Extracting relations from traditional Chinese medicine literature via heterogeneous entity networks

Wan, Huaiyu; Moens, Marie Francine; Luyten, Walter; Zhou, Xuezhong; Mei, Qiaozhu; Liu, Lu; Tang, Jie

doi:10.1093/jamia/ocv092

Cited by 28 publications

(24 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…They are a rich bio-resource for drugs of traditional medicinal systems, modern medicines, nutraceuticals, food supplements, folk medicines, pharmaceuticals, intermediates, and chemicals for synthetic drugs (Hammer et al, 1999 ). Medicinal uses are well-described in the Indian Ayurveda (Patwardhan, 2005 ), in Traditional Chinese Medicine (Wan et al, 2016 ), and in various European historical documents (Ginsburg and Deharo, 2011 ). However, indigenous knowledge in a particular region is an important component of traditional medicine, which is widely practiced by the tribal communities throughout India.…”

Section: Discussionmentioning

confidence: 99%

Antimicrobial, Anthelmintic, and Antiviral Activity of Plants Traditionally Used for Treating Infectious Disease in the Similipal Biosphere Reserve, Odisha, India

et al. 2017

Self Cite

View full text Add to dashboard Cite

In the present study, we tested in vitro different parts of 35 plants used by tribals of the Similipal Biosphere Reserve (SBR, Mayurbhanj district, India) for the management of infections. From each plant, three extracts were prepared with different solvents (water, ethanol, and acetone) and tested for antimicrobial (E. coli, S. aureus, C. albicans); anthelmintic (C. elegans); and antiviral (enterovirus 71) bioactivity. In total, 35 plant species belonging to 21 families were recorded from tribes of the SBR and periphery. Of the 35 plants, eight plants (23%) showed broad-spectrum in vitro antimicrobial activity (inhibiting all three test strains), while 12 (34%) exhibited narrow spectrum activity against individual pathogens (seven as anti-staphylococcal and five as anti-candidal). Plants such as Alangium salviifolium, Antidesma bunius, Bauhinia racemosa, Careya arborea, Caseria graveolens, Cleistanthus patulus, Colebrookea oppositifolia, Crotalaria pallida, Croton roxburghii, Holarrhena pubescens, Hypericum gaitii, Macaranga peltata, Protium serratum, Rubus ellipticus, and Suregada multiflora showed strong antibacterial effects, whilst Alstonia scholaris, Butea monosperma, C. arborea, C. pallida, Diospyros malbarica, Gmelina arborea, H. pubescens, M. peltata, P. serratum, Pterospermum acerifolium, R. ellipticus, and S. multiflora demonstrated strong antifungal activity. Plants such as A. salviifolium, A. bunius, Aporosa octandra, Barringtonia acutangula, C. graveolens, C. pallida, C. patulus, G. arborea, H. pubescens, H. gaitii, Lannea coromandelica, M. peltata, Melastoma malabathricum, Millettia extensa, Nyctanthes arbor-tristis, P. serratum, P. acerifolium, R. ellipticus, S. multiflora, Symplocos cochinchinensis, Ventilago maderaspatana, and Wrightia arborea inhibit survival of C. elegans and could be a potential source for anthelmintic activity. Additionally, plants such as A. bunius, C. graveolens, C. patulus, C. oppositifolia, H. gaitii, M. extensa, P. serratum, R. ellipticus, and V. maderaspatana showed anti-enteroviral activity. Most of the plants, whose traditional use as anti-infective agents by the tribals was well supported, show in vitro inhibitory activity against an enterovirus, bacteria (E. coil, S. aureus), a fungus (C. albicans), or a nematode (C. elegans).

show abstract

Section: Discussionmentioning

confidence: 99%

Antimicrobial, Anthelmintic, and Antiviral Activity of Plants Traditionally Used for Treating Infectious Disease in the Similipal Biosphere Reserve, Odisha, India

et al. 2017

Self Cite

View full text Add to dashboard Cite

show abstract

“…The data set of paired herbs was incorporated into the SHT topic modeling process. We conducted treatment pattern mining through LDA, LinkLDA and SHT model, and calculated perplexity on different number of topics, which vary from 10 to 100. Table 2 shows the probability distributions of 5 discovered topics in SHT model.…”

