Diabetes Prediction Machine

Project Using Machine Learning (ML)  With XGboost &  Deep Learning (DL)  With Neural Network (NN)

Diabetes Prediction Model:

1- Logistic Regression Machine Learning Model

2- Black Box model: XGboost

3- Black Box model: Neural Networks


Study Objectives


To formulate a predictive model showing how certain factors may contribute to the development of prediabetes and diabetes.


Using machine learning techniques, we aim to analyze the data, identify patterns, and build a predictive model that can estimate the likelihood of diabetes development based on the provided factors.


Such a model can provide valuable insights for healthcare professionals and researchers in developing targeted interventions and preventative strategies.


The Dataset


The data were collected from the Iraqi society, from the laboratory of Medical City Hospital (Specialist Center for Endocrinology and Diabetes-Al-Kindy Teaching Hospital). Patients' files were taken and data was extracted from them and entered into the database to construct the diabetes dataset. The data consists of patients’ medical information and laboratory analysis. You can find the dataset on the Mendeley website.

The dataset consists of 1000 patients and 14 measured or recorded variables. For more information and walk through the project you can click the button below.

First Click on the bottom below to access the Diabetes Prediction Machine file, then

PLEASE    OPEN    IT    WITH    GOOGLE    COLAB    :)

To see the presentation file (Presented by Prezi and saved as both a Prezi file and pdf),  please check my Github page down below: