Geographic convergence masks institutional stratification in the scale-up of medical AI in China

Ensuring the equitable deployment of artificial intelligence (AI) in healthcare represents a critical bottleneck following algorithm validation. Yet empirical evidence on the real-world allocation of capital-intensive AI diagnostic devices remains scarce. In this paper, we used the real-time public procurement data to track the spatial and institutional diffusion of AI-enabled diagnostic and screening devices (AIDSD) across Chinese public hospitals from 2020 to 2024. Using dual-dimensional Theil-T indices, spatial agglomeration metrics, and structural location quotients, we identify two coexisting but contradictory trends. At the macro-geographic level, national device volume expanded 37.25-fold—from 8 units to 298—driving sustained convergence in procurement-based spatial equity indices (T pop : from 0.425 to 0.047; T hosp : from 0.591 to 0.070), with zero-adoption regions fully eliminated by 2022. Beneath this encouraging surface, however, severe institutional stratification persisted. Even when restricting analysis strictly to hospitals with capital acquisition capacity, tertiary hospitals disproportionately accumulated AIDSD throughout (LQ tier3 = 2.154 in 2024), while secondary hospitals experienced persistent relative deprivation (LQ tier2: from 0.361 to 0.320). A nested decomposition of total inequality further revealed that the within-region, tier-based component rose from 47.3% to 84.5% of total inequality between 2020 and 2024, consistent with geographic convergence increasingly masking institutional stratification. Geographic convergence in the procurement of AIDSD coexisted with persistent concentration in tertiary hospitals, and tier-based differences within regions became the main source of remaining inequality. Regional equity indicators alone therefore understate institutional inequality. Monitoring and policy for AI diagnostics should consider hospital tier alongside geography, rather than focusing on device volume alone.

Authors

Institutions

Publication Details

Journal
BMC Health Services Research
Published
2026-11-12
DOI
https://doi.org/10.1186/s12913-026-16020-w
Primary Topic
Artificial Intelligence in Healthcare and Education
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Geographic convergence masks institutional stratification in the scale-up of medical AI in China

Panpan Jiang, Lisha Cheng
BMC Health Services Research
Artificial Intelligence in Healthcare and Education
article

Geographic convergence masks institutional stratification in the scale-up of medical AI in China

Panpan Jiang, Lisha Cheng
article en

Abstract

Ensuring the equitable deployment of artificial intelligence (AI) in healthcare represents a critical bottleneck following algorithm validation. Yet empirical evidence on the real-world allocation of capital-intensive AI diagnostic devices remains scarce. In this paper, we used the real-time public procurement data to track the spatial and institutional diffusion of AI-enabled diagnostic and screening devices (AIDSD) across Chinese public hospitals from 2020 to 2024. Using dual-dimensional Theil-T indices, spatial agglomeration metrics, and structural location quotients, we identify two coexisting but contradictory trends. At the macro-geographic level, national device volume expanded 37.25-fold—from 8 units to 298—driving sustained convergence in procurement-based spatial equity indices (T pop : from 0.425 to 0.047; T hosp : from 0.591 to 0.070), with zero-adoption regions fully eliminated by 2022. Beneath this encouraging surface, however, severe institutional stratification persisted. Even when restricting analysis strictly to hospitals with capital acquisition capacity, tertiary hospitals disproportionately accumulated AIDSD throughout (LQ tier3 = 2.154 in 2024), while secondary hospitals experienced persistent relative deprivation (LQ tier2: from 0.361 to 0.320). A nested decomposition of total inequality further revealed that the within-region, tier-based component rose from 47.3% to 84.5% of total inequality between 2020 and 2024, consistent with geographic convergence increasingly masking institutional stratification. Geographic convergence in the procurement of AIDSD coexisted with persistent concentration in tertiary hospitals, and tier-based differences within regions became the main source of remaining inequality. Regional equity indicators alone therefore understate institutional inequality. Monitoring and policy for AI diagnostics should consider hospital tier alongside geography, rather than focusing on device volume alone.

BMC Health Services Research
Fudan University (CN)
Reduced inequalities, Good health and well-being
Artificial Intelligence in Healthcare and Education
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon