Deep Neuro-Fuzzy Systems for Diabetes Prediction using Siddha Healthcare Practices with Algorithm Comparison
Year: 2025
Type: article
Abstract: Background: Diabetes Mellitus (Type-2) is a common chronic metabolic disease has a high-risk factor for global health. Early detection is critical for effective management and the reduction of associated complications. Siddha medicine, an ancient Indian healthcare system, offers a unique diagnostic approach that includes pulse-based assessments to evaluate the balance of three fundamental principles: Vatha, Pitta, and Kappa. Objective: This research introduces a novel approach to diabetes prediction by combining Siddha healthcare practices with machine learning and deep learning techniques. Method: The method incorporates various machine learning algorithms, including Decision Tree Classifier (DT), Support Vector Machine Classifier (SVM), and Random Forest Classifier (RF), compared with the deep learning technique Artificial Neural Networks (ANN) with Deep Neuro Fuzzy System (DNFS). This study developed a dataset based on Siddha diagnostic principles to extract relevant features, which are then input into the classifiers for diabetes prediction. The effectiveness of each algorithm is assessed using multiple performance metrics, such as precision, recall, accuracy, F1-score, specificity, False Positive Rate (FPR), and True Positive Rate (TPR). Findings: The results demonstrate that ANN had the best performance of 90% on test accuracy and precision and recall of 88% and 91%, respectively and was very close to the RF and SVM. Novelty: This study also demonstrates the medications used in the Siddha medical system, and the concept of combining Siddha with modern deep-learning techniques to predict Diabetes is a useful resource for healthcare professionals. Keywords: Diabetes Prediction, DNFS, Healthcare, Siddha Practices, SVM, Random Forest, Manikkadainool, Decision Tree Classifier, Artificial Neural Network, Deep Learning
Authors: S Jeevarathinam, P. L. Balasubramanian, A. B. Karthick Anand Babu
Institution: Tamil University
Language: English
Subfield: Health Information Management
Field: Health Professions
Domain: Health Sciences
Open Access status: diamond
OpenAlex ID: W4414101890