Section: Overall Performance and Discussionmentioning

confidence: 99%

“…This tripartite information network derived more accurate information than linking symptom and herb alone. Wan et al [10] used a heterogeneous factor graph model (HFGM) to infer the multiple types of relations (e.g., herb-syndrome, herbdisease) from the entire corpus of TCM literature. Zhao et al [13] found that a novel machine learning algorithm, minimum reference set-based multiple instance learning, was superior to other machine learning algorithms for TCM syndrome differentiation.…”

Section: Introductionmentioning

confidence: 99%

Discovering Symptom-herb Relationship by Exploiting SHT Topic Model

Wang

Xing

2017

IPSJ Transactions on Bioinformatics

View full text Add to dashboard Cite

TCM has been widely researched through various methods in computer science in past decades, but none digs into huge amount of clinical cases to discover the meaningful treatment patterns between symptoms and herbs. To meet the challenge, we explore the unstructured and intricate experiential data in clinical case, and propose a method to discover the treatment patterns by introducing a novel topic model named SHT (Symptom-Herb Topic model). Combinational rules are incorporated into the learning process. We evaluate our method on 3,765 TCM clinical cases. The experiment validates the effectiveness of our method compared with LDA model and LinkLDA model.

show abstract

“…Furthermore, from Table 3 we can see, there are not many entities about "tongue body" (7), "tongue coating" (10), "pulse" (22) and "both tongue body and tongue coating" (2), while they have large quantity of annotations (3789, 4978, 6148, 793). Hence one can see that the expressions of "tongue body", "tongue coating", "pulse" and "both tongue body and tongue coating" are relatively consistent and frequently-used in TCM clinical records.…”

Section: Examples Of Special Entities and Analysismentioning

confidence: 99%

Constructing a fine-grained entity recognition corpus based on clinical records of Traditional Chinese Medicine

Zhang

Wang

Yang

et al. 2019

Preprint

View full text Add to dashboard Cite

Objective In this paper, we focused on building a fine-grained entity annotation corpus with corresponding annotation guideline of Traditional Chinese Medicine ( TCM ) clinical records, and to provide an effective way to build more corpora of TCM clinical records in the future.Methods: Instead of previous research methods, we proposed four steps approach which is suitable for TCM medical records in our corpus construction work. Firstly, determine the entity types included in this study through sample annotation method; secondly, draft our fine-grained annotation guideline by summarizing the characteristics of the dataset and referring to some existing guidelines; thirdly, update the guideline through iterative annotations way until the inter-annotator agreement (IAA) value exceed 0.9, kappa value was used to measure the IAA; fourthly, comprehensive annotations were performed, if IAA value exceeds 0.9 stably . After above four method steps, we succeeded to construct the fine-grained entity recognition corpus of TCM clinical records.Results: There are 4 entity categories involving 13 entity types being determined finally. The finegrained annotated entity corpus consists of 1104 entities and 67799 tokens totally. The final IAAs are 0.93, 0.94, 0.94 respectively (between two of the three annotators), the IAA value show the finegrained entity recognition corpus are of high quality. We constructed a fine-grained annotated guideline and entity recognition corpus of TCM clinical records.Conclusions: The four-step method was of high quality, the corpus constructed in this study was an encouraging example . Based on this approach, more comprehensive corpus about TCM clinical records will be built to support the TCM named entity recognition tasks in future research.

show abstract

Extracting relations from traditional Chinese medicine literature via heterogeneous entity networks

Abstract: This study exploits the power of collective inference and proposes an HFGM based on heterogeneous entity networks, which significantly improved our ability to extract relations from TCM literature.

Cited by 28 publications

References 37 publications

Antimicrobial, Anthelmintic, and Antiviral Activity of Plants Traditionally Used for Treating Infectious Disease in the Similipal Biosphere Reserve, Odisha, India

Antimicrobial, Anthelmintic, and Antiviral Activity of Plants Traditionally Used for Treating Infectious Disease in the Similipal Biosphere Reserve, Odisha, India

Discovering Symptom-herb Relationship by Exploiting SHT Topic Model

Constructing a fine-grained entity recognition corpus based on clinical records of Traditional Chinese Medicine

Contact Info

Product

Resources

About