 💻 cs.AI

AI Models for Depressive Disorder Detection and Diagnosis: A Review

Dorsa Macky Aleagha , Payam Zohari , Mostafa Haghir Chehreghani This is my paper

Pith reviewed 2026-05-18 23:02 UTC · model grok-4.3

classification 💻 cs.AI
keywords depression detectionAI modelsmajor depressive disordergraph neural networkslarge language modelsmultimodal fusionsystematic reviewcomputational psychiatry
0
0 comments X

The pith

A review of 55 studies introduces a hierarchical taxonomy for AI-based depression detection organized by clinical task, data modality, and model class.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

This paper surveys current artificial intelligence techniques aimed at detecting and diagnosing major depressive disorder, which currently relies on subjective assessments. It reviews 55 key studies and proposes a novel taxonomy that categorizes approaches according to whether they focus on diagnosis or prediction, the type of data used such as text or brain images, and the kind of computational model employed. The analysis identifies growing use of graph neural networks for brain data, large language models for text and conversation, and efforts to combine multiple data types while addressing fairness and explainability. By also covering public datasets and evaluation metrics, the survey aims to guide researchers toward more objective and scalable tools for mental health. If accurate, this structure helps clarify where the field stands and what gaps remain for future work in computational psychiatry.

Core claim

We present a comprehensive survey of state-of-the-art AI methods for depression detection and diagnosis, based on a systematic review of 55 key studies. We introduce a novel hierarchical taxonomy that structures the field by primary clinical task (diagnosis vs. prediction), data modality (text, speech, neuroimaging, multimodal), and computational model class (e.g., graph neural networks, large language models, hybrid approaches). Our in-depth analysis reveals three major trends: the predominance of graph neural networks for modeling brain connectivity, the rise of large language models for linguistic and conversational data, and an emerging focus on multimodal fusion, explainability, and 0.0

What carries the argument

The novel hierarchical taxonomy that structures the field by primary clinical task (diagnosis vs. prediction), data modality (text, speech, neuroimaging, multimodal), and computational model class (e.g., graph neural networks, large language models, hybrid approaches).

Load-bearing premise

The 55 selected studies constitute a representative and unbiased sample of the field such that the taxonomy and trend analysis accurately reflect current capabilities and gaps without significant selection bias.

What would settle it

A follow-up systematic review that includes additional studies and identifies substantially different predominant models or trends would falsify the current analysis.

read the original abstract

Major Depressive Disorder is one of the leading causes of disability worldwide, yet its diagnosis still depends largely on subjective clinical assessments. Integrating Artificial Intelligence (AI) holds promise for developing objective, scalable, and timely diagnostic tools. In this paper, we present a comprehensive survey of state-of-the-art AI methods for depression detection and diagnosis, based on a systematic review of 55 key studies. We introduce a novel hierarchical taxonomy that structures the field by primary clinical task (diagnosis vs. prediction), data modality (text, speech, neuroimaging, multimodal), and computational model class (e.g., graph neural networks, large language models, hybrid approaches). Our in-depth analysis reveals three major trends: the predominance of graph neural networks for modeling brain connectivity, the rise of large language models for linguistic and conversational data, and an emerging focus on multimodal fusion, explainability, and algorithmic fairness. Alongside methodological insights, we provide an overview of prominent public datasets and standard evaluation metrics as a practical guide for researchers. By synthesizing current advances and highlighting open challenges, this survey offers a comprehensive roadmap for future innovation in computational psychiatry.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The manuscript presents a systematic review of 55 studies on AI methods for Major Depressive Disorder detection and diagnosis. It proposes a novel hierarchical taxonomy organized by clinical task (diagnosis vs. prediction), data modality (text, speech, neuroimaging, multimodal), and model class (e.g., GNNs, LLMs, hybrids). The analysis identifies three trends: GNN predominance for brain connectivity, rising use of LLMs for linguistic data, and growing emphasis on multimodal fusion, explainability, and fairness. It also surveys public datasets and evaluation metrics to guide researchers.

Significance. If the 55-study sample is representative, the taxonomy and trend synthesis could provide a useful organizing framework and roadmap for computational psychiatry, with practical value in the dataset and metric overviews. The work's significance is limited by the fast pace of the field, where even a well-chosen sample risks rapid obsolescence, and by the absence of quantitative performance aggregation across studies.

