Development of a digital cloud imaging platform based on a “pay-for-performance” model: an empirical study on cost control and resource conservation

Introduction Traditional physical film-based imaging services impose a heavy financial and operational burden on public hospitals. This study aims to evaluate the actual effectiveness of a digital cloud imaging platform—deployed under a model where “service providers make upfront investments and hospitals pay based on performance”—in terms of controlling medical costs, optimizing clinical workflows, and enhancing the patient experience, thereby providing a replicable management pathway and empirical evidence for the digital transformation of imaging services. Methods A single-center, quasi-experimental before-and-after study design was adopted, supplemented by interrupted time series (ITS) analysis as the primary method for causal inference, with Deyang People’s Hospital serving as the study site. Prior to project implementation, the study included data from 160,183 eligible imaging examinations; after implementation, it included data from 295,940 eligible imaging examinations. The platform was built on a hybrid cloud architecture. Through collaboration with a third-party technology provider, the hospital was able to deploy the system without upfront capital expenditures, replacing traditional physical films with “cloud films.” Operational data was collected from the hospital’s PACS system and the administrative backend of the digital cloud imaging service platform both before the platform’s launch (January 1, 2022–December 31, 2022) and after (August 12, 2024–August 11, 2025). Descriptive statistical methods, the Mann–Whitney U test, and interrupted time series analysis to compare differences in data before and after project implementation, thereby evaluating the system’s actual effectiveness. Results Within 12 months of the platform’s operation, the hospital achieved net savings of 4,159,500 yuan on film-related consumables; the median time from examination to reporting for physicians was reduced from 6.86 h to 2.41 h ( p < 0.001), a decrease of approximately 64.9%, resulting in a significant improvement in work efficiency. Patients experienced “zero wait time” for film retrieval, shortening their total visit duration; interrupted time series analysis showed that patient satisfaction exhibited a sustained quarterly upward trend following the intervention, increasing by an average of 0.46 percentage points per quarter ( P < 0.001), though no immediate jump in satisfaction was detected at the time of intervention implementation ( p = 0.440). Discussion Within the framework of this study’s single-center, quasi-experimental design, a temporal association was observed between the “service provider upfront investment, pay-for-performance” model of the digital cloud imaging platform and reductions in film and consumable costs, improvements in physicians’ image-reading efficiency, and enhancements in patient satisfaction. However, due to limitations in the study design, further evidence is needed to confirm the causality of these associations. The long-term financial sustainability and cross-institutional scalability of this model require validation through longer-term follow-up and multicenter studies.