Web Application for Screening and Evaluation of Chronic Kidney Disease Using Artificial Intelligence Technology

  • สกุนา ช่างบู่
  • ศนิ บุญญกุล
Keywords: Machine Learning: ML, Supervised Learning: SL, Chronic kidney disease: CKD, Diabetes


The objectives of this research were 1) to study the comparison of the efficiency of the model used to analyze the likelihood of chronic kidney disease in diabetic patients, 2) to study a model to analyze the stage of chronic kidney disease of diabetic patients, and 3) to develop a web application for assessing the stage of disease in patients with chronic kidney disease using artificial intelligence technology based on the principle of Machine Learning type and Supervised Learning. It was found that the appropriate algorithm was the algorithm supports vector machines and can be used to develop web applications by using Python with dataset of 1,528 diabetic patients from the Faculty of Medicine Vajira Hospital. The model has an accuracy at 0.9760, precision at 0.9726, recall at 0.9530, F1 -Score at 0.9627, and AUCs at 0.9701. Web Application Assessment’s average overall result is at 4.45, at a high level