The main aim of this project is to develop an intelligent system for the early prediction of diabetes using Machine Learning techniques with high accuracy through the Decision Tree J48 algorithm. Diabetes is one of the most common chronic diseases caused by increased blood glucose levels, which may lead to serious complications such as kidney failure, heart disease, blindness, and stroke if not detected at an early stage. This system combines hardware sensors and machine learning methods to monitor important health parameters and predict the possibility of diabetes effectively. The proposed system consists of a glucose sensor, temperature sensor, and MQ-135 acetone sensor to measure blood glucose level, body temperature, and breath acetone concentration respectively. These sensors are interfaced with an Arduino microcontroller, which collects and transfers the sensor data to a Raspberry Pi processor. The Raspberry Pi acts as the main controlling unit and is programmed using embedded Linux. A 4x4 keypad and selection switches are provided for entering dataset values and selecting automatic or manual operation modes. Using the collected sensor data and user input, the Decision Tree J48 algorithm processes the information and predicts the chances of diabetes in a person. The prediction result is displayed on an LCD screen. By integrating machine learning with embedded systems, the project provides an efficient, low-cost, and user-friendly solution for early diabetes prediction and health monitoring.    

Objectives:

1. To develop an intelligent diabetes prediction system using Machine Learning techniques.
2. To implement the Decision Tree J48 algorithm for accurate early prediction of diabetes.
3. To measure important health parameters such as blood glucose level, body temperature, and breath acetone concentration using sensors.
4. To interface sensors with Arduino and Raspberry Pi for real-time data acquisition and processing.
5. To provide both automatic and manual modes for user-friendly operation.
6. To allow users to enter dataset values through a 4x4 keypad for prediction analysis.
7. To display the diabetes prediction results clearly on an LCD screen.
8. To design a low-cost, efficient, and portable healthcare monitoring system.
9. To improve early diagnosis and reduce the risk of severe diabetes-related complications.
10. To integrate embedded systems and machine learning for smart healthcare applications.

        The major building blocks of the project are:  

1. Power supply.
2. Raspberry Pi processor.
3. Glucosensor.
4. Temperature Sensor
5. Arduino UNO.
6. 4X4 Keypad
7. Selection switch
8. LCD display
9. MQ-135.

    Software’s used:  

1. Raspbian OPERATING SYSTEM.
2. Python Language.
3. Express SCH for Circuit design.
4. Decision Tree algorithm.

    Block diagram:    

video: