Toward Zero-Egress Psychiatric AI: On-Device LLM Deployment for Privacy-Preserving Mental Health Decision Support
Old Dominion University(旧 Dominion 大学);AI Motion Labs(AI Motion 实验室);Department of Psychiatry and Neurobehavioral Sciences(精神病学与神经行为科学系);University of Virginia School of Medicine(弗吉尼亚大学医学院);Florida International University(佛罗里达国际大学);Blanchfield Army Community Hospital(Blanchfield陆军社区医院);McDonald Army Health Center(McDonald陆军医疗中心)
Eranga Bandara, Asanga Gunaratna, Ross Gore, Anita H. Clayton, Christopher K. Rhea, Sachini Rajapakse, Isurunima Kularathna, Sachin Shetty, Ravi Mukkamala, Xuep… 展开作者
Eranga Bandara, Asanga Gunaratna, Ross Gore, Anita H. Clayton, Christopher K. Rhea, Sachini Rajapakse, Isurunima Kularathna, Sachin Shetty, Ravi Mukkamala, Xueping Liang, Preston Samuel, Atmaram Yarlagadda
Privacy represents one of the most critical yet underaddressed barriers to AI adoption in mental healthcare -- particularly in high-sensitivity operational environments such as military, correctional, and remote healthcare settings, where the risk of patient data exposure can deter help-seeking behavior entirely. Existing AI-enabled psychiatric decision support systems predominantly rely on cloud-based inference pipelines, requiring sensitive patient data to leave the device and traverse external servers, creating unacceptable privacy and security risks in these contexts. In this paper, we propose a zero-egress, on-device AI platform for privacy-preserving psychiatric decision support, deployed as a cross-platform mobile application. The proposed system extends our prior work on fine-tuned LLM consortiums for psychiatric diagnosis standardization by fundamentally re-architecting the inference pipeline for fully local execution -- ensuring that no patient data is transmitted to, processed by, or stored on any external server at any stage. The platform integrates a consortium of three lightweight, fine-tuned, and quantized open-source LLMs -- Gemma, Phi-3.5-mini, and Qwen2 -- selected for their compact architectures and proven efficiency on resource-constrained mobile hardware. An on-device orchestration layer coordinates ensemble inference and consensus-based diagnostic reasoning, producing DSM-5-aligned assessments for conditions. The platform is designed to assist clinicians with differential diagnosis and evidence-linked symptom mapping, as well as to support patient-facing self-screening with appropriate clinical safeguards. Initial evaluation demonstrates that the proposed zero-egress deployment achieves diagnostic accuracy comparable to its server-side predecessor while sustaining real-time inference latency on commodity mobile hardware.
MentalSeek-Dx: Towards Progressive Hypothetico-Deductive Reasoning for Real-world Psychiatric Diagnosis
School of Computer Science Chongqing University(重庆大学计算机学院);MAIS Institute of Automation Chinese Academy of Sciences(中国科学院自动化研究所MAIS研究所);The First Affiliated Hospital of Chongqing Medical University(重庆医科大学第一附属医院)
Unveiling the Landscape of Clinical Depression Assessment: From Behavioral Signatures to Psychiatric Reasoning
School of Computer Science and Engineering, Central South University(中南大学计算机科学与工程学院);CoAI Group, DCST, IAI, BNRIST, Tsinghua University(清华大学);University of International Relations(国际关系大学);Central China Normal University(华中师范大学);Lingxin AI(灵心AI);Yuquan Hospital, Tsinghua University(清华大学友谊医院)
Lived Experience Not Found: LLMs Struggle to Align with Experts on Addressing Adverse Drug Reactions from Psychiatric Medication Use
College of Computing, Georgia Institute of Technology(佐治亚理工学院计算学院);Hospital General Universitario Gregorio Marañón(格雷戈里奥·马拉尼翁大学综合医院);Hofstra University(霍夫斯特拉大学);Columbia University(哥伦比亚大学)
Automated ICD Classification of Psychiatric Diagnoses: From Classical NLP to Large Language Models
Department of Sistemas Informáticos, Universidad Politécnica de Madrid, Spain(西班牙马德里理工大学信息系统系);KNODIS Research Group, Universidad Politécnica de Madrid, Spain(西班牙马德里理工大学KNODIS研究组);CIBERSAM ISCIII, Spain(西班牙ISCIII CIBERSAM);Department of Legal Medicine, Psychiatry and Pathology. Complutense University of Madrid, Spain(西班牙马德里康普顿斯大学法医学、精神病学与病理学系);Hospital Universitario de Móstoles, Universidad Rey Juan Carlos, Spain(西班牙雷阿尔皇家卡洛斯大学莫斯特oles大学医院);Department of Psychiatry, University Hospital Jimenez Díaz Fundation, Madrid, Spain(西班牙圣地亚哥· jiménez Díaz基金会精神病科部);Department of Psychiatry, University Hospital Rey Juan Carlos, Móstoles, Spain(西班牙雷阿尔皇家卡洛斯大学莫斯特oles医院精神病科部);Department of Psychiatry, General Hospital of Villalba, Madrid, Spain(西班牙维拉尔巴医院精神病科部);Department of Psychiatry, University Hospital Infanta Elena, Madrid, Spain(西班牙伊菲格尼亚医院精神病科部);Department of Psychology, Universidad Catolica del Maule, Talca, Chile(智利马尔学院心理学系);Department of Psychiatry, Madrid Autonomous University, Madrid, Spain(西班牙马德里自治大学精神病科部)
