The application of AI in psychiatric diagnosis faces significant challenges, including the subjective nature of mental health assessments, symptom overlap across disorders, and privacy constraints limiting data availability. To address these issues, we present MoodAngels, the first specialized multi-agent framework for mood disorder diagnosis. Our approach combines granular-scale analysis of clinical assessments with a structured verification process, enabling more accurate interpretation of complex psychiatric data. Complementing this framework, we introduce MoodSyn, an open-source dataset of 1,173 synthetic psychiatric cases that preserves clinical validity while ensuring patient privacy. Experimental results demonstrate that MoodAngels outperforms conventional methods, with our baseline agent achieving 12.3% higher accuracy than GPT-4o on real-world cases, and our full multi-agent system delivering further improvements. Evaluation in the MoodSyn dataset demonstrates exceptional fidelity, accurately reproducing both the core statistical patterns and complex relationships present in the original data while maintaining strong utility for machine learning applications. Together, these contributions provide both an advanced diagnostic tool and a critical research resource for computational psychiatry, bridging important gaps in AI-assisted mental health assessment.

Large Language Models (LLMs) have been shown to demonstrate imbalanced biases against certain groups. However, the study of unprovoked targeted attacks by LLMs towards at-risk populations remains underexplored. Our paper presents three novel contributions: (1) the explicit evaluation of LLM-generated attacks on highly vulnerable mental health groups; (2) a network-based framework to study the propagation of relative biases; and (3) an assessment of the relative degree of stigmatization that emerges from these attacks. Our analysis of a recently released large-scale bias audit dataset reveals that mental health entities occupy central positions within attack narrative networks, as revealed by a significantly higher mean centrality of closeness (p-value = 4.06e-10) and dense clustering (Gini coefficient = 0.7). Drawing from sociological foundations of stigmatization theory, our stigmatization analysis indicates increased labeling components for mental health disorder-related targets relative to initial targets in generation chains. Taken together, these insights shed light on the structural predilections of large language models to heighten harmful discourse and highlight the need for suitable approaches for mitigation.

In the clinical treatment of mood disorders, the complex behavioral symptoms presented by patients and variability of patient response to particular medication classes can create difficulties in providing fast and reliable treatment when standard diagnostic and prescription methods are used. Increasingly, the incorporation of physiological information such as neuroimaging scans and derivatives into the clinical process promises to alleviate some of the uncertainty surrounding this process. Particularly, if neural features can help to identify patients who may not respond to standard courses of anti-depressants or mood stabilizers, clinicians may elect to avoid lengthy and side-effect-laden treatments and seek out a different, more effective course that might otherwise not have been under consideration. Previously, approaches for the derivation of relevant neuroimaging features work at only one scale in the data, potentially limiting the depth of information available for clinical decision support. In this work, we show that the utilization of multi spatial scale neuroimaging features - particularly resting state functional networks and functional network connectivity measures - provide a rich and robust basis for the identification of relevant medication class and non-responders in the treatment of mood disorders. We demonstrate that the generated features, along with a novel approach for fast and automated feature selection, can support high accuracy rates in the identification of medication class and non-responders as well as the identification of novel, multi-scale biomarkers.

This research project aims to tackle the growing mental health challenges in today's digital age. It employs a modified pre-trained BERT model to detect depressive text within social media and users' web browsing data, achieving an impressive 93% test accuracy. Simultaneously, the project aims to incorporate physiological signals from wearable devices, such as smartwatches and EEG sensors, to provide long-term tracking and prognosis of mood disorders and emotional states. This comprehensive approach holds promise for enhancing early detection of depression and advancing overall mental health outcomes.

Personal sensing, leveraging data passively and near-continuously collected with wearables from patients in their ecological environment, is a promising paradigm to monitor mood disorders (MDs), a major determinant of worldwide disease burden. However, collecting and annotating wearable data is very resource-intensive. Studies of this kind can thus typically afford to recruit only a couple dozens of patients. This constitutes one of the major obstacles to applying modern supervised machine learning techniques to MDs detection. In this paper, we overcome this data bottleneck and advance the detection of MDs acute episode vs stable state from wearables data on the back of recent advances in self-supervised learning (SSL). This leverages unlabelled data to learn representations during pre-training, subsequently exploited for a supervised task. First, we collected open-access datasets recording with an Empatica E4 spanning different, unrelated to MD monitoring, personal sensing tasks -- from emotion recognition in Super Mario players to stress detection in undergraduates -- and devised a pre-processing pipeline performing on-/off-body detection, sleep-wake detection, segmentation, and (optionally) feature extraction. With 161 E4-recorded subjects, we introduce E4SelfLearning, the largest to date open access collection, and its pre-processing pipeline. Second, we show that SSL confidently outperforms fully-supervised pipelines using either our novel E4-tailored Transformer architecture (E4mer) or classical baseline XGBoost: 81.23% against 75.35% (E4mer) and 72.02% (XGBoost) correctly classified recording segments from 64 (half acute, half stable) patients. Lastly, we illustrate that SSL performance is strongly associated with the specific surrogate task employed for pre-training as well as with unlabelled data availability.

