Delivering Faster Patient Outcomes Leveraging Real-Time Surgical Planning Image Analytics on the Edge

By Sri Gangidi, Kevin Landwehr |  11TEN Innovation Partners

Ashely Keeley, Ashish Shah, Brian Warwick, Aditya Lolla | Amazon Web Services


The process of selecting implant devices prior to surgical procedures is cumbersome and time consuming. It may take days or weeks to consult with device representatives and manufacture an appropriately sized device. The process involves logistics, continuous communication, and manufacturing that will determine the success of the surgical procedure. Utilizing segmentations, measurements, clinical trial forms, and collaborative sharing tools, AngioCloud aims to alleviate this critical path for both providers and device representatives.


AngioCloud is a HIPAA-secured, cloud-based vascular imaging and care collaboration platform, cleared as an FDA Class II Medical Device. It offers a suite of web and mobile-enabled software tools designed for use by physicians and medical device representatives. The platform’s core functionality revolves around accessing, evaluating, sharing, and collaborating on 3D vascular cases. This makes it a valuable tool in various medical processes, including diagnosis, pre-interventional planning, surgical planning, collaborative medical care, and the administration of e-clinical trials.

  • AngioCloud drives value with its ability to facilitate the sharing and analysis of 3D vascular images with unprecedented ease and flexibility. By enabling access from any connected device, AngioCloud overcomes the limitation of needing dedicated hospital workstations for high-quality vascular imaging. This feature is particularly beneficial for medical procedures that require precise sizing, placement, and deployment of medical devices. Additionally, it supports e-consults, pre-surgical planning, and clinical decision-making processes, enhancing the overall efficiency and effectiveness of medical care. 
  • AngioCloud’s introduction of immediate 3D image sharing technology represents a significant advancement in medical imaging and collaboration, positioning it as a groundbreaking tool in the healthcare industry. AngioCloud is a centralized online vascular imaging and care collaboration platform, developed on AWS and used by physicians to access, analyze, store, and collaborate on vascular cases to support diagnosis, pre-operational planning, consultation, and e-clinical trial administration.
  • The platform provides a segmentation and analysis platform that can be accessed from the cloud and provides a de-identification and sharing mechanism that allows clinicians to analyze the images and create annotations, screenshots, and notes. 


Current AngioCloud architecture provided by Scopic Software

The AngioCloud surgical planning and collaboration solution was created on the AWS Cloud Platform. Developed to enable clinical collaboration between surgeons and device manufacturer representatives, the AngioCloud platform was built with the ability to scale as needed as the user base increases while remaining a cost-effective solution for the care team.

The current AngioCloud solution architecture consists of several components, that include: 

  1. AWS S3 bucket for storing images. 
  2. AWS Lambda function for processing images. 
  3. AWS DynamoDB table for storing image metadata. 
  4. AWS SQS queue for managing image processing requests. 
  5. AWS API Gateway for handling API requests. 
  6. AWS CloudWatch for monitoring the system. 
  7. AWS Batch for processing images in parallel. 
  8. Opensource Medical Imaging library for analyzing images. 
  9. AWS SNS for sending notifications. 
  10. AWS CloudFront for serving images to users. 
  11. AWS Cognito for Authentication and SSO. 
  12. AWS Route 53 for address resolution and DNS. 

This architecture is designed to process and analyze images in a scalable and efficient manner. The Amazon S3 bucket is used to store images, while the Amazon Lambda function processes the images. The Amazon DynamoDB table is used to store metadata about the images, and the Amazon SQS queue is used to manage image processing requests. The Amazon API Gateway is used to handle API requests, while the Amazon CloudWatch is used to monitor the system. The Amazon Batch is used to process images in parallel, and the Opensource Medical Imaging APIs are used to analyze images. Finally, the Amazon SNS is used to send notifications, and the Amazon CloudFront is used to serve images to users. 

