Publication

Explanations & Tutorials

How to Build a Medical Asset Tracking System with GCP

We're sharing how to build a hospital equipment tracking system using Google Cloud Products

June 28, 2018

As IoT applications continually move into major verticals including automotive and transportation, smart cities, and energy and utilities, healthcare has often been talked about as the next IoT market considering its size and need for connected solutions.

When thinking of healthcare and IoT, most people immediately conjure up images of remote health monitoring, wearables, and enhanced prescription management--and rightfully so. However, as discussed by Kayla Matthews in “6 Exciting IoT Use Cases in Healthcare,” a huge opportunity lies in the tracking of key hospital inventory and equipment.

In this article, we'll go over how to build a hospital asset tracking/equipment management system using Google Cloud Products, namely Cloud IoT Core, PubSub, BigQuery, and FireStore. Before jumping into the technical details, here’s a quick look at a prototype system we built in collaboration with Google:

 

According to the World Health Organization’s list of “Priority Medical Devices,” adult patient stretchers/beds, pediatric beds, wheelchairs, mobile workstations, and ventilators are the five most crucial pieces of equipment to track in a hospital. Hospital administrators want to track and locate equipment in real-time, inspect equipment history, and also set geofences to notify them when critical equipment is out of place.

To achieve this, you can use the following Google IoT Products:

Cloud IoT Core: This service manages the connection and security of devices that you intend to track. For example, if you have a BLE5 tracker on a pediatric bed sending location updates via HTTP or MQTT, Cloud IoT Core will manage that communication stream for you and automatically stream that message to Pub/Sub.

Cloud Pub/Sub: Pub/Sub is the data pipe that Google built to reliably send messages across the Internet at scale. Given Google’s experience in serving the data needs of Google, YouTube, and Gmail, Pub/Sub is a battle-tested, cloud-native messaging pipe.

BigQuery: In order to access historical data and also run analytics at scale, you need a data warehouse solution that can not only store data cheaply but also run queries fast.

FireStore: Lastly, to update the user interface in real-time, you need a flexible realtime database that will take in messages coming over via Pub/Sub and push changes to all the interfaces whether it be a desktop, tablet, or mobile interface.

All of these services are fully managed by Google so that you don't have to worry about scaling or healing the underlying infrastructure. While the reference architecture shown here is one that works for a simple equipment tracking solution, if you have more complex requirements, you can also find a list of Google’s products and other architecture suggestions on their solutions page.

Explore More from the Publication