LEARNING OBJECTIVES:

1. Define acute aortic syndrome (AAS) and explain the challenges of making definitive diagnosis, timely triage, and effective surgical planning for AAS.

2. Demonstrate the effective adoption of a network-of-care model, taking advantage of the collective expertise of a team of geographically-separated specialty providers, in offering the highest level of AAS-tailored patient management.

3. Illustrate the use of centralized cloud-based technologies in allowing remote and mobile access of real-time information by geographically-dispersed specialists thereby delivering synchronized care in a time-sensitive manner.

4. Discuss the multidimensional and integrated roles AI and machine learning may play in areas of automatic detection of AAS and non-AAS mimickers (e.g., pulmonary embolism and coronary stenosis), vitals-, level of consciousness-, symptomatology-, and clinical history-based triage and risk stratification, adverse event prediction, and coordination of constrained hospital assets such as the operating rooms and CT scanners. Longitudinal cloud-based data collection and storage further provide increased opportunities in continuous practice quality improvement via immediate data-driven feedback.        

BACKGROUND:  Since 24/7 AAS-specific rapid response teams are less frequently available within the hospital, a network of specialists situated at various locations are recruited to orchestrate care. Cloud-based technologies with high computing powers represent the next generation of remote information sharing and access, synchronizing the expertise within any given network and minimizing ambiguity in diagnosis, patient transfer, and staff coordination. Shortened total transfer time, streamlined patient handover, and lower mortality rates have been reported consequently.

CONCLUSION:   Cloud-based technologies enable geographically-dispersed specialist providers to make time-sensitive decisions by granting real-time remote access to critical patient information including past and current medical history, radiology studies, and laboratory test results. Such availability of large medical data sets has been shown to improve patient outcome and, in the near future, may provide an avenue for machine learning and AI-augmented clinical decision-making. Future studies focusing on network security especially pertaining to patient privacy and personal health information are critical to validate the continued development of sub-centralized healthcare cloud systems.

From big data to big value: cloud-based, real-time information sharing leads to improved critical decision making and management of acute aortic emergencies; Can AI play a role?