Blindness, kidney failure, heart attacks, stroke and loss of lower limbs are the main dangerous complications of diabetes. Even in some cases, type 3 diabetes can cause the death of the patient, so that between 2000 and 2019, the death rate due to type 3 diabetes has increased by 3%. Also, the number of people with diabetes has increased from 108 million people in 1980 to 422 million people in 2014.
You can see more statistical information in the World Health Organization report in April 2023, which can be accessed through this link “ World Health Organization – Diabetes ”, and of course, another important point raised in this report is how to deal with and control diabetes. You can delay or prevent this dangerous disease by controlling the diet, exercising and increasing physical activity and continuously monitoring the patient’s blood sugar status to detect the required drugs and the required dose of each drug.
Artificial intelligence technology and data mining methods can play a vital role in identifying diabetic patients. In a study published in BioMed Research International, accessible from this link ” BioMed Research International – A Novel Approach for Best Parameters Selection and Feature Engineering to Analyze and Detect Diabetes: Machine Learning Insights,” an RFWBP algorithm is presented, which is used with RF algorithms and engineering capabilities and can detect diabetic patients in the early stages of the disease. In this algorithm, 5-fold cross-validation was also used, and its accuracy was surprising. This algorithm has achieved 95.83% accuracy with 5-fold cross-validation and 90.68% without this validation.