Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2025 Feb;38(1):368-379.
doi: 10.1007/s10278-024-01200-z. Epub 2024 Jul 31.

A Cloud-Based System for Automated AI Image Analysis and Reporting

Neil Chatterjee  1   2 , Jeffrey Duda  3 , James Gee  3   4 , Ameena Elahi  5 , Kristen Martin  5 , Van Doan  5 , Hannah Liu  6 , Matthew Maclean  3 , Daniel Rader  7 , Arijitt Borthakur  3 , Charles Kahn  3   4 , Hersh Sagreiya  3   4 , Walter Witschey  3   4
Affiliations

A Cloud-Based System for Automated AI Image Analysis and Reporting

Neil Chatterjee et al. J Imaging Inform Med. 2025 Feb.

Abstract

Although numerous AI algorithms have been published, the relatively small number of algorithms used clinically is partly due to the difficulty of implementing AI seamlessly into the clinical workflow for radiologists and for their healthcare enterprise. The authors developed an AI orchestrator to facilitate the deployment and use of AI tools in a large multi-site university healthcare system and used it to conduct opportunistic screening for hepatic steatosis. During the 60-day study period, 991 abdominal CTs were processed at multiple different physical locations with an average turnaround time of 2.8 min. Quality control images and AI results were fully integrated into the existing clinical workflow. All input into and output from the server was in standardized data formats. The authors describe the methodology in detail; this framework can be adapted to integrate any clinical AI algorithm.

Keywords: AI; AI orchestrator; Informatics; Opportunistic screening; Steatosis.

PubMed Disclaimer

Conflict of interest statement

Declarations. Ethics Approval: The project was reviewed by the Institutional Review Board of the University of Pennsylvania on March 22, 2021 (IRB protocol #848519). A waiver of informed consent was requested and granted due to minimal risk, retrospective nature of the study, and deidentification of patient identifying information. The study was performed in accordance with the ethical standards as laid down in the 1964 Declaration of Helsinki and its later amendments. Consent to Participate: As detailed in the “Ethics Approval” section, this study was reviewed by our IRB and performed under a waiver for informed consent. Consent to Publish: As detailed in the “Ethics Approval” section, this study was reviewed by our IRB and performed under a waiver for informed consent. Competing Interest: The authors declare no competing interests.

Figures

Fig. 1
Fig. 1
Broad overview of AI orchestrator implementation. The AI orchestrator exchanges data with the PACS system, including AI-generated quality control images for radiologist review, and AI results are sent directly to the radiology report
Fig. 2
Fig. 2
a Overview of the enterprise radiology system with AI server integration. b Detailed view of operations performed by the AI orchestrator housed on the AI server, with the example hepatic steatosis AI algorithm. All information flow into and out of the server uses standardized formats
Fig. 3
Fig. 3
Sample quality control image sent to PACS from the AI Server. Transparent overlays show segmented liver, spleen, visceral fat, and subcutaneous fat from an AI segmentation algorithm used to screen for hepatic steatosis. This allows the radiologist to quickly verify accurate liver and spleen segmentations before including AI spleen and liver measurements in the clinical report
Fig. 4
Fig. 4
Number of non-contrast CTs processed per day over a 60-day period. All non-contrast abdominal CTs at multiple separate physical sites were sent to the AI server for opportunistic hepatic steatosis screening
Fig. 5
Fig. 5
Distribution of turnaround times (TAT) in a 60-day period. TAT is defined as the time from the start of the study transfer to the AI server to the completion of exporting AI results to PACS and the reporting engine. The average TAT was 2.8 min, and 94.9% of studies had a TAT of < 5 min
Fig. 6
Fig. 6
Distribution of hepatic attenuation of non-contrast abdominal CTs performed over a 60-day period. Attenuation of < 40 HU is consistent with hepatic steatosis
See this image and copyright information in PMC

References

    1. Artificial intelligence and machine learning (AI/ML)-enabled medical devices. https://www.fda.gov/medical-devices/software-medical-device-samd/artific.... - PMC - PubMed
    1. Bates DDB, Pickhardt PJ. CT-derived body composition assessment as a prognostic tool in oncologic patients: From opportunistic research to artificial intelligence–based clinical implementation. American Journal of Roentgenology. 2022;219(4):671–680. 10.2214/AJR.22.27749. - PubMed
    1. Rudie JD, Rauschecker AM, Xie L, et al. Subspecialty-level deep gray matter differential diagnoses with deep learning and Bayesian networks on clinical brain MRI: A pilot study. Radiol Artif Intell. 2020;2(5):e190146. 10.1148/ryai.2020190146. - PMC - PubMed
    1. Buls N, Watté N, Nieboer K, Ilsen B, de Mey J. Performance of an artificial intelligence tool with real-time clinical workflow integration – Detection of intracranial hemorrhage and pulmonary embolism. Physica Medica. 2021;83:154–160. 10.1016/j.ejmp.2021.03.015. - PubMed
    1. Shoshan Y, Bakalo R, Gilboa-Solomon F, et al. Artificial intelligence for reducing workload in breast cancer screening with digital breast tomosynthesis. Radiology. 2022;303(1):69–77. 10.1148/radiol.211105. - PubMed