Workflow Integration of Research AI Tools into a Hospital Radiology Rapid Prototyping Environment.

The field of artificial intelligence (AI) in medical imaging is undergoing explosive growth, and Radiology is a prime target for innovation. The American College of Radiology Data Science Institute has identified more than 240 specific use cases where AI could be used to improve clinical practice. In this context, thousands of potential methods are developed by research labs and industry innovators. Deploying AI tools within a clinical enterprise, even on limited retrospective evaluation, is complicated by security and privacy concerns. Thus, innovation must be weighed against the substantive resources required for local clinical evaluation. To reduce barriers to AI validation while maintaining rigorous security and privacy standards, we developed the AI Imaging Incubator. The AI Imaging Incubator serves as a DICOM storage destination within a clinical enterprise where images can be directed for novel research evaluation under Institutional Review Board approval. AI Imaging Incubator is controlled by a secure HIPAA-compliant front end and provides access to a menu of AI procedures captured within network-isolated containers. Results are served via a secure website that supports research and clinical data formats. Deployment of new AI approaches within this system is streamlined through a standardized application programming interface. This manuscript presents case studies of the AI Imaging Incubator applied to randomizing lung biopsies on chest CT, liver fat assessment on abdomen CT, and brain volumetry on head MRI.

Angiographic Efficacy of the Atriclip Left Atrial Appendage Exclusion Device Placed by Minimally Invasive Thoracoscopic Approach.

OBJECTIVES: This study sought to assess long-term left atrial appendage (LAA) closure efficacy of the Atriclip applied via totally thoracoscopic (TT) approach with computed tomographic angiography. BACKGROUND: LAA closure is associated with a low risk for atrial fibrillation-related embolic stroke. The Atriclip exclusion device allows epicardial LAA closure, avoiding the need for post-operative oral anticoagulation. Previous data with Atriclip during open chest procedures show a high efficacy rate of closure >95%. METHODS: Three-dimensional volumetric 2-phase computed tomographic angiography ≥90 days post-implantation was independently assessed by chest radiology for complete LAA closure on all consented subjects identified retrospectively as having had a TT-placed Atriclip at Vanderbilt University Medical Center from June 13, 2011, to October 6, 2015. RESULTS: Complete LAA closure (defined by complete exclusion of the LAA with no exposed trabeculations, and clip within 1 cm from the left circumflex artery) was found in 61 of 65 subjects (93.9%). Four cases had incomplete closure (6.2%). Two clips were placed too distally, leaving a large stump with exposed trabeculae. Two clips failed to address a secondary LAA lobe. No major complications were associated with TT placement of the Atriclip. Follow-up over 183 patient-years revealed 1 stroke in a patient with complete LAA closure and no thrombus (hypertensive cerebrovascular accident). CONCLUSIONS: Angiographic LAA closure efficacy with a TT-placed Atriclip is high (93.9%). The clinical significance of a remnant stump is unknown. Confirmation of complete LAA occlusion should be made before cessation of systemic anticoagulation.