Date of Award

2025

Degree Name

Healthcare Administration

College

College of Business

Type of Degree

M.S.

Document Type

Research Paper

First Advisor

Alberto Coustasse, Dr.PH. MD, MBA, MPH

Abstract

Introduction: There has been significant growth in the use of Artificial Intelligence (AI) in the healthcare industry, especially in Medical Imaging. Radiology has been the clear frontrunner in the adoption of AI in medicine, due in part to the massive amount of digital data available for use in Deep Learning (DL) AI integration has the potential to solve multiple challenges in radiology, address workload issues and transform the field.

Purpose of the Study: The purpose of the research was to evaluate the impact of implementing Artificial Intelligence in radiology to determine if these technologies have had an impact on workflow and efficiency, and also on the accuracy and precision of findings in disease detection compared to traditional radiological methods that had not utilized AI technologies.

Methodology: This study utilized a literature review. Four databases were used to identify sources. A total of 75 articles met the inclusion criteria for full-text review, of which 41 were used for the research. In addition, data from an anonymous online survey of practicing radiologists with experience using AI tools were analyzed to assess real-world applicability.

Results: The literature findings showed measurable improvements for radiologists working in conjunction with AI. Radiologists consistently experienced reduced interpretation times for digital images using AI assistance. AI usage demonstrated a higher accuracy in detecting lung and breast cancers and workload reductions of 40% to 86% were observed when AI was used to filter out normal studies, most notably in mammography and lung cancer screening.

Discussion: Survey responses indicated mixed perceptions of AI. While most respondents acknowledged its benefits in boosting diagnostic performance and identifying false negatives, concerns were raised about inefficiencies caused by poor integration, limitations in AI accuracy, and increased cognitive burden when AI results conflicted with their own interpretations.

Conclusion: The literature and survey data suggest that AI has improved efficiency, enhanced diagnostic accuracy, and reduced radiologist workload when properly integrated. However, continued refinement, better clinical integration, and updated regulatory standards are essential to fully realize AI’s potential in radiology.

Subject(s)

Health services administration.

Health facilities -- Business management.

Artificial intelligence -- Medical care.

Radiology.

Diseases -- Detection.

Diagnostic imaging.

Moderation -- Workload.

Download

Share

COinS