When his grandfather was diagnosed with Acute Myeloid Leukemia (AML) in 2018, Adam Milton-Baker felt compelled to try to do something to help other families that may face the same fight in the future.
A software developer, Adam was interested in creating systems that integrate Artificial Intelligence (AI), the Internet of Things (IoT), and Robotics in web-based user interfaces. He also had previous experience with developing Artificial Intelligence for the detection of Breast Cancer. When Adam's grandfather was diagnosed with AML, he started the Peter Moss (his grandfather's name) Acute Myeloid & Lymphoblastic Leukemia AI Research Project. Two years later, in 2020, the Asociación de Investigacion en Inteligencia Artificial Para la Leucemia Peter Moss was founded. Its mission is to advance leukemia medical research and medical tech using the latest technologies while keeping it open-source and free. One of the Association’s most important developments is the Hospital Intelligent Automation Server (HIAS).
Making decisions based on contextual information
HIAS is an open-source server that stores all the data gathered by various AI & IoT devices. To make sense of all that data, the team was looking for a tool that could put all that information into context.
"We work a lot on IoT devices," Adam told CEF. "The idea is that people can go into our repository, download a particular project, and set up on their own HIAS network as a device."
The role of CEF's Context Broker
The Context Broker holds all of the contextual data from those devices (configuration, name, etc.). What ends up being shown in the User Interface (UI) is all the information gathered by the Context Broker.
"When we are inside the UI, and we want to have a look at a device or change some of the information, the UI requests the Context Broker, and the Context Broker returns all the information for that device. That's how Context Broker powers the network, it basically holds all of the information, everything that's on the network, and we can use that information to display on the UI," Adam explained.
With the Context Broker’s help, the Magic Leap Acute Lymphoblastic Leukemia Detection System, one of the open-source projects the team developed, aims to make physical and digital objects co-exist and interact in real-time. In this so-called mixed reality, the goal is to utilize an artificial intelligence algorithm trained to detect Acute Lymphoblastic Leukemia (ALL) in images of peripheral blood samples and display the results in Mixed Reality. The Context Broker used is a Python implementation developed by the team to align with the NGSI V2 specs.
The dataset that made this project possible is the Acute Lymphoblastic Leukemia Image Database for Image Processing by Fabio Scotti, Full Professor at Università Degli Studi di Milano Dipartimento di Informatica. The classifier used - an artificial intelligence model that "classifies" data - achieves 98% percent accuracy and works well at classifying unseen images (images it has never seen before) on a small test set. However, to really validate the model, the Association needs to make partnerships with hospitals and medical centers that may provide the amount of data required to be confident that the model works well in the real world.
The classifier is homed on a Raspberry Pi 4, which hosts a local endpoint, making the classifier accessible to the Magic Leap Application. It then allows the near real-time classification of ALL in Mixed Reality over the local network. The Raspberry Pi 4 is connected to the HIAS IoT JumpWay MQTT broker; requests are authenticated using the HIAS private blockchain via the MQTT IoT Agent; and classifications are stored in the historical database with a hash of the category for data integrity.
In the future
It was not that long ago that applying these technologies to everyday medicine seemed like a distant dream. However, increased cooperation among the scientific community has come to accelerate technological advances, especially in healthcare.
Thanks to the internet, we now live in an era where information and data are more easily available to researchers. One of the challenges now is finding easy-to-use, regulated ways to aggregate that information, enabling them to arrive at the best decisions. In that regard, the Context Broker can play a vital role not just in the fight against Leukemia but also against many other diseases for which we have not yet found a cure.
The CEF Context Broker
Context Broker is a digital platform component that enables the integration of gathered data, including insights for further exploitation. It acts in three distinct ways:
- Data Broker - assembling information from different systems, eventually belonging to various organisations, instead of performing in separate silos.
- Leveraging investments - by enabling the Digital Single Market with portable and interoperable solutions in a data marketplace.
- Easy development - by reducing costs and time in developing context information-based platforms and solutions, it is possible to create meaningful dashboards in a few days.
How can CEF help you?
At the Connecting Europe Facility, we give you access to free tools, support, and funding to help you build your digital services. Here are some other Building Blocks that might be of interest to you.
Collect data from sources and support smart decisions at the right time
A free and secure translation tool to break language barriers in the EU
Offers digital services capable of electronically identifying users from all across Europe