GenAI-Powered Transformation: Optimizing Hospital Reporting for Improved Patient Care

Executive Summary

The client, a well-established healthcare organization in South Florida, grappled with significant challenges within their radiology department. Their manual quality assurance (QA) process for examining reports was slow and labor-intensive. Furthermore, the absence of a structured system to identify important observations unrelated to the primary condition (also known as incidental medical findings) led to a reactive approach to patient care.

Factspan’s team used GenAI to automate radiology reports, identifying critical cases and early incidental medical findings for proactive patient care. Additionally, by utilizing trend analytics, the organization gained valuable insights that informed patient care strategies and provided a holistic view of healthcare trends. This transformative approach significantly improved patient care coordination and responsiveness, addressing the organization’s radiology reporting challenges.

About the Client

The client is a respected faith-based not-for-profit healthcare organization and clinical care network in South Florida. With nearly 2 million patient visits annually and over 4,000 physicians, they are a trusted healthcare institution in the United States.

Known for their patient-centered approach, personalized treatment plans, advanced medical technologies, and team-based care, they operate multiple hospitals and physician practices. Their cuttingedge facilities and innovative telehealth services have set them apart in the healthcare industry. Additionally, their community outreach programs and healthcare education efforts benefit communities in Miami-Dade, Broward, and Palm Beach counties.

Business Challenge

1) The radiology team was severely handicapped by the manual process undertaken by the QA team to examine reports – manually flag findings and identifying critical patients for further assessment.

2) The hospital found itself experiencing delays in identifying critical patients promptly, which led to a reactive approach to patient care. The periodic capture of critical patient data reports was also too infrequent.

3) The lack of a structured system for incidental medical findings, which overlooked important observations, was concerning. Incidental medical findings refer to unexpected medical discoveries during a radiology exam. These findings are important to have a holistic understanding of a patient’s health and can be used to clarify the diagnostic uncertainties.

The team wanted to automate the manual QA process, for speed, efficiency, and effectiveness.

Our Solution

1) Instant Automated Reports: Factspan leveraged Language Learning Models (LLM) to automate the report analysis, providing the radiology team with instant access to this information.

2) Faster Data Capture: The team implemented faster and more frequent data capture, ensuring critical patients were not missed, resulting in fewer patient deteriorations.

3) Structured Incidental Reporting: Factspan implemented a structured and efficient way of reporting incidental medical findings.

4) Sharing Insights: Further, these unexpected findings were shared with peer hospitals to ensure patients received timely and appropriate care.

Business Impact

• GenAI automation made patient reports quickly accessible, eliminating delays in critical information retrieval
• It also reduced the risk of missing important patient data and observations, enabling healthcare providers to address patients’ needs immediately and effectively
• Improved efficiency is estimated to approximately save $1.2 million to $2.4 million annually for processing a million reports
• Insights from incidental medical findings and trend analyses went beyond radiology, benefiting various healthcare branches across the hospital