Discovering the key symptoms for identifying patterns in functional dyspepsia patients: Doctor's decision and machine learning

Abstract

Background: Pattern identification is a crucial diagnostic process in Traditional East Asian Medicine, classifying patients with similar symptom patterns. This study aims to identify key symptoms for distinguishing patterns in patients with functional dyspepsia (FD) using explicit (doctor's decision-based) and implicit (computational model-based) approaches.

Methods: Data from twenty-one FD patients were collected from local clinics of traditional Korean Medicine and provided to three doctors in a standardized format. Each doctor identified patterns among three types: spleen-stomach weakness, spleen deficiency with qi stagnation/liver-stomach disharmony, and food retention. Doctors evaluated the importance of the symptoms indicated by items in the Standard Tool for Pattern Identification of Functional Dyspepsia questionnaire. Explicit importance was determined through doctors' survey by general evaluation and by selecting specific information used for the diagnosis of patient cases. Implicit importance was assessed by feature importance from the random forest classification models, which classify three types for general differentiation and perform binary classification for specific types.

Results: Key symptoms for distinguishing FD patterns were identified using two approaches. Explicit importance highlighted dietary and nausea-related symptoms, while implicit importance identified complexion or chest tightness as generally crucial. Specific symptoms important for particular pattern types were also identified, and significant correlation between implicit and explicit importance scores was observed for types 1 and 3.

Conclusion: This study showed important clinical information for differentiating FD patients using real patient data. Our findings suggest that these approaches can contribute to developing tools for pattern identification with enhanced accuracy and reliability.

Keywords: Functional dyspepsia; Machine learning; Pattern identification; Symptom importance; Syndrome differentiation.

Fig. 1

Study procedure. Three Korean Medicine doctors completed the survey, which included scoring the overall importance of each questionnaire item on the Standard Tool for Pattern Identification of Functional Dyspepsia (STPI-FD), identifying the patterns of twenty-one FD patients, and ranking the five most important symptoms for identification. Overall importance scores indicate general explicit importance, and ranked symptoms were converted into scores to indicate specific explicit importance. For implicit knowledge, a random forest model was applied using variables from the STPI-FD questionnaire. General importance was derived from feature weights used to classify the three patterns, and specific importance was obtained from feature weights used in the binary classification for each pattern.

Fig. 2

Explicit and implicit importance scores for pattern identification and the patient-reported scores of symptoms. Importance scores for distinguishing three types of patterns in patients with FD are visualized using a heatmap. Red indicates high and blue indicates low scores. In each questionnaire item, A) explicit importance scores for general differentiations and specific to three pattern types, B) implicit importance scores for general differentiations and specific to three pattern types and, C) the mean score of symptoms from questionnaire for each pattern were presented. “v” represents the highest score among the 36 questionnaire items. “−” represents a symptom showing comparatively lower symptom severity with the highest implicit importance score, while “+” represents a symptom showing comparatively higher symptom severity with the highest implicit importance score.

Fig. 3

Radar plots of implicit and explicit importance scores. Distribution of importance scores in explicit and implicit approaches is presented in a radar plot for the three patterns. The green line represents implicit importance score and the blue line represents explicit importance score.