DEVELOPING A MACHINE LEARNING MODEL TO IDENTIFY KEY SYMPTOMS OF TRADITIONAL MEDICINE SYNDROMES IN PATIENTS WITH TYPE 2 DIABETES MELLITUS
Main Article Content
Abstract
Objective: To identify clinically significant symptoms associated with Traditional Medicine (TM) syndromes in patients with type 2 diabetes mellitus using a machine learning model.
Subjects and Methods: A cross-sectional descriptive study was conducted on 326 patients with type 2 diabetes mellitus receiving treatment at Thong Nhat Hospital, Traditional Medicine Hospital in Ho Chi Minh City, and the University Medical Center Ho Chi Minh City – Branch 3. The Boruta-SHAP algorithm was employed to develop a machine learning model for selecting important symptoms based on their contribution to each TM syndrome.
Results: Most patients exhibited concurrent manifestations of four TM syndromes. The most prevalent syndrome was Liver-Kidney Yin Deficiency, with key symptoms including dry mouth, blurred vision, and soreness or pain in the lower back and knees. The least common syndrome was the Phlegm-Heat Accumulation, characterized by key symptoms such as fatigue, palpitations, sticky and foul-smelling stools, and a deep, rapid, and slippery pulse.
Conclusion: Patients with type 2 diabetes mellitus in this study were predominantly elderly, retired, or had ceased working due to severe illness, and exhibited poor glycemic control. Most patients concurrently exhibited four TM syndromes, with Liver-Kidney Yin Deficiency being the most frequent. Furthermore, this study highlights the key symptoms associated with each TM syndrome in patients with type 2 diabetes mellitus.
Article Details
Keywords
type 2 diabetes mellitus, key symptoms, traditional medicine, machine learning, artificial intelligence..
References
[2]. Lian F, Ni Q, Shen Y, Yang S, Piao C, Wang J, Wei J, Duan J, Fang Z, Lu H, Yang G, Zhao L, Song J, Li Q, Zheng Y, Lyu Y, Tong X. International traditional Chinese medicine guideline for diagnostic and treatment principles of diabetes. Ann Palliat Med. 2020 Jul;9(4):2237-2250. doi: 10.21037/apm-19-271
[3]. American Diabetes Association Professional Practice Committee. 2. Diagnosis and Classification of Diabetes: Standards of Care in Diabetes-2024. Diabetes Care. 2024 Jan 1;47(Suppl 1):S20-S42. doi: 10.2337/dc24-S002
[4]. Dương Thị Ngọc Lan, Võ Trọng Tuân, Nguyễn Thị Bay. Bước đầu xây dựng tiêu chuẩn chẩn đoán các hội chứng y học cổ truyền trên bệnh nhân đái tháo đường type 2 bằng mô hình cây tiềm ẩn. Tạp chí Y học Việt Nam. 2024;542(Số chuyên đề):74-81
[5]. Althubaiti A. Sample size determination: A practical guide for health researchers. J Gen Fam Med. 2022 Dec 14;24(2):72-78. doi: 10.1002/jgf2.600
[6]. DeVore GR. Computing the Z Score and Centiles for Cross-sectional Analysis: A Practical Approach. J Ultrasound Med. 2017 Mar;36(3):459-473. doi: 10.7863/ultra.16.03025
[7]. Bùi Hương Giang, Trần Thị Hải Vân. Đặc điểm lâm sàng y học cổ truyền trên bệnh nhân đái tháo đường type 2 tại bệnh viện Y học cổ truyền Bộ Công an 2022-2023. Tạp chí Y học Việt Nam. 2024;535(2): 309-313. doi: 10.51298/vmj.v535i2.8537
[8]. Zhang GD, Liu XX, Liang JL, Hu QM. The Distribution Pattern of Traditional Chinese Medicine Syndromes in 549 Patients with Type 2 Diabetes. Diabetes Metab Syndr Obes. 2021 May 17;14:2209-2216. doi: 10.2147/DMSO.S295351
[9]. Hou C, Cui Y, Xu Y, Wang Y, Hao Y. TCM Syndrome Recognition Model of Type 2 Diabetes Mellitus in Shanghai Based on TCM Inquiry Information. Evid Based Complement Alternat Med. 2022 Mar 15;2022:2843218. doi: 10.1155/2022/2843218
[10]. Liu X, Huang X, Zhao J, Su Y, Shen L, Duan Y, Gong J, Zhang Z, Piao S, Zhu Q, Rong X, Guo J. Application of machine learning in Chinese medicine differentiation of dampness-heat pattern in patients with type 2 diabetes mellitus. Heliyon. 2023 Feb 13;9(2):e13289. doi: 10.1016/j.heliyon.2023.e13289