Connected Devices - How GE HealthCare Uses MongoDB to Manage IoT Device Lifecycle

Dr. Humza Akhtar
April 25, 2023 | Updated: April 26, 2023
#Healthcare

GE HealthCare, a global leader in medical technology, has turned to MongoDB to manage the lifecycle of its IoT devices, from deployment (Beginning of Life or BoL) to retirement (End of Life or EoL).

At GE HealthCare, MongoDB Atlas is used to persist device and customer data. These related data layers are utilized by the organization to develop customer experience strategies by providing greater efficiency, improving patient outcomes, and increasing access to care. The MongoDB document model easily combines data from diverse source systems while preserving its full fidelity. This flexibility allows seamless onboarding of new customers and related data sources without requiring time consuming schema modifications.

According to Emir Biser, Senior Data Architect at GE HealthCare, MongoDB Atlas is very appealing to the team because of its effective management, built-in monitoring and backup, global vertical and horizontal scalability, built-in security, and multi-cloud support.

MongoDB Atlas is a gamechanger. This technology stack is helping us streamline commercialization and bring market-ready solutions to deliver advanced healthcare.

Some of the recent tests resulted in an *83% decrease in retrieval time for critical data elements.

When all these features are put together, the tech stack is designed to help healthcare providers enhance productivity by reducing the complexity and time required to manage databases, enabling faster deployment of IoT devices.

Enhancing the IoT life cycle with MongoDB

GE HealthCare’s tech-stack is designed to accelerate the integration of healthcare applications by connecting IoT devices together with additional data sources into an aggregated clinical data layer. As the IoT device connections are established, multiple services are applied on the platform to support analytic and clinical applications.

Beginning of life - Device provisioning and configuration

As the device is being manufactured, the device parameters such as MAC and serial number are stored in MongoDB as a device digital representation. When the device is turned on, the GEHC team gets information about the device usage and the customer information. This information is used to validate the device. MongoDB is playing a crucial role in device provisioning by persisting the configuration information and making sure that the device is set up with the right configuration parameters. MongoDB change streams are used at this stage to make sure that the device gets the right parameters at the BoL stage.

Middle of life - Device usage and maintenance

Once the device comes online, it transmits both clinical and non-clinical information. The team at GE HealthCare uses MongoDB Atlas to help ensure clear separation between clinical and non-clinical as permissions, sensitivity, and access differs. Additionally, to understand how the device is being used compared to its standard configuration parameters. MongoDB’s real-time analytics capabilities help track key device performance metrics, such as battery life and identify trends and patterns in device usage. This enables the team to proactively address device issues, improve overall device performance and reliability for customers. GEHC is able to share these insights with customers to help optimize use of devices within their enterprise.

MongoDB Atlas Search is used to retrieve information about status of connected devices and usage patterns. Search Compound Geo JSON queries are used to look at products in a certain geographic region. Horizontal scalability with automatic sharding across clusters ensures Edison applications can continue to be cost effective while delivering real-time results.

MongoDB’s security features, including authorization, authentication and encryption, work with GEHC processes to enable teams working to protect device data from unauthorized access.

End of life - Device decommissioning and archiving

When the IoT device reaches the end of its lifecycle, GE HealthCare needs to decommission it and ensure that any data associated with the device is securely archived. By using MongoDB’s TTL (time-to-live) collections feature, the team automates the process of data deletion, reducing the data footprint. In addition, Atlas Online Archive helps to ensure that the data is always backed up and securely archived, reducing the risk of data loss and corruption. The authentication and authorization mechanisms help to ensure that decommissioned devices data can only be accessed by authorized personnel.

The future of GE HealthCare

According to Emir, the teams using MongoDB Atlas are excited about the benefits it brings, and they are looking forward to exciting new developments in Atlas platform.

We are helping teams achieve business goals across Imaging, Ultrasound Digital Solutions, and Patient Care Solutions. Our current strategy focuses on building solid pipelines to further help our medical device engineering teams deliver interoperability resulting in better care for our customers.

More on managing massive IoT devices

Internet of Things (IoT) is transforming the healthcare industry by providing real-time, actionable insights that improve patient outcomes and drive operational efficiencies. According to market analyst reports, the global IoT healthcare market is projected to reach around USD 446.52 billion by 2028 while exhibiting a CAGR of 25.9% between 2021 and 2028. In hospitals, IoT-enabled medical devices help improve patient safety and clinical experience by transmitting real-time monitoring and alerts in the event of device malfunctions or irregularities.

