Quantitative Assessment of Abdominal Physical Features Associated with Cold Pattern.

Background

Abdominal examination (AE) is an important component of clinical assessment in Traditional East Asian Medicine (TEAM), providing information on abdominal shape, tenderness, stiffness, and skin color. However, conventional AE relies largely on subjective judgment and lacks standardized quantitative indicators. This study aimed to quantitatively characterize abdominal physical features associated with the cold pattern (CP) using objective abdominal examination devices and to explore their potential role in supporting standardized pattern-related assessment.

Methods

A case-control study was conducted including 63 patients with functional dyspepsia and 60 healthy controls. Participants were classified according to CP status using a validated cold-heat pattern identification questionnaire. Abdominal features were quantified using a digital algometer and a depth-based geometric assessment system, measuring algometric (pressure tolerance, indentation depth, stiffness), geometric (abdominal depth and curvature), and chromatic (CIE L*a*b*) parameters. Group differences were analyzed using generalized linear models adjusted for confounders. A two-stage LASSO logistic regression with nested 10-fold cross-validation was applied.

Results

Individuals with CP showed significantly lower pressure tolerance, indentation depth, stiffness, and CIE a* values, along with a flatter abdominal contour. The integrated model achieved a cross-validated ROC-AUC of 0.777 (95% CI, 0.674-0.872), indicating moderate discriminative performance.

Conclusions

Quantitative algometric, geometric, and chromatic abdominal features were significantly associated with CP. Objective abdominal measurements may complement conventional AE by providing quantitative physical indicators that support more standardized and clinically relevant pattern-related assessment. These findings highlight the potential clinical utility of quantitative abdominal evaluation in improving diagnostic consistency. Trial registration: KCT0003369. Registered 23 November 2018.