Single-turn emergency psychiatric triage across 15 frontier AI chatbots
Max Planck UCL Centre for Computational Psychiatry and Ageing Research(马克斯·普朗克大学伦敦计算精神病学与衰老研究中心);UK AI Security Institute(英国人工智能安全研究所);Department of Experimental Psychology, University of Oxford(牛津大学实验心理学系);Microsoft AI(微软人工智能);Department of Psychiatry, University of Oxford(牛津大学精神病学系)
Patterns vs. Patients: Evaluating LLMs against Mental Health Professionals on Personality Disorder Diagnosis through First-Person Narratives
IDEAS Research Institute(IDEAS研究 institute);Adam Mickiewicz University(亚当·密茨凯维奇大学);AMU Center for Artificial Intelligence(AMU人工智能中心);Poznań University of Medical Sciences(波兹南医学科学大学);Maria Curie-Skłodowska University(玛丽·居里-斯克洛多夫斯卡大学)
LingxiDiagBench: A Multi-Agent Framework for Benchmarking LLMs in Chinese Psychiatric Consultation and Diagnosis
Tianqiao and Chrissy Chen Institute(天桥和克里斯西·陈研究所);EverMind AI Inc.(EverMind AI公司);Shanghai Mental Health Center, Shanghai Jiao Tong University School of Medicine(上海精神卫生中心,上海交通大学医学院)
WiseMind: a knowledge-guided multi-agent framework for accurate and empathetic psychiatric diagnosis
College of Biomedical Engineering(生物医学工程学院);Fudan University(复旦大学);Department of Electrical and Computer Engineering(电气与计算机工程系);University of Alberta(阿尔伯塔大学);University of Virginia(弗吉尼亚大学);McIntire School of Commerce(麦金尼商学院);School of Data Science(数据科学学院);Department of Psychiatry(精神病学系)
Multimodal Machine Learning in Mental Health: A Survey of Data, Algorithms, and Challenges
Queen Mary University of London United Kingdom;Queen Mary University of London \& Jo z ef Stefan Institute United Kingdom \& Slovenia;Queen Mary University of London;Queen Mary University of London \& Jo z ef Stefan Institute
Knowledge-augmented Agentic AI for Mental Health Medication Information Seeking
Department of Medicine and Therapeutics, The Chinese University of Hong Kong(香港中文大学内科及药物治疗学系);Department of Psychiatry, The Chinese University of Hong Kong(香港中文大学精神科学系);School of Control Science and Engineering, Shandong University(山东大学控制科学与工程学院);College of Information Science, University of Arizona(亚利桑那大学信息科学学院);Department of Medicine, Stanford University School of Medicine, Stanford University(斯坦福大学医学院医学系);Perelman School of Medicine, The University of Pennsylvania(宾夕法尼亚大学佩雷尔曼医学院);Department of Biostatistics, Vanderbilt University(范德堡大学生物统计学系);Li Ka Shing Institute of Health Sciences, Faculty of Medicine, The Chinese University of Hong Kong(香港中文大学医学院李嘉诚健康科学研究所)
MIND: Unified Inquiry and Diagnosis RL with Criteria Grounded Clinical Supports for Psychiatric Consultation
EverMind AI Inc.(EverMind AI公司);EverMind AI Inc., Tianqiao and Chrissy Chen Institute(EverMind AI公司、田桥与克里斯西·陈研究所);Shanghai Mental Health Center, Shanghai Jiao Tong University School of Medicine(上海精神卫生中心、上海交通大学医学院)
MoodAngels: A Retrieval-augmented Multi-agent Framework for Psychiatry Diagnosis
School of Artificial Intelligence, Wuhan University(武汉大学人工智能学院);Center for Language and Information Research, Wuhan University(武汉大学语言与信息研究所);School of Computer Science, Wuhan University(武汉大学计算机学院);Department of Psychiatry, Renmin Hospital of Wuhan University(武汉大学人民医院精神科);The Fin AI(Fin AI);Taikang Center for Life and Medical Sciences, Wuhan University(武汉大学泰康生命与医学科学中心)
A Safety-Gated Multimodal AI Backend for Mental-Health Support: Hierarchical State Representation, Conservative Risk Fusion, and Controlled Generation in Anian
EmoAgent: Assessing and Safeguarding Human-AI Interaction for Mental Health Safety
Department of Electrical & Computer Engineering, Princeton University(普林斯顿大学电气与计算机工程系);Department of Computer Science, Princeton University(普林斯顿大学计算机科学系);Department of Computer Science & Engineering, University of Michigan(密歇根大学计算机科学与工程系);Department of Data Science & Engineering, University of Michigan(密歇根大学数据科学与工程系);Department of Philosophy, Columbia University(哥伦比亚大学哲学系);AI Lab, Princeton University(普林斯顿大学人工智能实验室);Chen Frontier Lab for Al and Mental Health, Tianqiao and Chrissy Chen Institute(天桥及克里斯西·陈研究所人工智能与心理健康前沿实验室);Theta Health Inc.(Theta健康公司)
A Foundation Model for Wearable Movement Data in Mental Health Research
Dartmouth College(达特茅斯学院);National Institute of Diabetes and Digestive and Kidney Diseases(国家糖尿病、消化系统疾病和肾病研究所);National Institutes of Health(美国国立卫生研究院);Department of Computer Science at Dartmouth College(达特茅斯学院计算机科学系)