The life of an IoT device can be divided into three main stages: Beginning of Life (BoL), Middle of Life (MoL) and End of Life (EoL). During the BoL stage, the key activities are deployment design and provisioning. In this stage the device may be pre loaded with default credentials and configuration files. Once the device is installed and comes online, the focus in the MoL is to maintain its basic functional purpose as well as regularly updating firmware for reliability and security purposes. Over time, as new versions of devices are manufactured, the deployed devices need to be decommissioned by revoking the device certificate, archiving device data and disabling the model of device in the cloud as part of the EoL stage.

Figure 1: Three stages of IoT device lifecycle management

In each of the stages, the device has to be maintained to stay reliable, efficient, persistent and secure. Setting up telemetry from device to cloud/back end is just the tip of the iceberg. As the number of IoT devices deployed in healthcare continues to grow, so does the challenge of managing them efficiently. The large amount of data generated creates scalability challenges for IoT device management systems, which need to be able to handle large amounts of data and support the increased traffic. Different communication protocols make it challenging to integrate these devices into a unified system. Maintaining standard communication protocols and interoperability is critical to ensure seamless communication between devices and cloud backend. Finally, with the increasing number of cyber-attacks targeting IoT devices, it is critical to have robust security measures in place to protect against threats.

To learn more about GEHC digital offerings please visit https://apps.gehealthcare.com/

  • Test performed internally by GE HealthCare on company datasets and may not be replicable.

To learn more about MongoDB’s role in the healthcare and manufacturing industry, please visit our Manufacturing and Industrial IoT and Healthcare pages.

← Previous

Three Major IoT Data-Related Challenges and How to Address Them

IoT has formed itself a crucial component for future-oriented solutions and holds a massive potential of economic value. McKinsey & Company estimates that by 2030, IoT (Internet of Things) will enable $5.5 trillion to $12.6 trillion in value worldwide, including the value captured by consumers and customers. For proof of its growing popularity and consumers’ dependency on it, you likely don't need to look any further than your own wrist. From fitness bands to connected vehicles, smart homes, and fleet-management solutions in manufacturing and retail, IoT already connects billions of devices worldwide, with many more to come. As more IoT enabled devices come online, with increasingly sophisticated sensors, choosing the right underlying technology to make IoT solutions easier to implement and help companies seize new innovative opportunities is essential. In this blog we will discuss how MongoDB has successfully addressed three major IoT data-related challenges across various industries, including Manufacturing, Retail, Telecommunications, and Healthcare. The challenges are the following: Data Management Real-Time Analytics Supply Chain Optimization FIgure 1: MongoDB Atlas for IoT Let's dive right in! Data management Storing, transmitting, and processing the large amount of data that IoT devices produce is a significant challenge. Additionally, the data produced by IoT devices often comes in variable structures. This data must be carefully timestamped, indexed, and correlated with other data sources to make the context required for effective decision-making. This data volume and complexity combination makes it difficult to effectively and efficiently process data from IoT devices. Bosch Consider Bosch IoT Suite , a family of products and services in IoT device management, IoT data management, and IoT edge by Bosch Digital. These products and services hold over 250 international IoT projects and over 10 million connected devices. Bosch implemented MongoDB to store, manage, and analyze data in real time. MongoDB’s ability to handle structured, semi-structured, and unstructured data, and efficient data modeling with JSON make it easy to map the information model of each device to its associated document in the database. In addition, dynamic schemas support agile development methodologies and make it simple to develop apps and software. Adding new devices, sensors, and assets is easy, which means the team can focus on creating better software. ThingSpace Another example is that of ThingSpace , Verizon’s market-leading IoT connectivity management platform, which provides the network access required to deliver various IoT products and services. Verizon works with companies that purchase network access from it to connect their devices, bundled together with their own solutions, which they sell to end users. ThingSpace’s customers each sell an IoT product that needs reliable connectivity to ensure the devices always work, which WiFi cannot offer. Verizon’s monolithic RDBMS-based system would not be able to scale to handle both transactional and time-series workloads, so Verizon decided it needed a distributed database architecture going forward. MongoDB proved to be the only solution that scaled to meet Verizon’s requirements across different use cases and combinations of workload types. The immense processing needs resulting from the high number of devices and high velocity of messages coming in were only addressed by MongoDB’s highly available, scalable architecture. Native MongoDB Time Series allow for improved performance, through optimized storage with clustered indexes and optimized Time-Series query operators. MongoDB's advanced capabilities, such as flexible data modeling, powerful indexing and Time Series provide an effective solution for managing the complex and diverse data generated by IoT devices. Real-time analytics Real-time data analytics, one of the most crucial parts of big data analytics today, brings value to businesses for making more data-driven real-time decisions. However, despite its importance, very few can respond to changes in data minute by minute or second by second. Many challenges arise when it comes to the implementation of real-time analytics for enterprises. Storing such a huge volume of data and analyzing it in real time is an entirely different story. Thermo Fisher Cloud Let’s consider the Thermo Fisher Cloud, one of the largest cloud platforms for the scientific community on AWS. MS Instrument Connect allows Thermo Fisher customers to see live experiment results from any mobile device or browser. Each experiment produced millions of "rows" of data, which led to suboptimal performance with existing databases. Internal developers needed a database that could easily handle a wide variety of fast-changing data. MongoDB's expressive query language and rich secondary indexes provided the flexibility to support both ad-hoc and predefined queries customers needed for their scientific experiments. Anytime I can use a service like MongoDB Atlas, I’m going to take that so that we at Thermo Fisher can focus on what we’re good at, which is being the leader in serving science. Joseph Fluckiger, Sr. Software Architect @Thermo Fisher MongoDB Atlas scales seamlessly and is capable of ingesting enormous amounts of sensor and event data to support real-time analysis for catching any critical events or changes as they happen. That gives organizations new capabilities, including: Capturing streaming or batch data of all types without excessive data mapping Analyzing data easily and intuitively with a built-in aggregation framework Delivering data insights rapidly and at scale with ease With MongoDB organizations can optimize queries to quickly deliver results to improve operations and drive business growth. Supply chain optimization Items move through different locations in the supply chain, making it hard to maintain end-to-end visibility throughout their journey. The lack of control on any stage can dramatically harm the efficiency of planning, slow down the entire supply chain and ultimately result in lower return on investment. From optimizing warehouse space by sourcing raw materials as needed, to real-time supply chain insights, IoT-enabled supply chains can help significantly optimize these processes by eliminating blind spots and inefficiencies. Longbow Advantage Longbow Advantage delivers substantial business results by enabling clients to optimize their supply chains. Millions of shipments move through multiple warehouses every day, generating massive quantities of data throughout the day that must be analyzed for real-time visibility and reporting. Its flagship warehouse visibility platform, Rebus, combines real-time performance reporting with end-to-end warehouse visibility and intelligent labor management. Longbow needed a database solution that could process quantities of that scale and deliver real-time warehouse visibility and reporting at the heart of Rebus, and it knew it could not rely on monolithic, time-consuming spreadsheets to do so. It became clear that MongoDB’s document database model was a good match and would allow Rebus to gather, store, and build visibility into disparate data in near real time. Another key component of smart supply chain solutions is IoT-enabled mobile apps that provide real-time visibility and facilitate on-the-spot data-driven decisions. In such situations, the offline first paradigm becomes crucial, since staff need access to data in areas where connectivity is poor or nonexistent. Realm by MongoDB is a lightweight, object-oriented embedded database technology for resource constrained environments. It is an ideal solution for storing data on mobile devices. By utilizing MongoDB’s Realm SDKs, which wrap the Realm database, and Atlas Device Sync , which enables seamless data synchronization between MongoDB and Realm on your mobile phone with minimal developer efforts, businesses can rapidly develop mobile applications and drive innovation. MongoDB provides a powerful solution for IoT-enabled supply chains that can optimize processes and eliminate inefficiencies, enabling organizations to make data-driven decisions and improve supply chain efficiency. Conclusion The IoT industry is rapidly evolving, and as the number of connected devices grows, so do the challenges faced by businesses leveraging these solutions. Through a range of real-world use cases, we have seen how MongoDB has helped businesses deal with IoT data management, perform real-time analytics and optimize their supply chains, driving innovation in a variety of industries. With its unique features and capabilities, designed to manage the heavy lifting for you, MongoDB is well-positioned to continue playing a crucial role in the ongoing digital transformation of the IoT landscape. Want to learn more or get started with MongoDB? Check out our IoT resources MongoDB IoT Reference Architecture Migrate existing applications - with Relational Migrator MongoDB & IIoT ebook IoT webpage

April 24, 2023
Next →

Retrieval Augmented Generation for Claim Processing: Combining MongoDB Atlas Vector Search and Large Language Models