major comments (2)
  1. [Methods] Methods section: The literature search strategy, databases queried, keyword sets, date range, and explicit inclusion/exclusion criteria are not described in sufficient detail to permit replication or evaluation of selection bias. This directly affects the load-bearing claim that the derived taxonomy and trends (GNN predominance, LLM rise, multimodal focus) accurately reflect the field.
  2. [Results] Results, trend analysis paragraph: The assertion of 'predominance of graph neural networks for modeling brain connectivity' is stated qualitatively without a supporting count or proportion from the 55 studies (e.g., how many of the neuroimaging papers actually use GNNs versus other models) or citation of comparative performance numbers, weakening the trend's evidentiary basis.
minor comments (2)
  1. [Figure 1] Figure 1 (taxonomy diagram): The hierarchical levels are not labeled with explicit parent-child connectors or legend entries, making it difficult to trace how modalities map to model classes.
  2. [Section 5] Section 5 (datasets and metrics): Several dataset descriptions lack the original publication year or size statistics, reducing their utility as a practical guide.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the detailed and constructive report. We agree that greater methodological transparency and quantitative support for the identified trends will improve the manuscript's rigor and replicability. We have prepared revisions to address both major comments directly.

read point-by-point responses

  1. Referee: [Methods] Methods section: The literature search strategy, databases queried, keyword sets, date range, and explicit inclusion/exclusion criteria are not described in sufficient detail to permit replication or evaluation of selection bias. This directly affects the load-bearing claim that the derived taxonomy and trends (GNN predominance, LLM rise, multimodal focus) accurately reflect the field.

    Authors: We acknowledge this limitation in the current draft. The original manuscript summarized the review process at a high level but omitted a dedicated Methods section with operational details. In the revised version we will insert a new Methods section that explicitly lists: (1) databases queried (PubMed, IEEE Xplore, ACM Digital Library, arXiv, and Google Scholar), (2) the complete Boolean search strings (e.g., (“major depressive disorder” OR depression) AND (“machine learning” OR “deep learning” OR “graph neural network” OR “large language model” OR multimodal)), (3) the date range (1 January 2018 – 31 December 2024), and (4) the full inclusion/exclusion criteria (peer-reviewed original research on AI-based MDD detection or prediction in human subjects; English language; exclusion of reviews, conference abstracts without full text, and studies focused solely on other psychiatric conditions). These additions will enable replication and allow readers to assess selection bias. The taxonomy and trend claims will be explicitly caveated as reflecting the 55 studies that met these criteria. revision: yes

  2. Referee: [Results] Results, trend analysis paragraph: The assertion of 'predominance of graph neural networks for modeling brain connectivity' is stated qualitatively without a supporting count or proportion from the 55 studies (e.g., how many of the neuroimaging papers actually use GNNs versus other models) or citation of comparative performance numbers, weakening the trend's evidentiary basis.

    Authors: We accept the critique that the claim currently rests on qualitative description. In the revised manuscript we will replace the qualitative statement with quantitative counts drawn from our 55-study corpus. Specifically, we will report that 18 of the 55 studies used neuroimaging data and that 13 of those 18 (72 %) employed graph neural networks for connectivity modeling, compared with 3 using CNNs and 2 using traditional statistical models. We will add a supplementary table breaking down model class by data modality. Regarding performance, we will note that direct meta-analysis is precluded by heterogeneous datasets and metrics; however, we will cite representative studies in which GNN variants outperformed non-graph baselines on the same fMRI or EEG cohorts (e.g., accuracy gains of 4–9 % reported in cited works). These concrete figures and citations will strengthen the evidentiary basis while preserving the original interpretive framing. revision: yes

Circularity Check

0 steps flagged

No circularity: survey aggregates external studies without self-referential derivations

full rationale

This paper is a systematic literature review that selects and organizes 55 external studies into a hierarchical taxonomy based on clinical task, data modality, and model class. The trends (GNN predominance, LLM rise, multimodal focus) are reported as observations from the reviewed literature rather than derived via equations or predictions internal to the paper. No mathematical derivations, fitted parameters, or uniqueness theorems appear; the taxonomy is an organizational framework applied to independent prior work. Self-citations, if present, are not load-bearing for the central claims, and the selection process is described as systematic without reducing to author-defined inputs by construction. The survey is self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central claims rest on the assumption that the reviewed studies are representative; no free parameters, invented entities, or additional axioms beyond standard review methodology are introduced.

axioms (1)
  • domain assumption The 55 key studies provide a comprehensive and unbiased representation of state-of-the-art AI methods for depression detection and diagnosis.
    The taxonomy, trend analysis, and roadmap depend directly on this selection being representative.

pith-pipeline@v0.9.0 · 5732 in / 1228 out tokens · 68569 ms · 2026-05-18T23:02:09.222354+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

83 extracted references · 83 canonical work pages · 2 internal anchors

  1. [1]

    American Psychiatric Association, Diagnostic and statistical manual of mental disorders, 5th edition, text revision (DSM-5-TR), American Psychiatric Association Publishing, Washington, DC, 2022

    work page 2022

  2. [2]

    URL https://www.who.int/news-room/fact-sheets/detail/depression

    World Health Organization, Depressive disorder (depression), WHO NewsroomAccessed: July 19, 2025 (March 2023). URL https://www.who.int/news-room/fact-sheets/detail/depression

    work page 2025

  3. [3]

    G. S. Malhi, J. J. Mann, Depression, The Lancet 392 (10161) (2018) 2299–2312

    work page 2018

  4. [4]

    Evans-Lacko, M

    S. Evans-Lacko, M. Knapp, The economic impact of depression in high-income countries: a systematic review, The Journal of Mental Health Policy and Economics 19 (1) (2021) 37

    work page 2021

  5. [5]

    H. J. Lethus, D. Schouten, O. A. van den Heuvel, D. J. Veltman, R. van de Mheen, Artificial intelligence for mental health: a systematic review of opportunities, challenges, and future directions, Translational Psychiatry 13 (1) (2023) 344. 29

    work page 2023

  6. [6]

    A. Deng, R. Guidotti, M. A. Al-garadi, A. Sarker, Artificial intelligence in mental health, Artificial Intelligence in Medicine (2024) 102761

    work page 2024

  7. [7]

    Cheng, H

    J. Cheng, H. Li, J. Zhang, D. Tao, A review of multimodal data fusion for mental disorder detection, IEEE Transactions on Affective Computing (2024)

    work page 2024

  8. [8]

    J. W. Gichoya, I. Banerjee, H. Trivedi, M. Pugh, P. Film, C. Campos, et al., Ai recognition of patient race in medical imaging: a modelling study, The Lancet Digital Health 4 (6) (2022) e406–e414

    work page 2022

  9. [9]

    G. Shen, J. Jia, L. Nie, F. Feng, C. Zhang, T. Hu, T.-S. Chua, W.-l. Li, Depression detection from social media: A survey, IEEE Transactions on Affective Computing 12 (4) (2017) 915–935

    work page 2017

  10. [10]

    Apicella, A

    F. Apicella, A. Fagiolini, C. Gesi, Machine learning-based classification of major depressive dis- order using structural and functional neuroimaging: a systematic review, Molecular Psychiatry 28 (1) (2023) 262–274

    work page 2023

  11. [11]

    A. P. Association, What is depression?, Washington, DC: American Psychiatric Association (2018)

    work page 2018

  12. [12]

    T. N. Kipf, M. Welling, Semi-supervised classification with graph convolutional networks, in: 5th International Conference on Learning Representations, ICLR 2017, Toulon, France, April 24-26, 2017, Conference Track Proceedings, OpenReview.net, 2017. URL https://openreview.net/forum?id=SJU4ayYgl

    work page 2017

  13. [13]

    Z. Wu, S. Pan, F. Chen, G. Long, C. Zhang, P. S. Yu, A comprehensive survey on graph neural networks, IEEE transactions on neural networks and learning systems 32 (1) (2020) 4–24

    work page 2020

  14. [14]

    M. H. Chehreghani, Half a decade of graph convolutional networks, Nat. Mach. Intell. 4 (3) (2022) 192–193. doi:10.1038/S42256-022-00466-8 . URL https://doi.org/10.1038/s42256-022-00466-8

    work page doi:10.1038/s42256-022-00466-8 2022

  15. [15]

    Zohrabi, S

    M. Zohrabi, S. Saravani, M. H. Chehreghani, Centrality-based and similarity-based neigh- borhood extension in graph neural networks, J. Supercomput. 80 (16) (2024) 24638–24663. doi:10.1007/S11227-024-06336-X . URL https://doi.org/10.1007/s11227-024-06336-x

    work page doi:10.1007/s11227-024-06336-x 2024

  16. [16]

    Strong transitivity relations and graph neural networks

    Y. Mohamadi, M. H. Chehreghani, Strong transitivity relations and graph neural networks, CoRR abs/2401.01384 (2024). arXiv:2401.01384, doi:10.48550/ARXIV.2401.01384. URL https://doi.org/10.48550/arXiv.2401.01384

    work page doi:10.48550/arxiv.2401.01384 2024

  17. [17]

    Gholamzadeh Nasrabadi, A

    F. Gholamzadeh Nasrabadi, A. Kashani, P. Zahedi, M. Haghir Chehreghani, Content aug- mented graph neural networks, ACM Trans. WebJust Accepted (Oct. 2024). doi:10.1145/ 3700790. URL https://doi.org/10.1145/3700790

    work page doi:10.1145/3700790 2024

  18. [18]

