Dataset Information

Reliable multi-label learning via conformal predictor and random forest for syndrome differentiation of chronic fatigue in traditional Chinese medicine.

ABSTRACT:

Objective

Chronic Fatigue (CF) still remains unclear about its etiology, pathophysiology, nomenclature and diagnostic criteria in the medical community. Traditional Chinese medicine (TCM) adopts a unique diagnostic method, namely 'bian zheng lun zhi' or syndrome differentiation, to diagnose the CF with a set of syndrome factors, which can be regarded as the Multi-Label Learning (MLL) problem in the machine learning literature. To obtain an effective and reliable diagnostic tool, we use Conformal Predictor (CP), Random Forest (RF) and Problem Transformation method (PT) for the syndrome differentiation of CF.

Methods and materials

In this work, using PT method, CP-RF is extended to handle MLL problem. CP-RF applies RF to measure the confidence level (p-value) of each label being the true label, and then selects multiple labels whose p-values are larger than the pre-defined significance level as the region prediction. In this paper, we compare the proposed CP-RF with typical CP-NBC(Naïve Bayes Classifier), CP-KNN(K-Nearest Neighbors) and ML-KNN on CF dataset, which consists of 736 cases. Specifically, 95 symptoms are used to identify CF, and four syndrome factors are employed in the syndrome differentiation, including 'spleen deficiency', 'heart deficiency', 'liver stagnation' and 'qi deficiency'.

The results

CP-RF demonstrates an outstanding performance beyond CP-NBC, CP-KNN and ML-KNN under the general metrics of subset accuracy, hamming loss, one-error, coverage, ranking loss and average precision. Furthermore, the performance of CP-RF remains steady at the large scale of confidence levels from 80% to 100%, which indicates its robustness to the threshold determination. In addition, the confidence evaluation provided by CP is valid and well-calibrated.

Conclusion

CP-RF not only offers outstanding performance but also provides valid confidence evaluation for the CF syndrome differentiation. It would be well applicable to TCM practitioners and facilitate the utilities of objective, effective and reliable computer-based diagnosis tool.

SUBMITTER: Wang H

PROVIDER: S-EPMC4053362 | biostudies-literature | 2014

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Reliable multi-label learning via conformal predictor and random forest for syndrome differentiation of chronic fatigue in traditional Chinese medicine.

Wang Huazhen H Liu Xin X Lv Bing B Yang Fan F Hong Yanzhu Y

PloS one 20140611 6

<h4>Objective</h4>Chronic Fatigue (CF) still remains unclear about its etiology, pathophysiology, nomenclature and diagnostic criteria in the medical community. Traditional Chinese medicine (TCM) adopts a unique diagnostic method, namely 'bian zheng lun zhi' or syndrome differentiation, to diagnose the CF with a set of syndrome factors, which can be regarded as the Multi-Label Learning (MLL) problem in the machine learning literature. To obtain an effective and reliable diagnostic tool, we use C ...[more]

PMID: 24918430

Similar Datasets

Project description:Background:Traditional, complementary, and alternative medicine (TCAM) has attracted increasing attention in developed countries, but its mainstream status in China, the home of TCAM, is unclear. Over the period of 2004-2016, we analyze the health resources and health resource utilization of traditional medicine in traditional Chinese medicine (TCM) hospitals in China. Methods:Over 2004-2016, we obtained data from all TCM hospitals in all Chinese provinces to create a hospital-based, longitudinal dataset. TCM health resources and their utilization were measured by two outcome variables: (1) primary outcome variables comprising the proportion of TCM physicians, TCM pharmacists, revenue from TCM drugs, and TCM prescriptions and (2) the secondary outcome variables, as proxies of westernization for TCM hospitals, comprising the number of medical equipment above RMB 10,000 and the proportion of surgery in inpatient visits. We used linear regression models with hospital-fixed effects to analyze time trends for the outcome variables. Results:The number of public TCM hospitals remained stable from 2004 to 2016, while the number of private TCM hospitals increased from 294 in 2004 to 1560 in 2016. There was a small percentage increase in the proportion of TCM physicians (0.280%), TCM pharmacists (0.298%), and revenue from Chinese medicines (0.331%) and TCM prescriptions (1.613%) per hospital per year. Chinese drugs accounted for less than a half of the total drug prescriptions, and accordingly, just one-third of the drug revenue was from Chinese medicines at TCM hospitals. The proportions of physicians, pharmacists, revenue from Chinese drug sales, and traditional medicine prescriptions never reach the 60% benchmark target for mainstream in TCM hospitals. As proxies for Western medicine practices in TCM hospitals, the number of medical equipment above RMB 10,000 rapidly rose by over 13 percent per hospital per year, but the proportion of inpatient surgeries declined by 0.830 percentage points per hospital per year, reflecting a mixed trend in the use of Western medicine practices. Conclusion:For the 2004-2016 period, traditional medicine, although making progress towards the mainstream benchmark of 60% TCM services, was still not mainstream at TCM hospitals.

Project description:BackgroundAs the etiology of chronic fatigue syndrome (CFS) is unclear and the treatment is still a big issue. There exists a wide range of literature about acupuncture and moxibustion (AM) for CFS in traditional Chinese medicine (TCM). But there are certain doubts as well in the effectiveness of its treatment due to the lack of a comprehensive and evidence-based medical proof to dispel the misgivings. Current study evaluated systematically the effectiveness of acupuncture and moxibustion treatments on CFS, and clarified the difference among them and Chinese herbal medicine, western medicine and sham-acupuncture.MethodsWe comprehensively reviewed literature including PubMed, EMBASE, Cochrane library, CBM (Chinese Biomedical Literature Database) and CNKI (China National Knowledge Infrastructure) up to May 2016, for RCT clinical research on CFS treated by acupuncture and moxibustion. Traditional direct meta-analysis was adopted to analyze the difference between AM and other treatments. Analysis was performed based on the treatment in experiment and control groups. Network meta-analysis was adopted to make comprehensive comparisons between any two kinds of treatments. The primary outcome was total effective rate, while relative risks (RR) and 95% confidence intervals (CI) were used as the final pooled statistics.ResultsA total of 31 randomized controlled trials (RCTs) were enrolled in analyses. In traditional direct meta-analysis, we found that in comparison to Chinese herbal medicine, CbAM (combined acupuncture and moxibustion, which meant two or more types of acupuncture and moxibustion were adopted) had a higher total effective rate (RR (95% CI), 1.17 (1.09 ~ 1.25)). Compared with Chinese herbal medicine, western medicine and sham-acupuncture, SAM (single acupuncture or single moxibustion) had a higher total effective rate, with RR (95% CI) of 1.22 (1.14 ~ 1.30), 1.51 (1.31-1.74), 5.90 (3.64-9.56). In addition, compared with SAM, CbAM had a higher total effective rate (RR (95% CI), 1.23 (1.12 ~ 1.36)). In network meta-analyses, similar results were recorded. Subsequently, we ranked all treatments from high to low effective rate and the order was CbAM, SAM, Chinese herbal medicine, western medicine and sham-acupuncture.ConclusionsIn the treatment of CFS, CbAM and SAM may have better effect than other treatments. However, the included trials have relatively poor quality, hence high quality studies are needed to confirm our finding.

Dataset Information

Reliable multi-label learning via conformal predictor and random forest for syndrome differentiation of chronic fatigue in traditional Chinese medicine.

Objective

Methods and materials

The results

Conclusion

Publications

Reliable multi-label learning via conformal predictor and random forest for syndrome differentiation of chronic fatigue in traditional Chinese medicine.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets