Efficient Automated Disease Diagnosis Using Machine Learning Models.

Area of Science:

• Medical Informatics
• Machine Learning in Healthcare
• Public Health

Background:

• Automated diagnosis models are crucial for early disease detection to reduce mortality rates.
• Supervised learning models have shown promise in developing efficient diagnostic tools.
• Timely intervention based on early diagnosis significantly improves patient outcomes.

Purpose of the Study:

• To design an efficient automated disease diagnosis model for early detection of critical diseases.
• To integrate machine learning for real-time disease analysis via a mobile application.
• To assess the model's utility in assisting medical professionals with timely treatment decisions.

Main Methods:

• Development of an Android application for data input.
• Real-time data analysis using a pre-trained machine learning model deployed in Firebase.
• Utilization of logistic regression for disease prediction computation.
• Training and validation of the model on a dataset encompassing coronavirus, heart disease, and diabetes.

Main Results:

• The proposed model enables real-time disease detection through an Android application.
• Logistic regression effectively performs computations for predicting the risk of selected diseases.
• Comparative analysis suggests the model's potential to support clinical decision-making.

Conclusions:

• The developed automated diagnosis model facilitates early identification of coronavirus, heart disease, and diabetes risks.
• The integration of machine learning in a mobile app offers a practical tool for healthcare providers.
• Early detection and timely medication, supported by this model, can contribute to controlling disease-related death rates.