Qiong Hong
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Qiong Hong has a total of 3 co-authors. Here are the top co-authors by number of shared publications:
- Lailatul Qadri Zakaria
Faculty of Information Science and Technology, Universiti Kebangsaan Malaysia, Bangi, Malaysia
3 shared publications Sabrina Tiun
2 shared publications- Sabrina Tiun
Faculty of Information Science and Technology, Universiti Kebangsaan Malaysia (UKM), Bangi 43600, Malaysia
1 shared publication
Latest Publications
- Open Access
- 31 July 2026
- Published by PeerJ in PeerJ Computer Science
- Vol. 12, e3935
A systematic review of multimodal depression detection: from technological pipelines to the clinical depression management cycle
The effective management of depression—a high-prevalence mental disorder defined by persistent low mood and complex psychosomatic symptoms—poses a significant global challenge. It is necessary to conduct objective, multi-source assessments throughout the entire depression management cycle (DMC), from early screening to long-term prognosis. Multimodal depression detection (MMDD) has emerged as a critical enabling technology that leverages artificial intelligence (AI) to integrate these multi-source data streams. We conducted a Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA)-compliant search across major digital libraries (2015–2025; through Oct 23, 2025) and included 162 studies, synthesizing the technical pipeline—datasets, feature engineering, models, and fusion strategies. To unify fragmented perspectives, we introduce the AI–driven depression management cycle (AI-DMC), which maps MMDD applications onto two task paradigms: low-fidelity, large-scale monitoring (screening, prognosis) and high-fidelity, high-precision detection (diagnosis support, treatment assessment). Grounded in AI-DMC, we outline a clinically oriented roadmap and identify system-level gaps—data and annotation ecosystems, trustworthy fusion and interpretability, and translation from monitoring to clinical action—providing forward-looking guidance for research and deployment.
- 31 July 2026
- Published by PeerJ in PeerJ Computer Science
- Vol. 12, e3935
Open AccessA systematic review of multimodal depression detection: from technological pipelines to the clinical depression management cycle
The effective management of depression—a high-prevalence mental disorder defined by persistent low mood and complex psychosomatic symptoms—poses a significant global challenge. It is necessary to conduct objective, multi-source assessments throughout the entire depression management cycle (DMC), from early screening to long-term prognosis. Multimodal depression detection (MMDD) has emerged as a critical enabling technology that leverages artificial intelligence (AI) to integrate these multi-source data streams. We conducted a Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA)-compliant search across major digital libraries (2015–2025; through Oct 23, 2025) and included 162 studies, synthesizing the technical pipeline—datasets, feature engineering, models, and fusion strategies. To unify fragmented perspectives, we introduce the AI–driven depression management cycle (AI-DMC), which maps MMDD applications onto two task paradigms: low-fidelity, large-scale monitoring (screening, prognosis) and high-fidelity, high-precision detection (diagnosis support, treatment assessment). Grounded in AI-DMC, we outline a clinically oriented roadmap and identify system-level gaps—data and annotation ecosystems, trustworthy fusion and interpretability, and translation from monitoring to clinical action—providing forward-looking guidance for research and deployment.
- Open Access
- 22 May 2026
- Published by MDPI AG in Information
- Vol. 17 (6) , 516
NS-Dep-KAN: An Explainable Neuro-Symbolic Framework with Kolmogorov–Arnold Networks for DSM-Guided Depression Assessment
Automated depression assessment is critical for scalable mental healthcare but faces dual challenges: the lack of clinical interpretability in “black-box” deep learning models and the excessive computational cost of large-scale fusion architectures. To bridge this gap, we propose NS-Dep-KAN, a novel neuro-symbolic framework that harmonizes DSM-5-guided reasoning with Kolmogorov–Arnold Networks (KANs). Our approach leverages a Large Language Model (LLM) to extract symbolic symptom evidence aligned with diagnostic criteria, which then guides the aggregation of multimodal features from frozen pretrained encoders (WavLM and Qwen). Unlike traditional Multi-Layer Perceptrons, the proposed KAN prediction head employs learnable B-spline activation functions to capture complex nonlinear symptom–severity mappings with extreme parameter efficiency. Evaluations on the DAIC-WOZ benchmark demonstrate that NS-Dep-KAN achieves state-of-the-art performance among audio-text models (MAE 2.69, 13.5% improvement over the three-modality baseline MSGAF at MAE 3.11), with only ∼4.9 K trainable parameters. Moreover, the framework offers inherent interpretability, revealing granular symptom contribution profiles that align with clinical intuition. This work establishes a path toward explainable trustworthy AI for mental health screening.
- 22 May 2026
- Published by MDPI AG in Information
- Vol. 17 (6) , 516
Open AccessNS-Dep-KAN: An Explainable Neuro-Symbolic Framework with Kolmogorov–Arnold Networks for DSM-Guided Depression Assessment
Automated depression assessment is critical for scalable mental healthcare but faces dual challenges: the lack of clinical interpretability in “black-box” deep learning models and the excessive computational cost of large-scale fusion architectures. To bridge this gap, we propose NS-Dep-KAN, a novel neuro-symbolic framework that harmonizes DSM-5-guided reasoning with Kolmogorov–Arnold Networks (KANs). Our approach leverages a Large Language Model (LLM) to extract symbolic symptom evidence aligned with diagnostic criteria, which then guides the aggregation of multimodal features from frozen pretrained encoders (WavLM and Qwen). Unlike traditional Multi-Layer Perceptrons, the proposed KAN prediction head employs learnable B-spline activation functions to capture complex nonlinear symptom–severity mappings with extreme parameter efficiency. Evaluations on the DAIC-WOZ benchmark demonstrate that NS-Dep-KAN achieves state-of-the-art performance among audio-text models (MAE 2.69, 13.5% improvement over the three-modality baseline MSGAF at MAE 3.11), with only ∼4.9 K trainable parameters. Moreover, the framework offers inherent interpretability, revealing granular symptom contribution profiles that align with clinical intuition. This work establishes a path toward explainable trustworthy AI for mental health screening.