This architecture helps to process and analyze large volumes of images quickly and efficiently, while also providing a scalable and collaborative solution for image processing and analysis. 


Accurate sizing of medical devices such as stents, prosthetics, or orthopedic implants is crucial for successful medical interventions. However, sizing decisions are often based on off-site analysis of imaging data, leading to delays in identifying the right type/size of the medical implant. This could also lead to requiring re-imaging of the patient. Sometimes such delays can result in delaying the surgery or medical procedure by several weeks, delaying healthcare to the patient and revenue to the hospital. 

With a MEC based Edge solution deployed:  

  • The surgeon and devices reps will have real-time access to image analysis and medical device match, check against current supplies on hand and reduce rescans and bring the patient into surgeries sooner, saving lives, and improving OR efficiency.  
  • MEC solution can support doing multiple image segmentation and analysis at the same time and deliver results in real-time.  
  • MEC can support other imaging and clinical solutions being deployed on Outpost giving access to a variety of healthcare solutions to clinicians and surgeons.  

Proposed Architecture: AWS Outpost based Edge Solution 

The move to AWS Outposts brings AWS infrastructure, services, APIs, and tools to virtually any data center, co-location space, or on-premise facility. For healthcare providers, this means sensitive medical imaging data that was previously required to be stored and processed in distant data centers can now remain on-premise, satisfying the healthcare industry’s strict data residency laws and regulations. This local data residency is crucial for organizations handling sensitive patient information, ensuring that they meet legal and ethical standards for data sovereignty. 

Coupled with AWS Wavelength, which brings AWS services to the Edge of the telecommunications network, the latency is significantly reduced. This low-latency environment is paramount when healthcare professionals are accessing and analyzing large, complex medical images for surgical planning. By processing this data closer to its source and users, healthcare providers can ensure that surgeons and medical staff receive the crucial information they need without the delays that can occur with traditional cloud environments. 

In addition to delivering public access to the AngioCloud platform, when deployed on the Edge, the AngioCloud on-prem module can be accessed privately using a locally proxied access URL. 


The Cloud to Edge transformation includes a shift from using DynamoDB to AWS Relational Database Service (RDS). AWS RDS is a distributed relational database service that supports different database engines, offering the familiar management and scaling capabilities desired by enterprises but with the added benefit of being manageable locally via Outposts. This transition enables the AngioCloud platform to handle relational data more effectively, which is often the case with patient records and surgical planning data. 

Moreover, by leveraging Lambda functions within AWS Greengrass IoT, the system facilitates powerful image analytics directly on-premise. AWS Greengrass extends AWS to edge devices so they can process the medical images locally where they generate while still using the cloud for management, analytics, and durable storage. In this scenario, deploying Lambda functions to Greengrass allows complex image analytics algorithms to run adjacent to imaging devices and patient care areas, thereby enabling faster processing and immediate insights into medical imaging, crucial for time-sensitive and life-saving surgical planning. 

This local processing power not only accelerates the response times for imaging analytics but also opens the door to more advanced and immediate surgical planning capabilities. Surgeons can utilize enhanced 3D visualizations and potentially augmented reality interfaces to plan surgeries with greater precision and confidence. The combined benefit of meeting data residency requirements and accessing low-latency, high-performance computing resources fundamentally transforms the healthcare sector’s approach to medical image analysis and surgical preparation. 

By implementing these edge computing solutions, healthcare facilities can realize the vision of a modernized, efficient, and patient-centric surgical planning process that meets the needs of both practitioners and patients, ensuring that high-quality care is delivered promptly and securely, summarizing the value proposition as follows: 

Benefits and Value Proposition

Benefits of the Edge solution via Outpost and Wavelength include: 

  • Meet stringent Healthcare Data residency requirements. 
  • Private and local Access low latency access to AWS functions. 
  • Deliver local processing of DICOM and 3DRA images for faster response times.

Connect with us to learn more about how 11TEN is driving the art of what’s possible in healthcare innovation: