eon logo

Identify - Automate - Manage

eon is a system that combines deep learning and automated data extraction with intelligent patient tracking and management to defy disease
server side
Server Side
eon dashboard
The idea for Eon took birth in Dr. Akrum Alzubaidi’s mind in 2014, after he diagnosed a patient the same age as him with stage IV lung cancer. It was a humbling experience in many ways, but it got him thinking: what if there were a smarter, more effective way to collect, curate and distribute relevant data to caregivers and patients? And so, EonDirect was born – which has since become the market leader in making cancer screening programs more effective and more efficient, saving valuable time and money.
From an idea to a product that is now used by over a hundred hospitals in the US for managing numerous patients, from product research and spec to design and experience, all the way through to project management, engineering and scaling, OneByte has been proudly integral to Eon’s meteoric rise.
project life cycle

Product. Interface. Experience.

Healthcare tends to be extremely complex, tightly regulated, and tied up in legacy archaic systems. We worked closely with eon and potential client hospitals to identify the constraints, extract the pain-points and lay out the product roadmap. From there we conducted a thorough design process, that transformed the insights and research into wireframes, protoypes and high-fidelity design mockups.
eon product interface
machine learning

Machine Learning

In collaboration with eon’s data science team we developed state-of-the-art NLP incidental identification models that ingests patient exam reports, amplifies them by converting raw clinician speech text into structured data and flags potential disease findings.

Product Engineering

Parsing engine
Built on .net, the parsing engine allows each hospital to dynamically define its own rules on how data from its reports should be manipulated, extracted, and structured.
Template builder
Each hospital needs to collect different types of data. We built a flexible template builder in Node and Angular that allows them to define their own templates, data types, validations, even complex conditional logic, all through a simple intuitive UI.
Data rules middleware
Each incoming exam report needs to be treated differently per disease per hospital. We built a rules layer in Node that allows each hospital to dynamically define its own flow and data operations for reports as per its patient exam management process.
The dashboard
A highly intuitive, beautifully designed web-app for managing patient data, built with Angular, Node, Postgres and engineered to be performant even under immense loads and intensive data operations.
product engineering
infrastructure engineering

Infrastructure Engineering

CI/CD pipeline
All system components from the dashboard to ML Models are integrated into a CI/CD Pipeline, using Github, Jenkins, AWS Code-Deploy and Lambda, complete with production and staging environments for each hospital.
To manage hundreds of thousands of patients and millions of patient reports, we use a robust architecture that scales as needed, using Auto-Scaling Groups, Dockers, ECS, ABS, Fargate, Terraform, and Lambda.
Security & compliance
Our entire stack is fully HIPPA compliant and has a SOC-2 security certification.
VPN infrastructure
Every hospital gets its own dedicated Eon deployment and our team helps each hospital get connected with the VPN setup and implement it to stringent security standards.