Leveraging artificial intelligence (AI) to optimize multidisciplinary tumor board workflows at OSF HealthCare’s network of cancer centers.

Abstract

e13604 Background: In recent decades, technological advancements and personalized medicine have made cancer care increasingly complex, requiring collaboration across multiple specialties. Tumor boards (TBs) serve as essential platforms, bringing together experts to review patient data and treatment options, leveraging technology to streamline decision-making and improve patient outcomes. OSF HealthCare (OSF), an integrated health system with oncology services spanning seven hospitals in Illinois and Michigan, hosts 14 TB meetings at varying frequencies (weekly, bi-weekly, or monthly). In 2023, OSF implemented new software to enhance TB management. This study evaluates the outcomes of the software in optimizing and streamlining TB activities across its extensive network. Methods: A structured, stakeholder-driven approach guided the implementation, supported by steering and implementation committees comprising oncology leadership, nurses, and TB staff. Monthly stakeholder meetings ensured the platform was tailored to each hospital’s needs. Cancer Insights collaborated with OSF’s project management and IT teams through weekly virtual meetings, on-site assistance during go-live, and a post-rollout ticketing system. The AI platform integrates in real-time with the electronic health record (EHR), offering single sign-on access and features such as automated patient list management, case summary preparation, and streamlined documentation of recommendations. It incorporates multimodal data which includes clinical history, pathology, molecular/genomic data, and imaging to guide multidisciplinary treatment decisions. Results: Between February 2023 and January 2025, 887 TB meetings reviewed 4,285 cases involving 3,561 unique patients. Automated data integration and case presentation generation reduced manual data entry by 75%, significantly saving staff time. Preparation time decreased by 40%, enabling more cases to be addressed per meeting. In 90% cases, a Tumor Board Note was seamlessly integrated into the EHR, with accreditation documentation for COC, NAPBC, and NAPRC embedded in the automated workflow, improving compliance. Additionally, 90% of TB recommendations were communicated to patients efficiently, ensuring timely decision-making and enhanced patient-centered care. Conclusions: The AI platform successfully optimized TB workflows across OSF HealthCare’s Cancer Centers, demonstrating its ability to integrate multimodal data, streamline operations, and improve care delivery for complex cancer cases. Future efforts will leverage advanced analytics to predict case complexity, automate staging workflows, and clinical trial matching.

Ähnliche Arbeiten