DiaT2

Title of Funded Project: An Intelligent Low-cost Method for Early Detection of Type 2 Diabetes
Number: A2017A1-240
Agreement Duration: From 12/05/2017 To 12/05/2019
Principal Investigator: Professor Mohammad H. Nadimi-Shahraki
Co-principal Investigator: Hoda Zamani 

Importance: Type 2 diabetes is a serious condition that can lead to complications such as heart disease, stroke, kidney failure, blindness, and amputation. This chronic disease affects more than 450 million people worldwide, and its prevalence has rapidly increased and doubled in the last 15 years, which has challenged many countries' healthcare systems. Fortunately, there have been proposed methods for early detection, which can prevent or delay it by adopting a healthy lifestyle, such as eating well, moving more, and losing weight if needed. However, current methods of diagnosis, such as blood tests, are often invasive, costly, and time-consuming. Moreover, many people may not have access to adequate healthcare facilities or awareness of their risk status. Iran has one of the highest rates of type 2 diabetes among adults aged 20-79, ranking 14th in the world, and it is a national issue funded by government and healthcare ministries. Therefore, developing an intelligent mobile app that people can use for low-cost early detection of type 2 diabetes is crucial.

Objective: This project aims to develop an intelligent mobile app that people can use for low-cost early detection of type 2 diabetes since Iran has one of the highest rates of type 2 diabetes among adults aged 20-79, ranking 14th in the world. This mobile app can be used by people quickly and help them to prevent or reduce the severity of diabetes-related complications as soon as possible. This app can be used in other countries with a high prevalence of type 2 diabetes.

Project Phases: Data acquisition phase involves collecting relevant health information from the target population using various methods such as surveys, blood tests, and wearable devices. Two diabetologist consultants supervise this phase. The data preprocessing phase is crucial for preparing the data for the diagnosis model to involve cleaning, transforming, and analyzing the data using statistical and machine learning techniques

In the data modeling phase, we developed and validated diagnosis models that can accurately diagnose type 2 diabetes based on the data. In the final phase, we developed a mobile app that can provide personalized feedback and recommendations to users based on their risk level and health status. The project hopes to contribute to the global efforts to combat type 2 diabetes, which is one of the leading causes of death and disability in the world.

Outcomes and Impacts: We have created a mobile app for type 2 diabetes. In this app users can track their blood glucose, identify risk factors, detect type 2 diabetes and schedule a diabetologist visit, manage their diet and exercise, remind themselves of their medications, and receive tailored feedback and tips with our app. Our app is based on the latest scientific research and best practices for diabetes management. Our app is compatible with various devices and platforms and syncs with other health apps and devices. Our app empowers users to take charge of their health and well-being.

Our project dataset contains medical and demographic information from the targeted Najafabad Population, which we used to develop and test our diagnosis model. We could help to improve the accuracy and efficiency of diabetes diagnosis and management in this population by applying various machine learning techniques and evaluating their performance. We made this dataset available at the repository for further research and also shared this code and scripts for data analysis and visualization. As part of our planned strategy, we published some of our project's results in a paper that you can find at the bottom of this web page.