Mental health concerns are often expressed outside clinical settings, including in high-distress help seeking, where safety-critical guidance may be needed. Consumer health informatics systems increasingly incorporate large language models (LLMs) for mental health question answering, yet many evaluations underrepresent narrative, high-distress inquiries. We introduce UTCO (User, Topic, Context, Tone), a prompt construction framework that represents an inquiry as four controllable elements for systematic stress testing. Using 2,075 UTCO-generated prompts, we evaluated Llama 3.3 and annotated hallucinations (fabricated or incorrect clinical content) and omissions (missing clinically necessary or safety-critical guidance). Hallucinations occurred in 6.5% of responses and omissions in 13.2%, with omissions concentrated in crisis and suicidal ideation prompts. Across regression, element-specific matching, and similarity-matched comparisons, failures were most consistently associated with context and tone, while user-background indicators showed no systematic differences after balancing. These findings support evaluating omissions as a primary safety outcome and moving beyond static benchmark question sets.
Shivam Shukla, Emily Chen, Mahnaz Roshanaei
et al.
There has been a growing research interest in Digital Therapeutic Alliance (DTA) as the field of AI-powered conversational agents are being deployed in mental health care, particularly those delivering CBT (Cognitive Behaviour Therapy). Our proposition argues that the current design paradigm which seeks to optimize the bond between a patient in need of support and an AI agent contains a subtle but consequential trap: it risks producing an "appearance of connection" that unintentionally disrupts the fundamental human need for relatedness, which potentially displaces the authentic human relationships upon which long-term psychological recovery depends. We propose a reorientation from designing artificial intelligence tools that simulate relationships to designing AI that scaffolds them. To operationalize our argument, we propose an interdisciplinary model that translates the Responsible AI Six Sphere Framework through the lens of Self-Determination Theory (SDT), with a specific focus on the basic psychological need for relatedness. The resulting model offers the technical and other clinical communities a set of relationship-centered design guidelines and relevant provocations for building AI systems that function not just as companions, but as a catalyst for strengthening a patient's entire relational ecology; their connections with therapists, caregivers, family, and peers. In doing so, we discuss a model towards a more sustainable ecosystem of relationship-centered AI in mental health care.
Psychosocial online counselling frequently encounters generic subject lines that impede efficient case prioritisation. This study evaluates eleven large language models generating six-word subject lines for German counselling emails through hierarchical assessment - first categorising outputs, then ranking within categories to enable manageable evaluation. Nine assessors (counselling professionals and AI systems) enable analysis via Krippendorff's $α$, Spearman's $ρ$, Pearson's $r$ and Kendall's $τ$. Results reveal performance trade-offs between proprietary services and privacy-preserving open-source alternatives, with German fine-tuning consistently improving performance. The study addresses critical ethical considerations for mental health AI deployment including privacy, bias and accountability.
A BSTRACT This case study presents the therapeutic process and outcomes of an 8-year-old female client exhibiting emotional dysregulation, oppositional behaviors, and possible gender dysphoria following adoption disclosure and parental separation. Using nondirective, client-centered play therapy, the intervention focused on emotional expression, relational repair, and identity consolidation. Over 25 sessions with the child and six with the mother, play therapy helped the client access inner conflicts, develop self-regulation, and restore a sense of security. This case highlights the therapeutic use of play as a medium for integrating fragmented emotional experiences and emphasizes the importance of parallel parental coaching and reflective supervision in facilitating recovery. At 1-year follow-up, significant improvements in emotional regulation, attachment security, and behavioral functioning were sustained.
Samirah Bakker, Yao Ma, Seyed Sahand Mohammadi Ziabari
The complexity of mental healthcare billing enables anomalies, including fraud. While machine learning methods have been applied to anomaly detection, they often struggle with class imbalance, label scarcity, and complex sequential patterns. This study explores a hybrid deep learning approach combining Long Short-Term Memory (LSTM) networks and Transformers, with pseudo-labeling via Isolation Forests (iForest) and Autoencoders (AE). Prior work has not evaluated such hybrid models trained on pseudo-labeled data in the context of healthcare billing. The approach is evaluated on two real-world billing datasets related to mental healthcare. The iForest LSTM baseline achieves the highest recall (0.963) on declaration-level data. On the operation-level data, the hybrid iForest-based model achieves the highest recall (0.744), though at the cost of lower precision. These findings highlight the potential of combining pseudo-labeling with hybrid deep learning in complex, imbalanced anomaly detection settings.
Jiahui An, Sara Irina Fabrikant, Giacomo Indiveri
et al.
Accurately assessing mental workload is crucial in cognitive neuroscience, human-computer interaction, and real-time monitoring, as cognitive load fluctuations affect performance and decision-making. While Electroencephalography (EEG) based machine learning (ML) models can be used to this end, their high computational cost hinders embedded real-time applications. Hardware implementations of spiking neural networks (SNNs) offer a promising alternative for low-power, fast, event-driven processing. This study compares hardware compatible SNN models with various traditional ML ones, using an open-source multimodal dataset. Our results show that multimodal integration improves accuracy, with SNN performance comparable to the ML one, demonstrating their potential for real-time implementations of cognitive load detection. These findings position event-based processing as a promising solution for low-latency, energy efficient workload monitoring in adaptive closed-loop embedded devices that dynamically regulate cognitive load.
Large language models (LLMs) demonstrate proficiency across numerous computational tasks, yet their inner workings remain unclear. In theory, the combination of causal self-attention and multilayer perceptron layers allows every token to access and compute information based on all preceding tokens. In practice, to what extent are such operations present? In this paper, on mental math tasks (i.e., direct math calculation via next-token prediction without explicit reasoning), we investigate this question in three steps: inhibiting input-specific token computations in the initial layers, restricting the routes of information transfer across token positions in the next few layers, and forcing all computation to happen at the last token in the remaining layers. With two proposed techniques, Context-Aware Mean Ablation (CAMA) and Attention-Based Peeking (ABP), we identify an All-for-One subgraph (AF1) with high accuracy on a wide variety of mental math tasks, where meaningful computation occurs very late (in terms of layer depth) and only at the last token, which receives information of other tokens in few specific middle layers. Experiments on a variety of models and arithmetic expressions show that this subgraph is sufficient and necessary for high model performance, transfers across different models, and works on a variety of input styles. Ablations on different CAMA and ABP alternatives reveal their unique advantages over other methods, which may be of independent interest.
Ananya Bhattacharjee, Joseph Jay Williams, Miranda Beltzer
et al.
Challenges in engagement with digital mental health (DMH) tools are commonly addressed through technical enhancements and algorithmic interventions. This paper shifts the focus towards the role of users' broader social context as a significant factor in engagement. Through an eight-week text messaging program aimed at enhancing psychological wellbeing, we recruited 20 participants to help us identify situational engagement disruptors (SEDs), including personal responsibilities, professional obligations, and unexpected health issues. In follow-up design workshops with 25 participants, we explored potential solutions that address such SEDs: prioritizing self-care through structured goal-setting, alternative framings for disengagement, and utilization of external resources. Our findings challenge conventional perspectives on engagement and offer actionable design implications for future DMH tools.
Anxiety, depression, and suicidality are common mental health sequelae following concussion in youth patients, often exacerbating concussion symptoms and prolonging recovery. Despite the critical need for early detection of these mental health symptoms, clinicians often face challenges in accurately collecting patients' mental health data and making clinical decision-making in a timely manner. Today's remote patient monitoring (RPM) technologies offer opportunities to objectively monitor patients' activities, but they were not specifically designed for youth concussion patients; moreover, the large amount of data collected by RPM technologies may also impose significant workloads on clinicians to keep up with and use the data. To address these gaps, we employed a three-stage study consisting of a formative study, interface design, and design evaluation. We first conducted a formative study through semi-structured interviews with six highly professional concussion clinicians and identified clinicians' key challenges in remotely collecting patient information and accessing patient treatment compliance. Subsequently, we proposed preliminary clinician-facing interface designs with the integration of AI-based RPM technologies (AI-RPM), followed by design evaluation sessions with highly professional concussion clinicians. Clinicians underscored the value of integrating multi-modal AI-RPM technologies to support their decision-making while emphasizing the importance of customizable interfaces through collaborative design and multiple responsible design considerations.
Thurayya Zreik, Sandy Chaar, Michelle Lokot
et al.
Inclusive participation of mental health service users is critical for effective decision-making and governance, yet remains underexplored in humanitarian settings. Lebanon, facing protracted crises and hosting over 1.5 million Syrian refugees, provides a unique case to examine pathways of service user participation in mental health decision-making. This qualitative study investigates barriers, facilitators, and power dynamics influencing service user participation at the micro-, meso- (service), and macro- (policy) levels. Semi-structured interviews and focus group discussions were conducted with 33 purposively selected participants, including Syrian and Lebanese service users, NGO staff, and UN representatives. Data were collaboratively analysed using Dedoose software based on codes developed deductively and inductively. Our findings reveal that participation is limited and predominantly consultative, with power imbalances including gender, socioeconomic status, stigma, and displacement status creating significant barriers. Users reported decision-making power at the individual level, particularly in seeking services and treatment planning, but meaningful participation at service or governance levels was rare. Providers highlighted efforts to gather user input but often framed participation as part of routine monitoring and evaluation. Reluctance to engage formal governance structures, due to mistrust and structural and attitudinal barriers, further inhibited participation. Strategies to enhance meaningful participation include increasing awareness, capacity-building, promoting flexibility in service design, and strengthening user-led advocacy. Addressing power imbalances and promoting inclusive, user-centered approaches are essential to advancing inclusion in mental health systems, with valuable implications for humanitarian and crisis-affected settings globally.
Xiao-Xue Chen,1 Jing Chen,1,2 Bao-Liang Zhong1,2 1Research Center for Psychological and Health Sciences, China University of Geosciences (Wuhan), Wuhan, People’s Republic of China; 2Department of Psychiatry, Wuhan Mental Health Center, Wuhan, People’s Republic of ChinaCorrespondence: Jing Chen; Bao-Liang Zhong, Department of Psychiatry, Wuhan Mental Health Center, Wuhan, Hubei Province, People’s Republic of China, Email jingchenphd@21cn.com; haizhilan@gmail.comAbstract: Peer death is not uncommon during adolescence. Unlike the loss of family members, grief following the peer loss is often unrecognized and unsupported by society, making it difficult for adolescents to handle their grief. This may result in prolonged and intense reactions, negatively affecting academic performance and physical and mental health. In this article, we review the manifestations of, associated factors with, and interventions for grief after peer loss and discuss unaddressed questions. A literature search was conducted within three electronic databases (Web of Science, PubMed, and ScienceDirect) from January 1, 2014 to December 29, 2024. Finally, 13 English studies focusing on peer loss and grief among adolescents were identified and included. Grief over the death of a friend in adolescents could be intense, lasting up to 8.5 years, with symptoms such as post-traumatic stress disorder and sleep disturbances. The prolonged grief could impair mental health and social functioning. Girls tend to exhibit more complicated grief reactions and experience a longer duration of grief compared to boys. Personality traits also play a critical role: adolescents with high agreeableness tend to recover more quickly, while those with high neuroticism are less resilient. The quality of the friendship with the deceased and exposure to negative information also influence the grief intensity. Raising professional awareness, providing targeted interventions, and establishing effective social support are essential for grief recovery. Significant gaps still remain in understanding adolescent grief following peer loss, particularly in the mechanisms between different factors and grief, and the feasibility and effectiveness of specific treatment plans. Addressing these limitations is essential for advancing theoretical frameworks and developing targeted interventions. This review provides a foundational basis for future research and clinical practices, with the potential to inform therapeutic approaches and interventions that better support the healing and recovery processes of grieving adolescents.Keywords: grief, peer loss, adolescent, narrative review, associated factor
Lynette R. Goldberg, Kylie Radford, Kate Smith
et al.
