Thank you for wanting to learn more about us. Download the PDF
here.
To stay updated with our latest content, please follow us on LinkedIn.
Oops! Something went wrong while submitting the form.
HealthTech

Predicting Onset of Diabetes

with a HealthTech Innovator

Keywords

Diabetes Prevention, AI-ML

Services Rendered

Product Design

Product Engineering

Continuing Design

Tech stack

React, Material UI, Scikit-learn, Pytorch, PostgreSQL, Docker

Introduction

Over 96 million adults on the brink of developing diabetes. 1 in 3 people in the USA have diabetes. A healthtech innovator partnered with Zemoso to productize a solution that enables prevention using machine learning (ML) and artificial intelligence (AI).

iNDUSTRY CHALLENGE

Diabetes, if diagnosed early, can be prevented or delayed with individualized care programs, better treatments, and regular screening. However, conventional diabetes solutions, like continuous blood glucose monitoring (CGM) devices for diabetes are reactive, primarily built for managing diabetes, leaving a gap in prevention efforts. Our client identified this missed opportunity and developed a proactive solution that can predict the onset of diabetes up to 5 years in advance. By alerting individuals before symptoms even appear, the solution empowers people to make lifestyle changes and healthcare professionals to allocate appropriate resources and care to high-risk patients.

Zemoso Labs Partnership Challenge

The healthtech company partnered with Zemoso for our ability to build  nuanced AI-ML models, and take it to development with mature engineering and design practices. At every step, we evaluated the AI-ML model on accuracy, and its ability to be incorporated into the product. We helped our customer to navigate product innovation, and turn a good idea into a great product through:

  1. A self-organized pod with product, design, and engineering expertise that re-oriented based on needs at a moment in time.
  2. Continuously incorporating new volumes of data and released an update to the model every three months to iteratively improve outcomes for the product.
  3. Prioritizing model accuracy. Over our period of engagement, we trained and fine-tuned the model, moving the validation score of 0.60 to 0.98.
  4. Timeboxing every milestone to ensure meeting deadlines and pivot agility.

Impact created

Over our period of engagement, we helped train the solution on massive volumes of data that ultimately resulted in predicting the onset of diabetes with 85% accuracy using the Accelerated Failure Time (AFT) survival regression model, prioritizing individualized predictions.

85%

Model accuracy

What are our clients saying?

Our clients love what we do:

How did we do this?

Heading 1

Heading 2

Heading 3

Heading 4

Heading 5
Heading 6

Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure dolor in reprehenderit in voluptate velit esse cillum dolore eu fugiat nulla pariatur.

Block quote

Ordered list

  1. Item 1
  2. Item 2
  3. Item 3

Unordered list

  • Item A
  • Item B
  • Item C
Text link

Bold text

Emphasis

Superscript

Subscript

Our product, design, and engineering pods actively collaborated with stakeholders to launch the Minimum Viable Product (MVP) version. Our partnership with this healthtech innovator was split into two phases:

Phase 1 - Proof of Concept

The initial phase of our partnership was focused on validating the idea and creating a concrete proof of concept to demonstrate its feasibility.

Customized Design Sprint: We kicked off the partnership with a customized design sprint inspired by the Google Ventures Design Sprint to dig into the challenges faced by patients, healthcare providers, and insurance companies.

Diverge and Converge: Applying the double-diamond approach and deep research, we created space for brainstorming and prioritization — converging to focus on primary and immediate pain points.

Golden Path: We used the Jobs-to-be-Done (JTBD) framework to map out the golden path, followed by building, testing, and iterating the proof of concept via  a prototype with pilot customers.

Phase 2 - MVP to Scale

We had some early wins with positive response from early adopters with the prototype. We then put the ML algorithm into production and created a solution that could predict the onset of diabetes in patients with near-perfect accuracy, like no other solution in the market. The learning and improving the accuracy of the model was of course a priority.

Leveraging mature engineering practices, this is how we partnered with our customer to expand product capabilities:

Data Sourcing: We aggregated data from OpenEHR and a partner hospital. Of course all the data was obfuscated to maintain compliance with HIPAA standards of privacy. We cleaned it up and filled in the gaps in the data using imputation techniques.

Feature Selection and Engineering: We worked with the client to identify high-impact features from a wide range of potential predictors and built them for improved performance.

Algorithm Development: Our engineers built a fast-learning, self-improving algorithm utilizing the Accelerated Failure Time (AFT) survival regression model due to its suitability for modeling individual time-to-event data. This choice proved instrumental in achieving 85% accuracy in predicting the onset of diabetes.

Multi-modal Data Fusion: We utilized multiple modes of biometric and demographic data from various sources, including hospital visits and social history, followed by streamlining, processing, and feeding data into the model to facilitate continuous training.

Continuous Improvement: With each new iteration and implementation, we ensured that the algorithm continued to learn and improve its predictions, resulting in a continuously evolving and highly accurate solution.

Robust CI/CD pipelines: A Continuous Integration/Continuous Deployment (CI/CD) pipeline was deployed to shorten feedback loops and consistently meet product engineering milestones.

Scalability: We designed the solution to be easily scalable by streamlining data ingestion and processing, ensuring that it can handle increasing data volumes without compromising performance.

In Conclusion

Zemoso has always been at the forefront of AI innovation across industries as we enable innovators to solve critical problems by leveraging the power of data, AI, and ML.

If you have an idea with the potential to make a big impact, but it's been waiting for the right moment, let's talk. We can bring it to life with a proof of concept in 2 weeks. We have expertise in delivering realistic prototypes for AI/ML, and Generative AI use cases for startups and enterprises.  

P.S. We have strict Non-disclosure agreements (NDAs) with many of our clients. The data, insights, and capabilities discussed in this case study have been anonymized to protect our client’s identity and don’t include any proprietary information.

CASE STUDIES

We help you convert great ideas

Initiatives : All
Strategy & design
New product development
Product expansion
Initiatives : All
Strategy & design
Product expansion
Initiatives : All
Strategy & design

Got an idea?

Together, we’ll build it into a great product

Follow us on

Dallas, USA

London, UK

Waterloo, Canada

Hyderabad, India

©2024 Zemoso Technologies. All rights reserved.