    C. I. Kanatsoulis, E. Choi, S. Jegelka, J. Leskovec, A. Ribeiro, Learning efficient positional en- codings with graph neural networks, in: The Thirteenth International Conference on Learning Representations, ICLR 2025, Singapore, April 24-28, 2025, OpenReview.net, 2025. URL https://openreview.net/forum?id=AWg2tkbydO 30

    work page 2025

  19. [19]

    M. Gori, G. Monfardini, F. Scarselli, A new model for learning in graph domains, in: Pro- ceedings. 2005 IEEE International Joint Conference on Neural Networks, 2005., Vol. 2, IEEE, 2005, pp. 729–734

    work page 2005

  20. [20]

    W. X. Zhao, K. Zhou, J. Li, T. Tang, X. Wang, Y. Hou, Y. Min, B. Zhang, J. Zhang, Z. Dong, et al., A survey of large language models, arXiv preprint arXiv:2303.18223 (2023)

    work page internal anchor Pith review Pith/arXiv arXiv 2023

  21. [21]

    Vaswani, N

    A. Vaswani, N. Shazeer, N. Parmar, J. Uszkoreit, L. Jones, A. N. Gomez, L. Kaiser, I. Polo- sukhin, Attention is all you need, in: Advances in neural information processing systems, 2017, pp. 5998–6008

    work page 2017

  22. [22]

    Devlin, M.-W

    J. Devlin, M.-W. Chang, K. Lee, K. Toutanova, Bert: Pre-training of deep bidirectional trans- formers for language understanding, in: Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Tech- nologies, Volume 1 (Long and Short Papers), 2018, pp. 4171–4186

    work page 2019

  23. [23]

    Niedermeyer, F

    E. Niedermeyer, F. L. da Silva, Electroencephalography: basic principles, clinical applications, and related fields, Lippincott Williams & Wilkins, 2005

    work page 2005

  24. [24]

    Teplan, Fundamentals of eeg measurement, Measurement science review 2 (2) (2002) 1–11

    M. Teplan, Fundamentals of eeg measurement, Measurement science review 2 (2) (2002) 1–11

    work page 2002

  25. [25]

    Al-Farras, A

    M. Al-Farras, A. Al-Wabil, H. Al-Negheimish, Z. Al-Hussain, W. Al-Shehri, F. Al-Sultan, A review of eeg-based diagnosis of psychiatric disorders, Sensors 23 (17) (2023) 7481

    work page 2023

  26. [26]

    Deshpande, V

    M. Deshpande, V. Rao, Depression detection using emotion artificial intelligence, in: 2017 International Conference on Intelligent Sustainable Systems (ICISS), 2017, pp. 858–862. doi: 10.1109/ISS1.2017.8389299

    work page doi:10.1109/iss1.2017.8389299 2017

  27. [27]

    Huang, F

    X. Huang, F. Wang, Y. Gao, et al., Depression recognition using voice-based pre-training model, Scientific Reports 14 (2024) 12734. doi:10.1038/s41598-024-63556-0 . URL https://doi.org/10.1038/s41598-024-63556-0

    work page doi:10.1038/s41598-024-63556-0 2024

  28. [28]

    T. T. Prama, M. S. Islam, M. M. Anwar, I. Jahan, Ai-enabled deep depression detection and evaluation informed by dsm-5-tr, IEEE Transactions on Computational Social Systems 11 (5) (2024) 6453–6465. doi:10.1109/TCSS.2024.3382139

    work page doi:10.1109/tcss.2024.3382139 2024

  29. [29]

    M. O. Nusrat, W. Shahzad, S. A. Jamal, Multi class depression detection through tweets using artificial intelligence (2024). arXiv:2404.13104. URL https://arxiv.org/abs/2404.13104

    work page arXiv 2024

  30. [30]

    Patil, N

    S. Patil, N. Jagtap, K. Jadhav, A. Desai, A. Shaikh, Depression detection using convolutional neural network, International Journal for Research in Applied Science and Engineering Tech- nology (IJRASET)Accessed: Apr. 12, 2025 (2023)

    work page 2025

  31. [31]

    H. Liu, C. Li, X. Zhang, F. Zhang, W. Wang, F. Ma, H. Chen, H. Yu, X. Zhang, Depression de- tection via capsule networks with contrastive learning, in: Proceedings of the AAAI Conference on Artificial Intelligence, Vol. 38, 2024, pp. 22231–22239. doi:10.1609/aaai.v38i20.30228

    work page doi:10.1609/aaai.v38i20.30228 2024

  32. [32]

    Tejaswini, K

    V. Tejaswini, K. Sathya Babu, B. Sahoo, Depression detection from social media text analysis using natural language processing techniques and hybrid deep learning model, ACM Trans. Asian Low-Resour. Lang. Inf. Process. 23 (1) (Jan. 2024). doi:10.1145/3569580. URL https://doi.org/10.1145/3569580 31

    work page doi:10.1145/3569580 2024

  33. [33]

