Prediction of cold and heat patterns using anthropometric measures based on machine learning

Gyejigachulbutang (Gui-Zhi-Jia-Shu-Fu-Tang, Keishikajutsubuto, TJ-18) in Degenerative Knee Osteoarthritis Patients: Lessons and Responders from a Multicenter Randomized Placebo-Controlled Double-Blind Clinical Trial

Background

Gyejigachulbutang (GUI-ZHI-JIA-SHU-FU-TANG, GCB) is an herbal formula widely prescribed in traditional East Asian medicine practice for arthritis and muscle pain. We evaluated the efficacy and safety of GCB for degenerative knee osteoarthritis (KOA).

Methods

Eighty patients with KOA were randomly assigned to the GCB group or the placebo group in a 1 : 1 ratio in two Korean medicine hospitals. Patients took GCB or placebo three times a day for 4 weeks. Primary outcome was the change in the visual analogue scale (VAS) score for knee pain from baseline to 4^th week. Secondary outcomes were the change in the VAS score from baseline to 2^nd week and 8^th week, Korean Western Ontario and McMaster Universities Osteoarthritis Index (K-WOMAC), European Quality of Life Five Dimensions questionnaire (EQ-5D), and safety.

Results

There was no significant difference between the compared indicators of the GCB and placebo groups. However, in subgroup analysis, GCB was effective for subjects with a BMI lower than 25 kg/m². The dose of pain medication was significantly lower in the GCB group than in the placebo group after four weeks (p=0.016). There were no serious adverse events in the GCB group.

Conclusions

GCB was not effective in primary outcome analysis. In exploratory subgroup analysis, GCB might be effective for individuals with BMI lower than 25 kg/m² for the treatment of degenerative KOA. GCB may also help reduce the consumption of pain medication. Furthermore, research is required for our hypothesis. This trial is registered with KCT0003024.

Urinary Function of the Sasang Type and Cold-Heat Subgroup Using the Sasang Urination Inventory in Korean Hospital Patients

Introduction

The Sasang type-specific pathophysiological symptom is pivotal for the Sasang type classification and pattern identification. The Sasang Urination and Defecation Inventory (SUDI) for urinary function analysis was developed; however, the clinical usefulness of urination-related subscales of SUDI in the Sasang type and Cold-Heat subgroup was not reported with acceptable validation analysis.

Methods

The clinical diagnosis of the Sasang type and Cold-Heat subgroup, responses to SUDI items, and weight and height of the 350 hospital patients were acquired retrospectively. The Sasang Urination Inventory (SUI) with SUI-CHR (problematic physical characteristics of urine), SUI-HSS (hypersensitivity of urinary urgency and high frequency), and SUI-DIS (urinary discomfort of hesitancy and residual urine sense) subscales using 12 items of SUDI were improvised. The item and construct validity of the SUI were examined using item response theory and confirmatory factor analysis, and the clinical usefulness of the SUI in Sasang type and Cold-Heat subgroup differentiation was attested.

Results

The SUI and its subscales showed acceptable structural validity and have clinical usefulness in the Tae-Eum type. The Tae-Eum type has a significantly higher SUI-CHR score than did the So-Yang type, and the Heat subgroup has a significantly higher SUI-HSS score than did the Cold subgroup in the Tae-Eum type. Discussion. The distinctive Sasang type- and Cold-Heat subscale-specific pathological symptoms in urinary function were revealed using the SUI. The SUI combined with objective Sasang typology measures might be useful for integrative precision medicine combining Eastern and Western practice and for evidence-based clinical education for medical professions.

Discrimination of prediction models between cold-heat and deficiency-excess patterns

OBJECTIVE

The purpose of this study was to extract important patient questionnaire items by creating random forest models for predicting pattern diagnosis considering an interaction between deficiency-excess and cold-heat patterns.

DESIGN

A multi-centre prospective observational study.

SETTING

Participants visiting six Kampo speciality clinics in Japan from 2012 to 2015.

MAIN OUTCOME MEASURE

Deficiency-excess pattern diagnosis made by board-certified Kampo experts.

METHODS

We used 153 items as independent variables including, age, sex, body mass index, systolic and diastolic blood pressures, and 148 subjective symptoms recorded through a questionnaire. We sampled training data with an equal number of the different patterns from a 2 × 2 factorial combination of deficiency-excess and cold-heat patterns. We constructed the prediction models of deficiency-excess and cold-heat patterns using the random forest algorithm, extracted the top 10 essential items, and calculated the discriminant ratio using this prediction model.

RESULTS

BMI and blood pressure, and subjective symptoms of cold or heat sensations were the most important items in the prediction models of deficiency-excess pattern and of cold-heat patterns, respectively. The discriminant ratio was not inferior compared with the result ignoring the interaction between the diagnoses.

CONCLUSIONS

We revised deficiency-excess and cold-heat pattern prediction models, based on balanced training sample data obtained from six Kampo speciality clinics in Japan. The revised important items for diagnosing a deficiency-excess pattern and cold-heat pattern were compatible with the definition in the 11^th version of international classification of diseases.

Association of Cold-Heat Patterns with Tongue Features, Body Composition, Anthropometric Indices, and Blood Parameters in Tae-Eum Type

Introduction

The purpose of this study was to elucidate the relationship between cold-heat patterns and body composition, anthropometric indices, blood parameters, and tongue features in Tae-Eum type subjects. We also sought to determine whether significant indicators could be used as risk factors for predicting cold-heat patterns in a clinic.

