MULTIMODAL DATA-DRIVEN INTELLIGENT DECISION SUPPORT FOR TRADITIONAL MEDICINE
 https://doi.org/10.65725/RPSET/1/2/001

JOURNAL OF RESEARCH PERSPECTIVES IN MULTIDISCIPLINARY SCIENCE, EDUCATION AND TECHNOLOGY (RPSET)
Volume 1 Issue 2, April – June 2026

Abstract
Traditional medicine systems, including Traditional Chinese Medicine (TCM), Ayurveda, and Unani, have served humanity for centuries through holistic diagnostic approaches that combine visual inspection, pulse examination, patient history, and herbal formulations. However, the subjective nature of these practices often creates inconsistencies in diagnosis and treatment recommendations. With the rapid growth of artificial intelligence and multimodal data fusion techniques, there is now a strong opportunity to develop intelligent decision support systems that can preserve the wisdom of traditional medicine while improving accuracy and reproducibility. This study proposes a multimodal data-driven intelligent decision support framework that integrates tongue images, pulse signals, clinical text records, and herbal prescription databases to assist practitioners in diagnosis and treatment planning. We employed convolutional neural networks for image analysis, recurrent neural networks for pulse signal processing, and transformer-based language models for clinical text understanding. A late-fusion strategy combined these modalities into a unified decision layer. The system was tested on a dataset of 4,820 patient records collected from three traditional medicine clinics during 2023-2024. The proposed model achieved an overall diagnostic accuracy of 91.7%, outperforming single-modality baselines by 8-14%. Results suggest that multimodal fusion significantly improves syndrome differentiation, reduces practitioner variability, and provides interpretable treatment suggestions. This research contributes to bridging the gap between ancient medical wisdom and modern computational intelligence, offering a scalable tool for clinical practice, education, and research in traditional medicine.

Authors: Abdul Kuthus A M, Dr. N. Zackariah

Keywords: Traditional Medicine, Multimodal Learning, Intelligent Decision Support, Deep Learning, Syndrome Differentiation, Healthcare AI