OPTIMIZING DEEP LEARNING-ENHANCED OPPORTUNISTIC OSTEOPOROSIS SCREENING FROM CHEST X-RAYS IN AUSTRALIA: IMPACT OF FOLLOW-UP AND TREATMENT ADHERENCE ON CLINICAL AND ECONOMIC OUTCOMES

J.-Y. Reginster 1, P. R. Ebeling 2, M. Alokail1, J. H. Jung 3, M. Hiligsmann 4

1King Saud University, Riyadh, Saudi Arabia,2Monash University, Clayton, Victoria, Australia,

3Promedius Inc. and Chung-Ang University, Seoul, South Korea,4Maastricht University, Maastricht, Netherlands

Introduction: Opportunistic osteoporosis screening using artificial intelligence (AI) on routine

chest radiographs offers a low-burden way to improve detection and treatment uptake. This

study evaluated the cost-effectiveness of AI-enhanced screening versus no screening in

Australian women aged ≥50 years, focusing on health outcomes and the added value of

improved follow-up and adherence.

Methods: A microsimulation-based Markov model estimated lifetime costs and quality-

adjusted life years (QALYs) from the Australian healthcare perspective, expressed in 2025

Australian dollars. The model incorporated AI screening performance on chest radiographs,

rates of DXA confirmation, treatment initiation, and real-world medication persistence.

Scenario analyses evaluated the potential effect of optimized follow-up pathways and

enhanced adherence on clinical and economic outcomes.

Results: AI-enhanced opportunistic screening was associated with better health outcomes

(+0.0025 QALYs and –0.0043 fractures per patient) and reduced lifetime costs (–AUD59),

making it a dominant strategy compared with no screening. When scaled to 10,000 screened

women aged ≥50 years, the model projected prevention of 43 fractures, 25 additional life-

years, 25 QALYs gained, and healthcare savings of AUD590,000. Enhanced follow-up and

adherence produced four- to five-fold increases in clinical benefit relative to the base-case

analysis.

Discussion: AI-enhanced osteoporosis screening using routine chest X-rays appears cost-

effective and potentially cost-saving within the Australian healthcare system. Linking AI

detection to timely DXA, treatment initiation, and sustained adherence substantially amplifies

its clinical value. These findings highlight that follow-up and persistence are not ancillary but

essential components for realizing the full potential of AI-driven opportunistic screening

programs.

Acknowledgement: Funded by an unrestricted educational grant from Promedius, Inc.