PSO-FCM based data mining model to predict diabetic disease.

Area of Science:

• Medical Informatics
• Computational Biology
• Data Mining

Background:

• Rising global prevalence of diabetes mellitus necessitates improved diagnostic tools.
• Data mining offers potential for analyzing complex health datasets to improve diabetes prediction.
• Existing methods for diabetes prediction require enhanced accuracy and efficiency.

Purpose of the Study:

• To propose a novel data mining model for forecasting type 2 diabetes mellitus (T2DM).
• To evaluate the performance of the proposed model against existing methods using a benchmark dataset.

Main Methods:

• Development of a hybrid model combining Particle Swarm Optimization (PSO) and Fuzzy Clustering Means (FCM) (PSO-FCM).
• Application of the PSO-FCM model to the Pima Indians Diabetes Database for T2DM prediction.
• Evaluation of model effectiveness using sensitivity, specificity, and accuracy metrics.

Main Results:

• The proposed PSO-FCM model demonstrated superior performance in forecasting T2DM.
• Experimental results showed an 8.26% increase in accuracy compared to other evaluated methods.
• The model achieved high reliability in diagnosing diabetes based on the Pima Indians Diabetes Database.

Conclusions:

• The PSO-FCM model offers a significant advancement in data mining for diabetes prediction.
• The proposed hybrid approach provides greater accuracy and performance than existing models.
• This method holds promise for more effective T2DM diagnosis and management.