[Paper Review] Empowering Medical Equipment Sustainability in Low-Resource Settings: An AI-Powered Diagnostic and Support Platform for Biomedical Technicians

Bernes Lorier Atabonfack, Ahmed Tahiru Issah+6 more|arXiv (Cornell University)|Jan 23, 2026

Quality and Safety in Healthcare0 citations

TL;DR

The paper presents INGENZI Tech, an AI-powered, offline-capable diagnostic and support platform for biomedical technicians in LMICs, validated on the Philips HDI 5000 ultrasound with 100% error-code interpretation accuracy and 80% actionable-troubleshooting accuracy.

ABSTRACT

In low- and middle-income countries (LMICs), a significant proportion of medical diagnostic equipment remains underutilized or non-functional due to a lack of timely maintenance, limited access to technical expertise, and minimal support from manufacturers, particularly for devices acquired through third-party vendors or donations. This challenge contributes to increased equipment downtime, delayed diagnoses, and compromised patient care. This research explores the development and validation of an AI-powered support platform designed to assist biomedical technicians in diagnosing and repairing medical devices in real-time. The system integrates a large language model (LLM) with a user-friendly web interface, enabling imaging technologists/radiographers and biomedical technicians to input error codes or device symptoms and receive accurate, step-by-step troubleshooting guidance. The platform also includes a global peer-to-peer discussion forum to support knowledge exchange and provide additional context for rare or undocumented issues. A proof of concept was developed using the Philips HDI 5000 ultrasound machine, achieving 100% precision in error code interpretation and 80% accuracy in suggesting corrective actions. This study demonstrates the feasibility and potential of AI-driven systems to support medical device maintenance, with the aim of reducing equipment downtime to improve healthcare delivery in resource-constrained environments.

Motivation & Objective

• Address the maintenance gap for medical equipment in LMICs caused by limited technician expertise and scarce manufacturer support.
• Develop an AI-assisted platform that provides step-by-step troubleshooting using error codes, symptoms, and manuals.
• Enable offline, multilingual access and peer-to-peer knowledge sharing to improve device uptime.
• Pilot the concept on a common imaging device (Philips HDI 5000) with a plan to scale to MRI/CT/X-ray devices.

Proposed method

• Build a Retrieval-Augmented Generation (RAG) framework with an LLM (GPT-3.5 Turbo) and segmented FAISS vector stores for user manuals, service manuals, and error codes.
• Provide a multilingual, offline-capable web interface for inputting error codes or device symptoms and receiving guided troubleshooting.
• Integrate tools for error-code lookup, log parsing, self-test simulation, and maintenance scheduling to support actionable repair workflows.
• Incorporate a peer-to-peer technician forum to enable knowledge exchange and crowd-sourced model improvement.
• Evaluate phase-0 performance with a proof-of-concept on the Philips HDI 5000 ultrasound, measuring error-code retrieval precision and instructional guidance accuracy.

Experimental results

Research questions

• RQ1 How accurately can the AI platform interpret device error codes and fetch relevant documented guidance?
• RQ2 Can the system generate usable, step-by-step repair guidance from unstructured instructional queries?
• RQ3 Does offline, multilingual deployment improve accessibility and technician usability in LMIC settings?
• RQ4 What is the feasibility and pathway for scaling from ultrasound to other diagnostic devices (MRI, CT, X-ray) across vendors?

Key findings

• Error-code interpretation achieved 100% precision on 90 tested codes.
• Instructional query guidance achieved 80% accuracy across 30 queries.
• Phase-0 prototype maintained sub-10-second latency in retrieval-and-generation cycles.
• Offline, multilingual, and forum-enabled design supports LMIC contexts and continuous model improvement.
• Phase-0 validates the feasibility of an LMIC-focused, AI-assisted maintenance platform and sets the stage for multi-device expansion.