    J. Zhu, C. Jiang, J. Chen, X. Lin, R. Yu, X. Li, B. Hu, Eeg based depression recognition using improved graph convolutional neural network, Computers in Biology and Medicine 148 (2022) 105815. doi:https://doi.org/10.1016/j.compbiomed.2022.105815. URL https://www.sciencedirect.com/science/article/pii/S0010482522005765

    work page doi:10.1016/j.compbiomed.2022.105815 2022

  34. [34]

    S. Liu, J. Zhou, X. Zhu, Y. Zhang, X. Zhou, S. Zhang, Z. Yang, Z. Wang, R. Wang, Y. Yuan, X. Fang, X. Chen, Y. Wang, L. Zhang, G. Wang, C. Jin, An objective quantitative diagnosis of depression using a local-to-global multimodal fusion graph neural network, Patterns 5 (12) (2024) 101081. doi:https://doi.org/10.1016/j.patter.2024.101081. URL https://www.sc...

    work page doi:10.1016/j.patter.2024.101081 2024

  35. [35]

    Q. Yu, R. Wang, J. Liu, L. Hu, M. Chen, Z. Liu, Gnn-based depression recognition using spatio- temporal information: A fnirs study, IEEE Journal of Biomedical and Health Informatics 26 (10) (2022) 4925–4935. doi:10.1109/JBHI.2022.3195066

    work page doi:10.1109/jbhi.2022.3195066 2022

  36. [36]

    Z. Chen, J. Deng, J. Zhou, J. Wu, T. Qian, M. Huang, Depression detection in clinical in- terviews with LLM-empowered structural element graph, in: K. Duh, H. Gomez, S. Bethard (Eds.), Proceedings of the 2024 Conference of the North American Chapter of the Associ- ation for Computational Linguistics: Human Language Technologies (Volume 1: Long Pa- pers), ...

    work page doi:10.18653/v1/2024.naacl-long.452 2024

  37. [37]

    Zhang, H

    X. Zhang, H. Liu, K. Xu, Q. Zhang, D. Liu, B. Ahmed, J. Epps, When LLMs meets acoustic landmarks: An efficient approach to integrate speech into large language models for depression detection, in: Y. Al-Onaizan, M. Bansal, Y.-N. Chen (Eds.), Proceedings of the 2024 Confer- ence on Empirical Methods in Natural Language Processing, Association for Computati...

    work page doi:10.18653/v1/2024.emnlp-main.8 2024

  38. [38]

    Zheng, S

    K. Zheng, S. Yu, B. Li, R. Jenssen, B. Chen, Brainib: Interpretable brain network-based psy- chiatric diagnosis with graph information bottleneck, IEEE Transactions on Neural Networks and Learning Systems (2024) 1–14 doi:10.1109/TNNLS.2024.3449419

    work page doi:10.1109/tnnls.2024.3449419 2024

  39. [39]

    Nguyen, A

    T. Nguyen, A. Yates, A. Zirikly, B. Desmet, A. Cohan, Improving the generalizability of depression detection by leveraging clinical questionnaires (2022). arXiv:2204.10432. URL https://arxiv.org/abs/2204.10432

    work page arXiv 2022

  40. [40]

    Danner, B

    M. Danner, B. Hadzic, S. Gerhardt, S. Ludwig, I. Uslu, P. Shao, T. Weber, Y. Shiban, M. Ratsch, Advancing mental health diagnostics: Gpt-based method for depression detec- tion, in: 2023 62nd Annual Conference of the Society of Instrument and Control Engineers (SICE), 2023, pp. 1290–1296. doi:10.23919/SICE59929.2023.10354236

    work page doi:10.23919/sice59929.2023.10354236 2023

  41. [41]

    T. Zhao, G. Zhang, Enhancing major depressive disorder diagnosis with dynamic-static fusion graph neural networks, IEEE Journal of Biomedical and Health Informatics 28 (8) (2024) 4701–4710. doi:10.1109/JBHI.2024.3395611

    work page doi:10.1109/jbhi.2024.3395611 2024

  42. [42]

    Sadeghi, B

    M. Sadeghi, B. Egger, R. Agahi, R. Richer, K. Capito, L. H. Rupp, L. Schindler-Gmelch, M. Berking, B. M. Eskofier, Exploring the capabilities of a language model-only approach for 32 depression detection in text data, in: 2023 IEEE EMBS International Conference on Biomedical and Health Informatics (BHI), 2023, pp. 1–5. doi:10.1109/BHI58575.2023.10313367

    work page doi:10.1109/bhi58575.2023.10313367 2023

  43. [43]

