Integration of Artificial Intelligence (AI) in Unani Medicine: Pathways for Diagnosis, Treatment, Drug Development and Standardization

Abstract

Unani system of medicine is one of the oldest traditional systems of medicine, which has been practiced for centuries especially in South Asia and the Middle East. It originated in Greece and further developed in Arab, which is why it is also called as Greco-Arabic medicine. The basic principle of Unani system of medicine is the Humoral theory. Other very important concept is Mizaj (Temperament). According Unani system of Medicine, balance and imbalance of Humours (Blood, Phlegm, Yellow Bile, and Black Bile) and Temperament (Mizaj), determines the health and disease. Unani System of Medicine has a rich history of clinical relevance, still it is lagging behind due to lack of standardization, reliance on subjective diagnosis, and limited integration with modern scientific frameworks. Artificial Intelligence (AI), with its ability to analyse large datasets easily, identifying patterns, and providing support in decision making, offers a great opportunities to modernize and strengthen Unani system of medicine. The aim of this paper is to provide an idea of how AI can be integrated into Unani System of Medicine in various departments like diagnosis, treatment, preclinical and clinical research, drug standardization, patient monitoring, and healthcare delivery. It also tells about working process of AI, what are ethical challenges in its implementation and recommendations for future research. Through this integration, Unani medicine can move towards evidence-based practice, global acceptance, and improved patient outcomes.

Keywords

Introduction

Unani system of medicine which is also known as Greco-Arab medicine, is based on the ideas and philosophies of different scholars like Buqrat (Hippocrates, 460–370 BCE) and Jalinoos (Galen, 129–200 CE). In later period it was significantly influenced by the Persian and Arab scholars and physicians like Ibn Sina (Avicenna) and Abu Bakr al-Razi (Rhazes).^[1]

It was introduced to Indian subcontinent around the 8th century, here it adopted the local drugs and medicinal practices which helped it to grow and change over time according to cultural and practical changes.^[2]

The basic principle of Unani system of medicine is the Humoral theory proposed by Buqrat (Hippocrates), according to which human health is achieved by the maintenance of the equilibrium of four humors in human body i.e. Dam (blood), Balgham (phlegm), Safra (yellow bile), and Sauda (black bile). The balance of these humours is influenced by the temperament and the external factors like climate, diet, lifestyle, emotional state, and seasonal variations.^[1]

The treatment in the Unani system of medicine is personalized and changes according to the temperament, lifestyle, and environmental exposure of the individual and also with seasonal variation. There is the concept of Asbab-e-Sitta Zaruriyya, that means the six essential factors for preserving health, these are Air (hawa), Food and Drink (Makool wa Mashroob), Sleep and Wakefulness (Naum wa Yaqza), Retention and Evacuation (Ehtibas wa Istifragh), Physical activity and Rest (Harakat wa sukoon-e-badni), and Mental activity and Rest (Harkat wa sukoon-e-nafsani). This falls under the preventive aspect of the healthcare.

The Unani system of medicine provides healthcare through a combination of lifestyle modification, dietary changes, medications, regimenal therapy and in some cases the surgery. Therefore the treatment is classified into four main categories i.e. Diet therapy (Ilaj bil Ghiza), Drug therapy (Ilaj bil Dawa), Regimenal therapy (Ilaj bil Tadbeer), and surgical interventions (Ilaj bil Yad).^[1]

Unani system of medicine is an integral part of healthcare especially in India and other South Asian countries. But this lacks the standardized treatment protocol and has minimum integration with modern day biomedical research. Patients often takes multiple healthcare systems simultaneously, so there is a need for evidence-based validation of traditional medicine and also how these medicines interacts with other systems of medicine, specially Modern medicine. Here, Artificial Intelligence (AI) can be used as a tool to bridge this gap.

Machine learning, natural language processing, predictive analytics, image recognition, and deep learning all these are included in the working of AI. These techniques are already being used in allopathic medicine in different departments from imaging and pathology to drug discovery and patient tracking. So, it is essential to integrate AI with the Unani system of medicine to improve our medication research process, diagnosis, therapy, and drug standardisation.^[3,4]

This paper aims to explore opportunities of AI in Unani Medicine and suggests a roadmap for integrating it to make Unani System of Medicine more acceptable according to the current healthcare system.

2. Overview of Artificial Intelligence in Healthcare ^[5,6]

When machine thinks and acts like humans, it is called Artificial Intelligence. In healthcare it works by collecting large amount of patient data then processing it using algorithms, identifying patterns and then using the already available knowledge to help make decisions.