Large language models, particularly GPT-3, are able to produce high quality summaries of general domain news articles in few- and zero-shot settings. However, it is unclear if such models are similarly capable in more specialized, high-stakes domains such as biomedicine. In this paper, we enlist domain experts (individuals with medical training) to evaluate summaries of biomedical articles generated by GPT-3, given zero supervision. We consider both single- and multi-document settings. In the former, GPT-3 is tasked with generating regular and plain-language summaries of articles describing randomized controlled trials; in the latter, we assess the degree to which GPT-3 is able to \emph{synthesize} evidence reported across a collection of articles. We design an annotation scheme for evaluating model outputs, with an emphasis on assessing the factual accuracy of generated summaries. We find that while GPT-3 is able to summarize and simplify single biomedical articles faithfully, it struggles to provide accurate aggregations of findings over multiple documents. We release all data and annotations used in this work.

In my experience, one of the biggest opportunities for AI in the field of neuroscience is doing more biologically plausible research and creating a visual perception of the unique features of human cognition. That will lead us in understanding and knowing so much more about the human brain which still is a mystery to a huge extent. This also eventually will lead us to understand mental illnesses, triggers to mental illnesses and the best way to recover from them. Biologically-inspired artificial neural networks and computational neuroscience approaches that attempt to elucidate brain networks are gradually getting closer to experimental neuroscience. However this would definitely require a lot more research and more conversations and collaborations between neuroscientists and AI researchers.

Objectives Psychological distress is a worldwide problem and a serious problem that needs to be addressed in the field of occupational health. This study aimed to use artificial intelligence (AI) to predict psychological distress among workers using sociodemographic, lifestyle and sleep factors, not subjective information such as mood and emotion, and to examine the performance of the AI models through a comparison with psychiatrists. Design Cross-sectional study. Setting We conducted a survey on psychological distress and living conditions among workers. An AI model for predicting psychological distress was created and then the results were compared in terms of accuracy with predictions made by psychiatrists. Participants An AI model of the neural network and six psychiatrists. Primary outcome The accuracies of the AI model and psychiatrists for predicting psychological distress. Methods In total, data from 7251 workers were analysed to predict moderate and severe psychological distress. An AI model of the neural network was created and accuracy, sensitivity and specificity were calculated. Six psychiatrists used the same data as the AI model to predict psychological distress and conduct a comparison with the AI model. Results The accuracies of the AI model and psychiatrists for predicting moderate psychological distress were 65.2% and 64.4%, respectively, showing no significant difference. The accuracies of the AI model and psychiatrists for predicting severe psychological distress were 89.9% and 85.5%, respectively, indicating that the AI model had significantly higher accuracy. Conclusions A machine learning model was successfully developed to screen workers with depressed mood. The explanatory variables used for the predictions did not directly ask about mood. Therefore, this newly developed model appears to be able to predict psychological distress among workers easily, regardless of their subjective views. No data are available.

Summary: New AI technology can detect a patient's stroke depression type, and improve treatment options. An AI developed by Japanese researchers might soon help stroke survivors get the right treatment by detecting a patient's post-stroke depression (PSD) type, a frequently seen but often overlooked neuropsychiatric manifestation after a stroke that could impair functional recovery. The AI was developed by Hiroshima University (HU) researchers using a probabilistic artificial neural network called log-linearized Gaussian mixture network. The neural network was trained to distinguish between depression, apathy, or anxiety based on 36 evaluation indices obtained from functional, physical, and cognitive tests on 274 patients. Details about their research that analyzed the relationship between PSD and activities of daily living independence, degree of paralysis, stress awareness, and higher brain function using machine learning are published in Scientific Reports.

However, what if we add to this and take a more holistic approach to health, describing it as more than just the "absence of illness?" Wellness is a somewhat elusive concept, defined by the WHO as "a state of complete physical, mental, and social well-being, and not merely the absence of disease or infirmity." Can machine learning contribute to our wellness? Health does feature prominently in the overall wellness of a person for the simple reason that being free of illness is the main prerequisite. In other words, suffering from a disease of some kind will trump everything else you do for your wellness. There is no doubt that machine learning has established its role in healthcare with its capabilities.