Purpose of research: Data show that many Aboriginal and Torres Strait Islander peoples experiencing dementia receive services at mainstream health organizations and from non-Indigenous health care providers. It is imperative that non-Indigenous health care providers are educated about culturally respectful and safe care for Aboriginal and Torres Strait Islander peoples with dementia. The purpose of this research was to partner with Aboriginal Elders to co-design and implement an online unit on culturally respectful and safe care to educate non-Indigenous health care providers. Principal results: Twelve Aboriginal Elders from four Australian states, along with state-based Aboriginal project officers, partnered with the national, interdisciplinary research team to co-create and co-deliver the 13-week unit. Elders formed a Governance Group to guide the research team and ensure the content, delivery and methods of assessment of the unit privileged the spirit, voices, and diverse cultures of Aboriginal and Torres Strait Islander peoples. A team of Aboriginal markers, including Elders and project officers, was established to evaluate students’ assessments. The unit commenced in late July 2024 with 375 students enrolled. Major conclusions: Comments from both Elders and students affirmed the importance of Elders' presence in the unit through their weekly zoom sessions with students and participation in evaluation of students' learnings. Elders’ guidance in the co-creation and co-delivery of the unit has been recognized at program, college and university levels. The unit is available nationally and internationally through the online Diploma of Dementia Care offered by the University of Tasmania, Australia.
Iftitah Saadati, I Gusti Ngurah Antaryama, FX Teddy Badai Samodra
Abstract: Stress causes a person's mental state to be unstable. A contributing factor to stress is emotional exhaustion caused by work routines. Recovery facilities that support the psychological needs of individuals are essential, but existing public facilities still have shortcomings, especially in implementing designs that consider the perspectives of users and stress issues. In the context of psychology, the natural environment is identified as an effective solution to cope with stress. Architecture, as a physical manifestation of the environment, is important in creating positive energy to overcome this. This research aims to evaluate building and environmental design elements integrated with user needs. This research uses a qualitative approach with literature studies and interviews as research methods. Restorative Healing Architecture design principles as a study aspect to produce architecture criteria. The results showed that to achieve emotional stability, natural and healthy environmental conditions are needed that are by user needs. Architectural criteria generated from natural elements can manage and control physiological stress, thereby improving self-quality.Keywords: Architecture, Environment, Psychology, and StressAbstrak: Stres mempengaruhi mental seseorang menjadi tidak stabil. Faktor penyebab stres adalah kelelahan emosianal yang dsebabkan oleh rutinitas pekerjaan. Fasilitas pemulihan yang mendukung kebutuhan psikologis individu menjadi kritis, namun fasilitas publik yang ada masih memiliki kekurangan, terutama dalam penerapan desain yang memperhatikan perspektif pengguna dan permasalahan stres. Dalam konteks psikologi, lingkungan alami diidentifikasi sebagai solusi yang efektif untuk mengatasi stres. Arsitektur, sebagai manifestasi fisik lingkungan, memiliki peran kunci dalam menciptakan energi positif terhadap kendala tersebut. Penelitian ini bertujuan untuk mengevaluasi elemen desain bangunan dan lingkungan yang terintegrasi dengan kebutuhan pengguna. Penelitian ini menggunakan pendekatan kualitatif dengan kajian literatur dan wawancara sebagai metode penelitian. Prinsip desain Arsitektur Pemulih sebagai aspek kajian untuk menghasilkan kriteria Arsitektur. Hasil penelitian menunjukkan bahwa untuk mencapai kestabilan emosional diperlukan kondisi lingkungan yang alami dan sehat yang sesuai bagi penggunanya. Kriteria Arsitektur yang dihasilkan dari elemen alami dapat mengelola dan mengontrol stres fisiologis, sehingga dapat meningkatkan kualitas diri.Kata Kunci: Arsitektur, Lingkungan, Psikologi, dan Stres
Francesca Bianco, Silvia Rigato, Maria Laura Filippetti
et al.
Theory of Mind (ToM), the ability to attribute beliefs, intentions, or mental states to others, is a crucial feature of human social interaction. In complex environments, where the human sensory system reaches its limits, behaviour is strongly driven by our beliefs about the state of the world around us. Accessing others' mental states, e.g., beliefs and intentions, allows for more effective social interactions in natural contexts. Yet, these variables are not directly observable, making understanding ToM a challenging quest of interest for different fields, including psychology, machine learning and robotics. In this paper, we contribute to this topic by showing a developmental synergy between learning to predict low-level mental states (e.g., intentions, goals) and attributing high-level ones (i.e., beliefs). Specifically, we assume that learning beliefs attribution can occur by observing one's own decision processes involving beliefs, e.g., in a partially observable environment. Using a simple feed-forward deep learning model, we show that, when learning to predict others' intentions and actions, more accurate predictions can be acquired earlier if beliefs attribution is learnt simultaneously. Furthermore, we show that the learning performance improves even when observed actors have a different embodiment than the observer and the gain is higher when observing beliefs-driven chunks of behaviour. We propose that our computational approach can inform the understanding of human social cognitive development and be relevant for the design of future adaptive social robots able to autonomously understand, assist, and learn from human interaction partners in novel natural environments and tasks.
With the rise of social media and peer-to-peer networks, users increasingly rely on crowdsourced responses for information and assistance. However, the mechanisms used to rank and promote responses often prioritize and end up biasing in favor of timeliness over quality, which may result in suboptimal support for help-seekers. We analyze millions of responses to mental health-related posts, utilizing large language models (LLMs) to assess the multi-dimensional quality of content, including relevance, empathy, and cultural alignment, among other aspects. Our findings reveal a mismatch between content quality and attention allocation: earlier responses - despite being relatively lower in quality - receive disproportionately high fractions of upvotes and visibility due to platform ranking algorithms. We demonstrate that the quality of the top-ranked responses could be improved by up to 39 percent, and even the simplest re-ranking strategy could significantly improve the quality of top responses, highlighting the need for more nuanced ranking mechanisms that prioritize both timeliness and content quality, especially emotional engagement in online mental health communities.
José Carlos M. Silva, Diogo H. Silva, Francisco A. Rodrigues
et al.
Infections diseases are marked by recovering time distributions which can be far from the exponential one associated with Markovian/Poisson processes, broadly applied in epidemic compartmental models. In the present work, we tackled this problem by investigating a susceptible-infected-recovered-susceptible model on networks with $η$ independent infectious compartments (SI$_η$RS), each one with a Markovian dynamics, that leads to a Gamma-distributed recovering times. We analytically develop a theory for the epidemic lifespan on star graphs with a center and $K$ leaves showing that the epidemic lifespan scales with a non-universal power-law $τ_{K}\sim K^{α/μη}$ plus logarithm corrections, where $α^{-1}$ and $μ^{-1}$ are the mean waning immunity and recovering times, respectively. Compared with standard SIRS dynamics with $η=1$ and the same mean recovering time, the epidemic lifespan on star graphs is severely reduced as the number of stages increases. In particular, the case $η\rightarrow\infty$ leads to a finite lifespan. Numerical simulations support the approximated analytical calculations. For the SIS dynamics, numerical simulations show that the lifespan increases exponentially with the number of leaves, with a nonuniversal rate that decays with the number of infectious compartments. We investigated the SI$_η$RS dynamics on power-law networks with degree distribution $P(K)\sim k^{-γ}$. When $γ<5/2$, the epidemic spreading is ruled by a maximum $k$-core activation, the alteration of the hub activity time does not alter either the epidemic threshold or the localization pattern. For $γ>3$, where hub mutual activation is at work, the localization is reduced but not sufficiently to alter the threshold scaling with the network size. Therefore, the activation mechanisms remain the same as in the case of Markovian healing.
Julia Ive, Paulina Bondaronek, Vishal Yadav
et al.
Introduction: Healthcare AI models often inherit biases from their training data. While efforts have primarily targeted bias in structured data, mental health heavily depends on unstructured data. This study aims to detect and mitigate linguistic differences related to non-biological differences in the training data of AI models designed to assist in pediatric mental health screening. Our objectives are: (1) to assess the presence of bias by evaluating outcome parity across sex subgroups, (2) to identify bias sources through textual distribution analysis, and (3) to develop a de-biasing method for mental health text data. Methods: We examined classification parity across demographic groups and assessed how gendered language influences model predictions. A data-centric de-biasing method was applied, focusing on neutralizing biased terms while retaining salient clinical information. This methodology was tested on a model for automatic anxiety detection in pediatric patients. Results: Our findings revealed a systematic under-diagnosis of female adolescent patients, with a 4% lower accuracy and a 9% higher False Negative Rate (FNR) compared to male patients, likely due to disparities in information density and linguistic differences in patient notes. Notes for male patients were on average 500 words longer, and linguistic similarity metrics indicated distinct word distributions between genders. Implementing our de-biasing approach reduced diagnostic bias by up to 27%, demonstrating its effectiveness in enhancing equity across demographic groups. Discussion: We developed a data-centric de-biasing framework to address gender-based content disparities within clinical text. By neutralizing biased language and enhancing focus on clinically essential information, our approach demonstrates an effective strategy for mitigating bias in AI healthcare models trained on text.
We propose a pipeline for gaining insights into complex diseases by training LLMs on challenging social media text data classification tasks, obtaining explanations for the classification outputs, and performing qualitative and quantitative analysis on the explanations. We report initial results on predicting, explaining, and systematizing the explanations of predicted reports on mental health concerns in people reporting Lyme disease concerns. We report initial results on predicting future ADHD concerns for people reporting anxiety disorder concerns, and demonstrate preliminary results on visualizing the explanations for predicting that a person with anxiety concerns will in the future have ADHD concerns.
Az iskoláskorú gyermekek körében egyre gyakoribb a helytelen testtartás, amely hosszú távon különböző mozgásszervi problémákhoz vezethet. A növekedés intenzív időszakában a gyermekek törzsizomzata gyakran nem elég erős, amit tovább súlyosbít az ülő életmód és a nem megfelelő iskolai környezet. A tartáshibák és gerincbántalmak prevenciója érdekében kulcsfontosságú a korai felismerés és a megfelelő intervenciós programok alkalmazása.
Jelen kutatás célja, hogy felmérje egy célzott tartásjavító program hatását serdülőkorú tanulók gerincszakaszainak változásaira, és vizsgálja a pandémia alatt bekövetkező otthoni inaktivitás hatását. A kutatás során a Spinal Mouse eszközzel vizsgáltuk serdülő korú gyermekek gerincét, valamint az Idiag Posture Score és a Matthiass teszt eredményeit figyelembe véve értékeltük a kapott eredményeket.
Lina-Estelle Linelle Louis, Saïd Moussaoui, Vincent Roualdes
et al.
During an activity, knowing the mental workload (MWL) of the user allows to improve the Human-Machine Interactions (HMI). Indeed, the MWL has an impact on the individual and its interaction with the environment. Monitoring it is therefore a crucial issue. In this context, we have created the virtual game Back to Pizza which is based on the N-back task (commonly used for measuring MWL). In this more playful variant, users must carry out orders from customers of a pizza food truck. It is an interactive game that involves the audience of the IHM'23 conference, choosing several parameters like the number of ingredients. During this experience, the objective is to measure MWL in real time through an ElectroEncephaloGraph (EEG) and visual feedback on MWL level is given to the audience. With this demonstration, we propose to present a concept of a virtual interactive game that measures MWL in real time.