T

he deployment of AI in the medical field is growing. A recent review on AI published in Complementary and Alternative Medicine (CAM) analyzed Pubmed, Embase, and Cochrane databases to find studies on AI in CAM. AI was used in diverse fields, such as acupressure treatment, tongue and lip diagnosis, herbal medicines, pulse diagnosis, and TCM syndromes^[3]. Another study has reported how thousands of images of Diabetic Retinopathy (DR) were used for training to achieve DR diagnosis with sensitivity and specificity^[4]. Furthermore, there has been more and more AI integration happening in the conventional medical field, and so is introduction of AI in complementary and alternative medicine (CAM) is also being attempted^[5]

AI has potential to transform ancient science of Ayurveda. Opportunities for interdisciplinary research that joins AI and Ayurveda knowledge base can have breakthrough outcomes in terms of the prediction & prevention of diseases and in delivering personalized therapy. The Trisutra (three fundamental databases) of AI for Ayurveda are human data, Ayurveda therapeutics data, and disease data [Figure 1]. The first dataset on humans includes Dosha Prakriti (Somatic constitution) and its characteristics, Sapta Dhatu (major structural components of the body), 13 types of Agni (digestive/metabolic factors), 13 Srotas (structural or functional body channels), 3 Mala (Body waste products), and Oja (the essence of all seven Dhatu). The second set of Ayurveda therapeutics includes medicinal plants with their Rasa (taste), Guna (property/ quality/ attribute), Virya (Potency), Vipaka (bio-transformed rasa), and procedures like Panchakarma (five internal bio-cleansing therapies) as well as Pathyapathya (compatible diet & regimen). Lastly are the disease symptoms. These three datasets, or Trisutra for AI in Ayurveda, could be brought together to diagnose, predict, and prevent disease through drugs, diet, Dinacharya (daily regimen), and Ritucharya (seasonal regimens) concepts taking care of chronobiology.