Following up on our previous blog, AI, Vectors, and the Future of Claims Processing: Why Insurance Needs to Understand The Power of Vector Databases , we’ll pick up the conversation right where we left it. We discussed extensively how Atlas Vector Search can benefit the claim process in insurance and briefly covered Retrieval Augmented Generation (RAG) and Large Language Models (LLMs). MongoDB.local NYC Join us in person on May 2, 2024 for our keynote address, announcements, and technical sessions to help you build and deploy mission-critical applications at scale. Use Code Web50 for 50% off your ticket! Learn More One of the biggest challenges for claim adjusters is pulling and aggregating information from disparate systems and diverse data formats. PDFs of policy guidelines might be stored in a content-sharing platform, customer information locked in a legacy CRM, and claim-related pictures and voice reports in yet another tool. All of this data is not just fragmented across siloed sources and hard to find but also in formats that have been historically nearly impossible to index with traditional methods. Over the years, insurance companies have accumulated terabytes of unstructured data in their data stores but have failed to capitalize on the possibility of accessing and leveraging it to uncover business insights, deliver better customer experiences, and streamline operations. Some of our customers even admit they’re not fully aware of all the data in their archives. There’s a tremendous opportunity to leverage this unstructured data to benefit the insurer and its customers. Our image search post covered part of the solution to these challenges, opening the door to working more easily with unstructured data. RAG takes it a step further, integrating Atlas Vector Search and LLMs, thus allowing insurers to go beyond the limitations of baseline foundational models, making them context-aware by feeding them proprietary data. Figure 1 shows how the interaction works in practice: through a chat prompt, we can ask questions to the system, and the LLM returns answers to the user and shows what references it used to retrieve the information contained in the response. Great! We’ve got a nice UI, but how can we build an RAG application? Let’s open the hood and see what’s in it! Figure 1: UI of the claim adjuster RAG-powered chatbot Architecture and flow Before we start building our application, we need to ensure that our data is easily accessible and in one secure place. Operational Data Layers (ODLs) are the recommended pattern for wrangling data to create single views. This post walks the reader through the process of modernizing insurance data models with Relational Migrator, helping insurers migrate off legacy systems to create ODLs. Once the data is organized in our MongoDB collections and ready to be consumed, we can start architecting our solution. Building upon the schema developed in the image search post , we augment our documents by adding a few fields that will allow adjusters to ask more complex questions about the data and solve harder business challenges, such as resolving a claim in a fraction of the time with increased accuracy. Figure 2 shows the resulting document with two highlighted fields, “claimDescription” and its vector representation, “claimDescriptionEmbedding” . We can now create a Vector Search index on this array, a key step to facilitate retrieving the information fed to the LLM. Figure 2: document schema of the claim collection, the highlighted fields are used to retrieve the data that will be passed as context to the LLM Having prepared our data, building the RAG interaction is straightforward; refer to this GitHub repository for the implementation details. Here, we’ll just discuss the high-level architecture and the data flow, as shown in Figure 3 below: The user enters the prompt, a question in natural language. The prompt is vectorized and sent to Atlas Vector Search; similar documents are retrieved. The prompt and the retrieved documents are passed to the LLM as context. The LLM produces an answer to the user (in natural language), considering the context and the prompt. Figure 3: RAG architecture and interaction flow It is important to note how the semantics of the question are preserved throughout the different steps. The reference to “adverse weather” related accidents in the prompt is captured and passed to Atlas Vector Search, which surfaces claim documents whose claim description relates to similar concepts (e.g., rain) without needing to mention them explicitly. Finally, the LLM consumes the relevant documents to produce a context-aware question referencing rain, hail, and fire, as we’d expect based on the user's initial question. So what? To sum it all up, what’s the benefit of combining Atlas Vector Search and LLMs in a Claim Processing RAG application? Speed and accuracy: Having the data centrally organized and ready to be consumed by LLMs, adjusters can find all the necessary information in a fraction of the time. Flexibility: LLMs can answer a wide spectrum of questions, meaning applications require less upfront system design. There is no need to build custom APIs for each piece of information you’re trying to retrieve; just ask the LLM to do it for you. Natural interaction: Applications can be interrogated in plain English without programming skills or system training. Data accessibility: Insurers can finally leverage and explore unstructured data that was previously hard to access. Not just claim processing The same data model and architecture can serve additional personas and use cases within the organization: Customer Service: Operators can quickly pull customer data and answer complex questions without navigating different systems. For example, “Summarize this customer's past interactions,” “What coverages does this customer have?” or “What coverages can I recommend to this customer?” Customer self-service: Simplify your members’ experience by enabling them to ask questions themselves. For example, “My apartment is flooded. Am I covered?” or “How long do windshield repairs take on average?” Underwriting: Underwriters can quickly aggregate and summarize information, providing quotes in a fraction of the time. For example, “Summarize this customer claim history.” “I Am renewing a customer policy. What are the customer's current coverages? Pull everything related to the policy entity/customer. I need to get baseline info. Find relevant underwriting guidelines.” If you would like to discover more about Converged AI and Application Data Stores with MongoDB, take a look at the following resources: RAG for claim processing GitHub repository From Relational Databases to AI: An Insurance Data Modernization Journey Modernize your insurance data models with MongoDB and Relational Migrator

April 18, 2024