    T. Chen, R. Hong, Y. Guo, S. Hao, B. Hu, Ms ²-gnn: Exploring gnn-based multimodal fusion network for depression detection, IEEE Transactions on Cybernetics 53 (12) (2023) 7749–7759. doi:10.1109/TCYB.2022.3197127

    work page doi:10.1109/tcyb.2022.3197127 2023

  44. [44]

    J. Kim, M. Sohn, Graph representation learning-based early depression detection framework in smart home environments, Sensors 22 (4) (2022). doi:10.3390/s22041545. URL https://www.mdpi.com/1424-8220/22/4/1545

    work page doi:10.3390/s22041545 2022

  45. [45]

    Zhang, D

    M. Zhang, D. Long, Z. Chen, C. Fang, Y. Li, P. Huang, F. Chen, H. Sun, Multi-view graph network learning framework for identification of major depressive disorder, Computers in Biol- ogy and Medicine 166 (2023) 107478. doi:https://doi.org/10.1016/j.compbiomed.2023. 107478. URL https://www.sciencedirect.com/science/article/pii/S0010482523009435

    work page doi:10.1016/j.compbiomed.2023 2023

  46. [46]

    M. Li, X. Sun, M. Wang, Detecting depression with heterogeneous graph neural network in clinical interview transcript, IEEE Transactions on Computational Social Systems 11 (1) (2024) 1315–1324. doi:10.1109/TCSS.2023.3263056

    work page doi:10.1109/tcss.2023.3263056 2024

  47. [47]

    Z. Xia, Y. Fan, K. Li, Y. Wang, L. Huang, F. Zhou, Depressiongraph: A two-channel graph neural network for the diagnosis of major depressive disorders using rs-fmri, Electronics 12 (24) (2023). doi:10.3390/electronics12245040. URL https://www.mdpi.com/2079-9292/12/24/5040

    work page doi:10.3390/electronics12245040 2023

  48. [48]

    T. Chen, Y. Guo, S. Hao, R. Hong, Exploring self-attention graph pooling with eeg-based topological structure and soft label for depression detection, IEEE Transactions on Affective Computing 13 (4) (2022) 2106–2118. doi:10.1109/TAFFC.2022.3210958

    work page doi:10.1109/taffc.2022.3210958 2022

  49. [49]

    Noman, C.-M

    F. Noman, C.-M. Ting, H. Kang, R. C.-W. Phan, H. Ombao, Graph autoencoders for em- bedding learning in brain networks and major depressive disorder identification, IEEE Journal of Biomedical and Health Informatics 28 (3) (2024) 1644–1655. doi:10.1109/JBHI.2024. 3351177

    work page doi:10.1109/jbhi.2024 2024

  50. [50]

    Venkatapathy, M

    S. Venkatapathy, M. Votinov, L. Wagels, S. Kim, M. Lee, U. Habel, I.-H. Ra, H.-G. Jo, Ensemble graph neural network model for classification of major depressive disorder using whole-brain functional connectivity, Frontiers in Psychiatry Volume 14 - 2023 (2023). doi:10.3389/fpsyt.2023.1125339. URL https://www.frontiersin.org/journals/psychiatry/articles/10...

    work page doi:10.3389/fpsyt.2023.1125339 2023

  51. [51]

    W. Ma, Y. Wang, N. Ma, Y. Ding, Diagnosis of major depressive disorder using a novel interpretable gcn model based on resting state fmri, Neuroscience 566 (2025) 124–131. doi: 10.1016/j.neuroscience.2024.12.045. URL https://www.sciencedirect.com/science/article/pii/S0306452224007528

    work page doi:10.1016/j.neuroscience.2024.12.045 2025

  52. [52]

    Y. Wang, W. Zheng, Y. Li, H. Yang, A hybrid graph neural network for enhanced eeg-based depression detection (2024). arXiv:2410.18103. URL https://arxiv.org/abs/2410.18103 33

    work page internal anchor Pith review Pith/arXiv arXiv 2024

  53. [53]

    C. Sun, M. Jiang, L. Gao, Y. Xin, Y. Dong, A novel study for depression detecting using audio signals based on graph neural network, Biomedical Signal Processing and Control 88 (2024) 105675. doi:https://doi.org/10.1016/j.bspc.2023.105675. URL https://www.sciencedirect.com/science/article/pii/S1746809423011084

    work page doi:10.1016/j.bspc.2023.105675 2024

  54. [54]

    S. Teng, J. Liu, R. K. Jain, S. Chai, R. Hou, T. Tateyama, L. Lin, Y. wei Chen, Enhancing depression detection with chain-of-thought prompting: From emotion to reasoning using large language models (2025). arXiv:2502.05879. URL https://arxiv.org/abs/2502.05879

    work page arXiv 2025

  55. [55]