Methods

This prospective, case-control pilot study was conducted at a single center. The subjects were males and females aged 19 years or older who had been analyzed as the Tae-Eum type. After screening, subject allocation was performed. The body composition, 11 anthropometric indices, blood parameters, and tongue features of the subjects were measured by well-trained practitioners. An independent t-test was conducted to compare the cold- and heat-pattern groups. Binary logistic regression was performed to determine significant differences between the two groups after adjusting for age, sex, and systolic blood pressure, with a focus on identifying significant indicators.

Results

Eighty-nine participants were recruited, 39 of whom were excluded from the analysis. Ultimately, 20 cold-pattern and 30 heat-pattern subjects were included in the final analysis. We found that alanine aminotransferase and all anthropometric indices, except for neck and forehead circumference, were significant predictive factors in both groups according to the binary regression analysis. Additionally, the tongue body color in the cold-pattern group was significantly paler than that in the heat-pattern group.

Conclusions

This study found that cold and heat patterns were significantly associated with body composition, anthropometric indices, blood parameters, and tongue features. We suggest that these factors could thus be used as objective indicators and predictors of cold-heat patterns. Our findings provide fundamental but also applicable data that will be useful for diagnosing and monitoring cold-heat patterns in Tae-Eum type patients.

Etiologies of the Relationships Among Body Mass Index and Cold-Heat Patterns: A Twin Study

The phenotypic relationships between body mass index (BMI) and cold-heat patterns have been frequently reported, but the etiology of these relationships remains unknown. We previously demonstrated that the cold pattern (CP) and the heat pattern (HP) were heritable traits. In the present study, we explored underlying genetic and environmental structures of the relationships among BMI and the CP and the HP. Twins (N = 1,752) drawn from the South Korean twin registry completed a cold-heat pattern questionnaire via a telephone interview. The phenotypic correlations among the three phenotypes were moderate but significant. Cross-twin, cross-trait correlations among BMI and the CP and the HP were consistently greater in monozygotic than in dizygotic twins, suggesting the presence of genetic effects on the relationships between BMI and the two patterns. A trivariate Cholesky model was applied to the raw data. The results indicated that the phenotypic relationship between the HP and BMI was completely determined by common genetic influences, while the relationship between the CP and BMI was explained by both common genetic and common individual-specific environmental influences. The genetic correlation between the HP and the CP was not significant, suggesting that the two patterns may be genetically independent from each other. Genetic correlations were 0.31 between the HP and BMI, and -0.22 between the CP and BMI. The individual-specific environmental correlation was -0.22 between HP and CP, and between CP and BMI.

Biopsychological and pathophysiological features of Cold-Heat subgroup of Sasang typology with Sasang Digestive Function Inventory, Sasang Personality Questionnaire and Body Mass Index

Background

The Sasang typology is a traditional Korean personalized medicine and its Cold-Heat subgroup identification is essential for effective use of medical herbs and acupuncture. The purpose of this study was to discover differences between Cold-Heat subgroups with objective clinical measures and to examine its clinical usefulness.

Methods

The pathophysiological symptoms of the digestive system, temperament and body shape of 241 patients were measured using the Sasang Digestive Function Inventory (SDFI), Sasang Personality Questionnaire (SPQ) and Body Mass Index (BMI). The differences between Cold and Heat subgroups of each Sasang types were tested by Analysis of Covariance considering age and sex, while the associations of SDFI, SPQ and BMI with Cold-Heat subgroup were examined by logistic regression analysis.

Results

There were significant differences between Cold and Heat subgroups in SDFI, SPQ and BMI for the So-Yang, SDFI and BMI for the Tae-Eum type and SDFI-Digestion subscale for the So-Eum type. Moreover, the SDFI-Digestion was a substantial predictor for Cold-Heat subgroup identification in three Sasang types. The logistic regression model with SDFI, SPQ and BMI correctly predicted 81.9%, 77% and 75.5% of the Cold-Heat subgroups in So-Yang, Tae-Eum and So-Eum types, respectively.

Conclusion

The results of the present study showed that the objective and validated clinical measures of SDFI, SPQ and BMI would be useful for differentiating Cold-Heat subgroups of Sasang typology. Further clinical studies on pathophysiological mechanisms in Cold-Heat subgroup are required to generalize these results.

The Association of Quantitative Facial Color Features with Cold Pattern in Traditional East Asian Medicine

Introduction

Facial diagnosis is a major component of the diagnostic method in traditional East Asian medicine. We investigated the association of quantitative facial color features with cold pattern using a fully automated facial color parameterization system.

Methods

The facial color parameters of 64 participants were obtained from digital photographs using an automatic color correction and color parameter calculation system. Cold pattern severity was evaluated using a questionnaire.

Results

The a^⁎ values of the whole face, lower cheek, and chin were negatively associated with cold pattern score (CPS) (whole face: B = -1.048, P = 0.021; lower cheek: B = -0.494, P = 0.007; chin: B = -0.640, P = 0.031), while b^⁎ value of the lower cheek was positively associated with CPS (B = 0.234, P = 0.019). The a^⁎ values of the whole face were significantly correlated with specific cold pattern symptoms including cold abdomen (partial ρ = -0.354, P < 0.01) and cold sensation in the body (partial ρ = -0.255, P < 0.05).

Conclusions

a^⁎ values of the whole face were negatively associated with CPS, indicating that individuals with increased levels of cold pattern had paler faces. These findings suggest that objective facial diagnosis has utility for pattern identification.