An Apriori Algorithm-Based Association Rule Analysis to Identify Acupoint Combinations for Treating Diabetic...

Area of Science:

• Integrative Medicine
• Data Mining in Healthcare
• Gastroenterology

Background:

• Diabetic gastroparesis (DGP) is a debilitating complication of diabetes.
• Acupuncture is a potential therapeutic option for DGP, but effective acupoint combinations require identification.
• Data mining techniques can uncover patterns in clinical data to guide treatment strategies.

Purpose of the Study:

• To identify frequent acupoint combinations associated with treating diabetic gastroparesis (DGP).
• To apply data mining algorithms, specifically Apriori and frequent pattern growth, to analyze acupoint usage in DGP randomized controlled trials (RCTs).

Main Methods:

• A systematic review of 17 randomized controlled trials (RCTs) on acupuncture for DGP was conducted.
• The Apriori algorithm and a frequent pattern growth algorithm were used to mine association rules among 29 distinct acupoints.
• Frequency patterns of acupoint usage were analyzed to determine significant combinations.

Main Results:

• The top 10 frequently used acupoints included CV12, ST36, PC6, ST25, BL21, BL20, BL23, SP6, BL18, and ST21.
• Key associated acupoint rules identified were {BL23, BL18} ≥ {SP6}, {BL20, BL18} ≥ {PC6}, {PC6, BL18} ≥ {BL20}, and {SP6, BL18} ≥ {BL23}.
• Acupoints BL23, BL18, SP6, BL20, and PC6 emerged as core elements in DGP treatment combinations.

Conclusions:

• Specific acupoint combinations demonstrate significant associations in the treatment of diabetic gastroparesis.
• The identified core acupoints (BL23, BL18, SP6, BL20, PC6) provide a basis for developing standardized acupuncture protocols for DGP.
• Data mining offers a valuable approach for uncovering evidence-based acupoint selection in traditional medicine.