    Zhang, Z

    L. Zhang, Z. Gao, D. Zhou, Y. He, Explainable depression detection in clinical interviews with personalized retrieval-augmented generation (2025). arXiv:2503.01315. URL https://arxiv.org/abs/2503.01315

    work page arXiv 2025

  56. [56]

    Zhang, H

    X. Zhang, H. Liu, Q. Zhang, B. Ahmed, J. Epps, Speecht-rag: Reliable depression detection in llms with retrieval-augmented generation using speech timing information (2025). arXiv: 2502.10950. URL https://arxiv.org/abs/2502.10950

    work page arXiv 2025

  57. [57]

    Z. Xu, C. L. P. Chen, T. Zhang, Tfagl: A novel agent graph learning method using time- frequency eeg for major depressive disorder detection, IEEE Transactions on Affective Com- puting (2025) 1–14 doi:10.1109/TAFFC.2025.3527459

    work page doi:10.1109/taffc.2025.3527459 2025

  58. [58]

    C. Yang, X. Dong, X. Zong, Frequency feature fusion graph network for depression diagnosis via fnirs (2025). arXiv:2504.21064. URL https://arxiv.org/abs/2504.21064

    work page arXiv 2025

  59. [59]

    S. Kim, S. H. Bong, S. Yun, D. Kim, J. H. Yoo, K. S. Choi, H. Park, H. J. Jeon, J.-H. Kim, J. H. Jang, B. Jeong, Neurobiologically interpretable causal connectome for predicting young adult depression: A graph neural network study, Journal of Affective Disorders 377 (2025) 225–234. doi:10.1016/j.jad.2025.02.076. URL https://www.sciencedirect.com/science/a...

    work page doi:10.1016/j.jad.2025.02.076 2025

  60. [60]

    J. Cha, S. Kim, D. Kim, E. Park, Mogam: A multimodal object-oriented graph attention model for depression detection (2024). arXiv:2403.15485. URL https://arxiv.org/abs/2403.15485

    work page arXiv 2024

  61. [61]

    Z. Yan, F. Peng, D. Zhang, Decen: A deep learning model enhanced by depressive emotions for depression detection from social media content, Decision Support Systems 191 (2025) 114421. doi:10.1016/j.dss.2025.114421. URL https://www.sciencedirect.com/science/article/pii/S0167923625000223

    work page doi:10.1016/j.dss.2025.114421 2025

  62. [62]

    F. Zhu, J. Zhang, R. Dang, B. Hu, Q. Wang, MTNet: Multimodal transformer network for mild depression detection through fusion of EEG and eye tracking, Biomedical Signal Processing and Control 100 (2025) 106996. doi:10.1016/j.bspc.2024.106996. URL https://www.sciencedirect.com/science/article/pii/S1746809424010541

    work page doi:10.1016/j.bspc.2024.106996 2025

  63. [63]

    Naseem, A

    U. Naseem, A. Dunn, J. Kim, M. Khushi, Detection of depression severity in social media text using transformer-based models, Information 16 (2) (2025) 114. doi:10.3390/info16020114. URL https://www.mdpi.com/2078-2489/16/2/114 34

    work page doi:10.3390/info16020114 2025

  64. [64]

    E. Lim, M. Jhon, J.-W. Kim, S.-H. Kim, S. Kim, H.-J. Yang, A lightweight approach based on cross-modality for depression detection, Computers in Biology and Medicine 186 (2025) 109618. doi:10.1016/j.compbiomed.2024.109618. URL https://www.sciencedirect.com/science/article/pii/S0010482524017037

    work page doi:10.1016/j.compbiomed.2024.109618 2025

  65. [65]

    Cheong, A

    J. Cheong, A. Bangar, S. Kalkan, H. Gunes, U-fair: Uncertainty-based multimodal multitask learning for fairer depression detection (2025). arXiv:2501.09687. URL https://arxiv.org/abs/2501.09687

    work page arXiv 2025

  66. [66]

    A. M. H. Kwok, J. Cheong, S. Kalkan, H. Gunes, Machine learning fairness for depression detection using eeg data (2025). arXiv:2501.18192. URL https://arxiv.org/abs/2501.18192

    work page arXiv 2025

  67. [67]

    T. Xing, Y. Dou, X. Chen, et al., An adaptive multi-graph neural network with multimodal feature fusion learning for mdd detection, Scientific Reports 14 (2024) 28400. doi:10.1038/ s41598-024-79981-0

    work page 2024

  68. [68]

    A.-T. Kuo, H. Chen, Y.-H. Kuo, W.-S. Ku, Dynamic graph representation learning for depres- sion screening with transformer (2023). arXiv:2305.06447. URL https://arxiv.org/abs/2305.06447

    work page arXiv 2023

  69. [69]