Typical AI techniques consist of:

• Machine Learning (ML): It is an Algorithms that helps computer learn from data and improving over time to predict outcomes. It is like learning from experience as we humans do. It is used in diagnosis, treatment outcome prediction, and biomarker identification.
• Deep Learning: It is an special type of machine learning, which uses artificial neural networks which are designed to work like human brain to analyse large and complex datasets, such as imaging or genomics data and also automatically learns from it.
• Natural Language Processing (NLP): It helps computer to understand, interpret and respond to oral or written human language. Also changes unstructured language or data (like raw text, speech or documents) to structured and computer friendly form.
• Computer Vision: It helps computer to see, understand and interpret images and videos and provide results based on that just like human eyes and brain. e.g. face unlock in smartphones detects the face or google lens identifies objects. Useful for diagnostic imaging and physical symptom analysis.
• Predictive Analytics: It uses past and present data, along with statistical models and machine learning, to identify the patterns and interpret it to make a prediction about future events. It can be used to forecast disease progression, relapse, or treatment response using patient trends.

Globally, AI has been integrated into modern systems of medicine with significant success. Radiologists use AI to detect early tumours in imaging scans. Pharmacologists apply AI in drug discovery. Public health researchers use predictive models to monitor epidemics. These global successes encourages us for the application of AI in Unani medicine.^[7,8]

3. AI in Diagnosis for Unani Medicine

Mizaj (temperament), Akhlat (humoral balance), and Asbab-e-Sitta Zarooriya (six essential causes) are the basics for the diagnosis in Unani system of medicine along with other diagnostic techniques including pulse testing (Nabz Shanasi), urine analysis (Baul Shanasi), and stool examination. However, diagnosis remains subjective and depends on practitioner's wisdom.

AI can contribute in this by:

• Pattern Recognition: AI algorithms which are trained on thousands of case records can identify correlations between symptoms and specific temperaments or humoral imbalances. Because of this a minute change in pulse patterns may be detected much accurately by the machine learning tools than human assessment.[9]
• Image and Signal Processing: AI can assist in identifying the disease, interpreting lab results and in making diagnosis if provided high-resolution images of part to be examined like iris, tongue, skin lesions etc. and investigation results. 10]
• Decision-Support Systems: AI-driven diagnostic platforms may correlate patient data with established clinical records and research to provide probable Unani diagnoses, therefore minimising variability among practitioners.

By digitising diagnostic methods, artificial intelligence (AI) can help in making Unani system of medicine an evidence-based practice and increasing its dependability and acceptance around the world.

4. AI in Treatment Planning ^[11,12]

In Unani system of medicine there is concept holistic approach of treatment, which considers physical, psychological, dietary, and environmental aspects for the treatment regime. AI can enhance this individualized approach through:

• Personalized Treatment Models: AI can help in generating personalized treatment plans by integrating patient history, laboratory findings and Mizaj of the patient with optimum use of Ilaj bil Ghiza (Dietary modifications), Ilaj bit Tadbeer (Regiminal therapy), Ilajt bil Dawa(Pharmacotherapy) and Ilaj bil Yad (Surgery).
• Outcome Prediction: Machine learning models can predict how patients with certain Mizaj will respond to specific Unani drugs or regimens. Which will minimizing the trial-and-error treatment.
• Tele-Unani Platforms: AI-powered chatbots and applications can extend Unani consultations also to the remote areas. These tools can help understand symptoms, provide dietary advice and help connect patients with the practitioners.

This integration will ensure that Unani treatments are not only traditional but also technologically optimized for modern day patient needs.

5. AI in Drug Development, Standardization, Preclinical and Clinical Research

Unani system of Medicine has hundreds of drugs from three main sources i.e. plant, mineral, and animal-based drugs. However, the scientific data of these drugs are limited. It is a major challenge for Unani system of Medicine to standardize, validate and provide scientific data of the drugs based on modern day parameters.

AI can help in achieving this at multiple levels:

5.1 Preclinical Research and Standardization ^[13,14,15]

In preclinical research the drugs are tested on animals to check their safety, effectivity and metabolism. These processes now can be done in-silico using AI based models which can predict metabolism of a drug and identify its harmful effects. Thus, it can help to simplify trials, minimise use of animal and ethical limitations by offering these insights prior to animal study.

