Abstract
Healthcare systems with limited resources face critical shortages of imaging capabilities, which compound diagnostic disparities. While artificial intelligence is a potential solution, the data and technology required to implement typical AI models make them impractical in many resource-limited medical facilities. This paper describes the theoretical framework for the resource-aware, agentically frugal intelligent (RAFI) development of medical imaging Artificial Intelligence (AI), with the goal of developing efficient, independent, and just medical imaging AI systems in low-resource settings. RAFI combines frugal machine learning techniques, agentic decision-making, and trustworthy concepts to help overcome data scarcity, computational limitations, and bias. We synthesise the literature and identify the main gaps. The RAFI framework includes principles (e.g., resource prioritisation, fairness by design, and continuous local adaptation) and sets objectives informed by resource and ethical constraints. A comparison between RAFI and current paradigms highlights differences in scalability, equity, and sustainability. We provide four hypotheses (P1-P4) for future research; we present a roadmap for the implementation and validation of RAFI; and we discuss the implications for policy and capacity building. The development of RAFI provides a means to build a scalable, trustworthy AI system for equitable global health.
Links to Paper and Supplementary Materials
Main Paper (Open Access Version): https://papers.miccai.org/miccai-2026-sat/paper/MIRASOL_022.pdf
SharedIt Link: Not yet available
SpringerLink (DOI): Not yet available
Supplementary Material: Not Submitted
Link to Open Review
Open Review Page: https://openreview.net/forum?id=8aW1AkLYMt
BibTex
@InProceedings{OguRos_Frugal_MICCAISAT2026,
author = { Ogundokun, Roseline Oluwaseun AND Bello, Rotimi-Williams AND Owolawi, Pius Adewale},
title = { { Frugal and Agentic AI for Medical Imaging in Resource-Constrained Settings: A Sustainable Framework } },
booktitle = {Medical Image Computing and Computer Assisted Intervention -- MICCAI 2026 Workshops and Challenges},
year = {2026},
publisher = {Springer Nature Switzerland},
volume = {LNCS 17264},
month = {pending},
page = {pending}
}