    Kamoji, S

    S. Kamoji, S. Rozario, S. Almeida, S. Patil, S. Patankar, H. Pendhari, Mental health pre- diction using machine learning models and large language model, in: 2024 Second Inter- national Conference on Inventive Computing and Informatics (ICICI), 2024, pp. 185–190. doi:10.1109/ICICI62254.2024.00040

    work page doi:10.1109/icici62254.2024.00040 2024

  70. [70]

    2023 , journal =

    K. Sabaneh, M. A. Salameh, F. Khaleel, M. M. Herzallah, J. Y. Natsheh, M. Maree, Early risk prediction of depression based on social media posts in arabic, in: 2023 IEEE 35th International Conference on Tools with Artificial Intelligence (ICTAI), 2023, pp. 595–602. doi:10.1109/ ICTAI59109.2023.00094

    work page arXiv 2023

  71. [71]

    G. Luo, H. Rao, P. An, Y. Li, R. Hong, W. Chen, S. Chen, Exploring adaptive graph topologies and temporal graph networks for eeg-based depression detection, IEEE Transactions on Neural Systems and Rehabilitation Engineering 31 (2023) 3947–3957. doi:10.1109/TNSRE.2023. 3320693

    work page doi:10.1109/tnsre.2023 2023

  72. [72]

    Bidja, Depressiongnn: Depression prediction using graph neural network on smartphone and wearable sensors, Honors scholar thesis, University of Connecticut (May 2019)

    P. Bidja, Depressiongnn: Depression prediction using graph neural network on smartphone and wearable sensors, Honors scholar thesis, University of Connecticut (May 2019). URL https://digitalcommons.lib.uconn.edu/srhonors_theses/607/

    work page 2019

  73. [73]

    W. Qin, Z. Chen, L. Wang, Y. Lan, W. Ren, R. Hong, Read, diagnose and chat: Towards explainable and interactive llms-augmented depression detection in social media (2023).arXiv: 2305.05138. URL https://arxiv.org/abs/2305.05138

    work page arXiv 2023

  74. [74]

    S. M. Shah, S. A. Gillani, M. S. A. Baig, M. A. Saleem, M. H. Siddiqui, Advancing depression detection on social media platforms through fine-tuned large language models, Online Social Networks and Media 46 (2025) 100311. doi:10.1016/j.osnem.2025.100311. URL https://www.sciencedirect.com/science/article/pii/S2468696425000126 35

    work page doi:10.1016/j.osnem.2025.100311 2025

  75. [75]

    S. Teng, J. Liu, H. Sun, S. Chai, T. Tateyama, L. Lin, Y.-W. Chen, Enhanced multi- modal depression detection with emotion prompts, in: ICASSP 2025 - 2025 IEEE Inter- national Conference on Acoustics, Speech and Signal Processing (ICASSP), 2025, pp. 1–5. doi:10.1109/ICASSP49660.2025.10889035

    work page doi:10.1109/icassp49660.2025.10889035 2025

  76. [76]

    P. A. P´ erez-Toro, J. Dineley, R. Iniesta, et al., Exploring biases related to the use of large language models in a multilingual depression corpus, Research SquarePreprint (jan 2025). doi:10.21203/rs.3.rs-5731243/v1

    work page doi:10.21203/rs.3.rs-5731243/v1 2025

  77. [77]

    Y. Jiao, K. Zhao, X. Wei, et al., Deep graph learning of multimodal brain networks defines treatment-predictive signatures in major depression, Molecular Psychiatry (2025). doi:10. 1038/s41380-025-02974-6

    work page 2025

  78. [78]

    Kumar, A

    A. Kumar, A. Sharma, S. R. Sangwan, Transformer-based abstractive summarization for de- pression detection literature for enhanced medical insights, AuthoreaPreprint (January 09 2025)

    work page 2025

  79. [79]

    Ibrahimov, T

    Y. Ibrahimov, T. Anwar, T. Yuan, Depressionx: Knowledge infused residual attention for explainable depression severity assessment (2025). arXiv:2501.14985. URL https://arxiv.org/abs/2501.14985

    work page arXiv 2025

  80. [80]

    D. E. Merzougui, G. Dias, J. Pantin, F. Maurel, Evaluating large language models for depres- sion symptom estimation, in: R. Bellazzi, J. M. Juarez Herrero, L. Sacchi, B. Zupan (Eds.), Artificial Intelligence in Medicine. AIME 2025, Vol. 15735 of Lecture Notes in Computer Sci- ence, Springer, Cham, 2025, pp. 589–599. doi:10.1007/978-3-031-95841-0_51 . URL...

    work page doi:10.1007/978-3-031-95841-0_51 2025

Showing first 80 references.