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

TL;DR

This paper introduces a privacy-preserving, on-device AI platform for psychiatric decision support that operates entirely locally on mobile devices, ensuring data privacy without sacrificing diagnostic accuracy or speed.

Contribution

It presents a novel zero-egress, fully local inference pipeline for psychiatric AI, integrating lightweight LLMs on mobile hardware to enhance privacy in sensitive healthcare settings.

Findings

Achieves diagnostic accuracy comparable to server-based systems.

Maintains real-time inference latency on mobile devices.

Ensures no patient data leaves the device during processing.

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.