Toward Zero-Egress Psychiatric AI: On-Device LLM Deployment for Privacy-Preserving Mental Health Decision Support

• URL: http://arxiv.org/abs/2604.18302v1
• Date: Mon, 20 Apr 2026 14:09:01 GMT
• Title: Toward Zero-Egress Psychiatric AI: On-Device LLM Deployment for Privacy-Preserving Mental Health Decision Support
• Abstract summary: Privacy represents one of the most critical yet under barriers to AI adoption in mental healthcare.<n>Existing AI-enabled psychiatric decision support systems rely on cloud-based inference pipelines.<n>We propose a zero-egress, on-device AI platform for privacy-preserving psychiatric decision support.
• Score: 3.560737321459875
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: 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.

Related papers

• MedCTA: A Benchmark for Clinical Tool Agents [47.47354499871572]
  We introduce MedCTA, a benchmark for evaluating medical tool agents on clinician-validated, step-implicit tasks.<n>MedCTA comprises 107 real-world clinical tasks with clinician-verified executable trajectories over 5 deployed tools.<n>We benchmark 18 open- and closed-source multimodal models and find that even frontier systems remain brittle in multi-step clinical tool use.
  arXiv  Detail & Related papers  (2026-06-10T06:26:52Z)
• An Agentic LLM-Based Framework for Population-Scale Mental Health Screening [0.3568466510804538]
  This paper proposes an agentic framework for building robust LLM-based pipelines.<n>A proof-of-concept in transcript-based detection demonstrates that the framework converges to stable configurations.<n>Results highlight the potential of agentic AI to enable population-level mental health screening over large clinical datasets.
  arXiv  Detail & Related papers  (2026-05-13T06:08:43Z)
• Key-Embedded Privacy for Decentralized AI in Biomedical Omics [43.072683741556126]
  We present a lightweight federated learning method, INFL, based on Implicit Neural Representations.<n>INFL achieves strong, controllable privacy while maintaining utility, preserving utility for downstream scientific and clinical applications.
  arXiv  Detail & Related papers  (2026-03-30T12:04:50Z)
• Federated Learning for Privacy-Preserving Medical AI [0.0]
  This dissertation investigates privacy-preserving federated learning for Alzheimer's disease classification using three-dimensional MRI data.<n>Existing methodologies often suffer from unrealistic data partitioning, inadequate privacy guarantees, and insufficient benchmarking.<n>To address these gaps, this research proposes a novel site-aware data partitioning strategy that preserves institutional boundaries.
  arXiv  Detail & Related papers  (2026-03-16T20:46:21Z)
• PrivacyBench: Privacy Isn't Free in Hybrid Privacy-Preserving Vision Systems [1.31032910112552]
  We introduce PrivacyBench, a benchmarking framework that reveals striking failures in privacy technique combinations with severe deployment implications.<n>We show that FL + DP combinations exhibit severe convergence failure, with accuracy dropping from 98% to 13% while compute costs and energy consumption substantially increase.<n>Our framework provides the first systematic platform for evaluating privacy-utility-cost trade-offs through automated YAML configuration, resource monitoring, and reproducible protocols.
  arXiv  Detail & Related papers  (2026-02-21T16:45:56Z)
• SoK: Privacy-aware LLM in Healthcare: Threat Model, Privacy Techniques, Challenges and Recommendations [0.6533091401094101]
  Large Language Models (LLMs) are increasingly adopted in healthcare to support clinical decision-making and enhance patient care.<n>This work examines the evolving threat landscape across the three core LLM phases: Data preprocessing, Fine-tuning, and Inference within realistic healthcare settings.<n>We present a detailed threat model that characterizes adversaries, capabilities, and attack surfaces at each phase, and we systematize how existing privacy-preserving techniques (PPTs) attempt to mitigate these vulnerabilities.
  arXiv  Detail & Related papers  (2026-01-15T02:28:57Z)
• Toward an AI Reasoning-Enabled System for Patient-Clinical Trial Matching [0.0]
  We present a secure, scalable proof-of-concept system for Artificial Intelligence (AI)-augmented patient-trial matching.<n>The system generates structured eligibility assessments with interpretable reasoning chains that support human-in-the-loop review.<n>The system aims to reduce coordinator burden, intelligently broaden the set of trials considered for each patient and guarantee comprehensive auditability of all AI-generated outputs.
  arXiv  Detail & Related papers  (2025-12-08T20:35:51Z)
• The Locally Deployable Virtual Doctor: LLM Based Human Interface for Automated Anamnesis and Database Conversion [0.0]
  MedChat is a locally deployable virtual physician framework for AI-assisted clinical anamnesis.<n>Unlike existing cloud-based systems, this work demonstrates the feasibility of a fully offline, locally deployable LLM-diffusion framework for clinical anamnesis.
  arXiv  Detail & Related papers  (2025-11-23T22:12:35Z)
• Timely Clinical Diagnosis through Active Test Selection [49.091903570068155]
  We propose ACTMED (Adaptive Clinical Test selection via Model-based Experimental Design) to better emulate real-world diagnostic reasoning.<n>LLMs act as flexible simulators, generating plausible patient state distributions and supporting belief updates without requiring structured, task-specific training data.<n>We evaluate ACTMED on real-world datasets and show it can optimize test selection to improve diagnostic accuracy, interpretability, and resource use.
  arXiv  Detail & Related papers  (2025-10-21T18:10:45Z)
• MoodAngels: A Retrieval-augmented Multi-agent Framework for Psychiatry Diagnosis [58.67342568632529]
  MoodAngels is the first specialized multi-agent framework for mood disorder diagnosis.<n>MoodSyn is an open-source dataset of 1,173 synthetic psychiatric cases.
  arXiv  Detail & Related papers  (2025-06-04T09:18:25Z)
• Towards Privacy-aware Mental Health AI Models: Advances, Challenges, and Opportunities [58.61680631581921]
  Mental health disorders create profound personal and societal burdens, yet conventional diagnostics are resource-intensive and limit accessibility.<n>This paper examines these challenges and proposes solutions, including anonymization, synthetic data, and privacy-preserving training.<n>It aims to advance reliable, privacy-aware AI tools that support clinical decision-making and improve mental health outcomes.
  arXiv  Detail & Related papers  (2025-02-01T15:10:02Z)
• Optimal discharge of patients from intensive care via a data-driven policy learning framework [58.720142291102135]
  It is important that the patient discharge task addresses the nuanced trade-off between decreasing a patient's length of stay and the risk of readmission or even death following the discharge decision. This work introduces an end-to-end general framework for capturing this trade-off to recommend optimal discharge timing decisions. A data-driven approach is used to derive a parsimonious, discrete state space representation that captures a patient's physiological condition.
  arXiv  Detail & Related papers  (2021-12-17T04:39:33Z)
• MedPerf: Open Benchmarking Platform for Medical Artificial Intelligence using Federated Evaluation [110.31526448744096]
  We argue that unlocking this potential requires a systematic way to measure the performance of medical AI models on large-scale heterogeneous data. We are building MedPerf, an open framework for benchmarking machine learning in the medical domain.
  arXiv  Detail & Related papers  (2021-09-29T18:09:41Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.