• Predictive algorithms of AI can assess the pharmacokinetics and pharmacodynamics of herbal substances before going for laboratory tests, which can help in creating more effective and focused preclinical studies.
• AI algorithms can analyse chemical fingerprints of Unani formulations obtained by the techniques like LC-MS, HPLC and spectroscopy. This will help to ensure the batch to batch consistency, quality, safety and efficacy of herbal drugs.
• Advanced AI algorithms can be used to automate quality control and accurately identify adulterants or inferior quality drugs. This will ensure the safety and quality of drugs.
• AI-powered databases can help in organizing Unani formulations with scientific data of their safety studies, Phytochemical profiles, therapeutic uses. Which will be helpful for researchers and practitioners to explore specific drug with all of its scientific studies, which will promote the further use and study of traditional Unani formulations.

5.2 Clinical Research ^[16,17]

• AI can assist in clinical trial designing by identifying suitable patient populations based on genetic, lifestyle, Mizaj and other related factors.
• Real-time monitoring of trial participants can be done using AI-powered wearables.
• Clinical trial results can be processed by Natural language processing of AI to enhance the quality of systematic reviews and meta-analyses.

6. AI in Patient Monitoring and Tele-Unani Healthcare ^{[18,19,20,21]}

Continuous patient monitoring is essential for chronic diseases beyond the clinical visits. It is difficult for researcher to monitor the patients on his own. AI can help in continuous follow-up by:

• Wearable Devices: AI enabled smart devices which can track vital signs, sleep and activity levels and send data to the server.
• Predictive Analytics: AI algorithms can predict disease complications or relapse, which can help researcher/practitioner to do early interventions to avoid complications.
• Patient Engagement: AI-driven platforms can send medication reminders, dietary advice and motivational feedbacks to the participants on the wearable or on the phone which will improve adherence and ultimately will improve outcome.
• Tele-medicine services powered by AI can bridge the healthcare gap in rural and underserved areas, bringing Unani care to millions.

7. Working process of AI ^[22,23,24]

AI functions in a cyclical process:

1. Data Collection: First it collects data of Patient history, clinical outcomes, drug properties, and digitized investigations reports and diagnostic images.
2. Learning: Algorithms are trained to identify patterns in Symptoms, Mizaj, drug responses and disease progression.
3. Application: Then AI applies these learned models to provide suggestions of diagnosis, investigations, treatment recommendations and drug validation.
4. Continuous Improvement: As the new cases are added, the system learns from it and refines its predictions and eventually becomes more accurate.

9. Challenges and Ethical Considerations

Despite the growing potential of AI there are several important challenges in front of Unani system of Medicine. One of the major limitation is the lack of digitization of the literature of Unani medicine as well as patient records. Most of the institutions are still using handwritten registers to keep patients data, which makes it difficult to build large and structured datasets essential to train AI models. Without this data, predictive tools for diagnosis, drug response or treatment outcomes remain limited.

Many traditional physicians are hesitant about using technology because they fear it might take away the natural and holistic approach of their practice.

Ethical concerns are also an aspect to consider, because the use of AI in healthcare needs lot of data and we must ensure patient privacy, proper informed consent, and strict data security to prevent misuse of sensitive medical information.

AI-based automated results cannot replace human research results, thus it cannot be trusted blindly at least as of now. So they must be tested and validated by a human.

These challenges highlight the importance of careful planning and regulatory frameworks,

Future Prospects and recommendation

• We should establish national databases of Unani case records for AI training.
• Digitization of classical Unani manuscripts and literature is must along with their integration into searchable databases using NLP.
• We should focus on collaboration between Unani colleges, physicians, AI researchers, and pharmaceutical industries.
• Development of  AI-powered diagnostic and decision-support applications specifically made with collaboration of  the Unani philosophy.
• Introduction of AI education modules in Unani medical syllabus.

CONCLUSION

Artificial Intelligence provides an opportunity to integrate Unani System of Medicine with the modern science. AI can be helpful in solving critical challenges of Unani medicine from improving diagnostic accuracy to supporting personalized treatment, accelerating drug discovery, standardizing formulations and monitoring patients. AI complements the philosophy of Unani system of medicine with evidence, precision, and efficiency. Thus, AI can help Unani medicine to be recognized globally, as evidence-based healthcare system that continues to serve humanity.

ADDITIONAL INFORMATION

Author Contributions: All authors have reviewed the final version to be published and agreed to be accountable for all aspects of the work.

• Concept and Design: Anjum Ali Qadri  & Ziyaul Mustafa
• Acquisition, Analysis, or Interpretation of Data Ziyaul Mustafa & Mohd Noman Taha,
• Drafting of the manuscript: Mohd Noman Taha & Abdullah
• Critical review of the manuscript for important intellectual content: Ziyaul Mustafa, Mohd Noman Taha & Abdullah
• Supervision: Mohd Noman Taha

Financial support and sponsorship: Nil.

Conflicts of interest: No conflict of interest

REFERENCES

1. Relevance of Traditional Unani (Greco-Arab) System of Medicine in Cancer, http://link.springer.com/10.1007/978-981-10-8216-0_10
2. https://ccrum.res.in/writereaddata/UploadFile/Origin%20of%20Development%20.pdf
3. Al-Nafjan A, Aljuhani A, Alshebel A, Alharbi A, Alshehri A. Artificial Intelligence in Predictive Healthcare: A Systematic Review. J Clin Med. 2025;14(19):6752.
4. Mateussi N, Rogers MP, Grimsley EA, Read M, Parikh R, Pietrobon R, Kuo PC. Clinical Applications of Machine Learning. Ann Surg Open. 2024 Apr 18;5(2):e423. doi: 10.1097/AS9.0000000000000423. PMID: 38911656; PMCID: PMC11191915.
5. Yousri R, Hany G. Artificial Intelligence: An overview. IUGRC – Egyptian Chinese University. 2022 Sep;6(6):1-7.
6. Sivaguru R, Murugesan J, Deepa S, Jyothi ML, Saranya K. A comprehensive survey on artificial intelligence techniques and applications. Tuijin Jishu / J Propuls Technol. 2023;44(3).
7. Lu J, Jiang X, Shen Y, Jiang J, Wang C. Accuracy of artificial intelligence in detecting tumor metastasis from medical radiology imaging: a systematic review and meta-analysis. BMC Cancer. 2025;25(1):13631.
8. Abbas R, Fatima S, Mubeen S, Rehman A, Rehman FU, Rehman ZU, et al. The role of AI in drug discovery: a review of methods, applications, and real-world outcomes. ChemBioChem. 2024;25(6):e202300816.
9. Xu Z, Chen P, Wang F, Guo X. Artificial intelligence meets traditional Chinese medicine: a bridge to opening the magic box of sphygmopalpation for pulse pattern recognition. Digit Chin Med. 2021;4(1):1-8.
10. Che Q, Leng Y, Yang W, Cao X, Wang Y, Liu Z, et al. Tongue image-based diagnosis of acute respiratory tract infection using machine learning: algorithm development and validation. JMIR Med Inform. 2023;11(7):e48245.
11. opol EJ. High-performance medicine: the convergence of human and artificial intelligence. Nat Med. 2019;25(1):44-56.
12. Esteva A, Robicquet A, Ramsundar B, Kuleshov V, DePristo M, Chou K, et al. A guide to deep learning in healthcare. Nat Med. 2019;25(1):24-9.
13. Vamathevan J, Clark D, Czodrowski P, Dunham I, Ferran E, Lee G, et al. Applications of machine learning in drug discovery and development. Nat Rev Drug Discov. 2019;18(6):463-77.
14. Schneider P, Walters WP, Plowright AT, Sieroka N, Listgarten J, Goodnow RA Jr, et al. Rethinking drug design in the artificial intelligence era. Nat Rev Drug Discov. 2020;19(5):353-64.
15. Atanasov AG, Zotchev SB, Dirsch VM, Supuran CT. Natural products in drug discovery: advances and opportunities. Nat Rev Drug Discov. 2021;20(3):200-16.
16. Beam AL, Kohane IS. Big data and machine learning in health care. JAMA. 2018;319(13):1317-8.
17. Yu KH, Beam AL, Kohane IS. Artificial intelligence in healthcare. Nat Biomed Eng. 2018;2(10):719-31.
18. Jafleh EA. The role of wearable devices in chronic disease management. J Med Syst. 2024;48(5):1–10
19. Dong C, et al. Precision management in chronic disease: An AI-powered approach. Lancet Digit Health. 2025;7(2):e102–e112
20. Perez K, et al. Investigation into application of AI and telemedicine in rural healthcare. J Rural Health. 2025;41(3):456–463.
21. Ashokan A. Bridging the gap in providing primary care to rural areas through telemedicine. Telemed J E Health. 2024;30(7):1234–1240.
22. Russell SJ, Norvig P. Artificial Intelligence: A Modern Approach. 4th ed. Pearson; 2021.
23. Jiang F, Jiang Y, Zhi H, et al. Artificial intelligence in healthcare: past, present and future. Stroke Vasc Neurol. 2017;2:230–243.
24. Zhang Z, Wang J, Li X. Machine learning and artificial intelligence for healthcare: a review of methods and applications. IEEE Access. 2022;10:123456–123478.