scholarly journals A Knowledge Graph of Combined Drug Therapies Using Semantic Predications From Biomedical Literature: Algorithm Development

10.2196/18323 ◽  
2020 ◽  
Vol 8 (4) ◽  
pp. e18323
Author(s):  
Jian Du ◽  
Xiaoying Li
Keyword(s):  
Machine Learning ◽  
Clinical Trial ◽  
Clinical Practice ◽  
Practice Guidelines ◽  
Biomedical Literature ◽  
Knowledge Graph ◽  
Combination Therapies ◽  
Combination Drug ◽  
Automated Knowledge ◽  
Semantic Predications

Background Combination therapy plays an important role in the effective treatment of malignant neoplasms and precision medicine. Numerous clinical studies have been carried out to investigate combination drug therapies. Automated knowledge discovery of these combinations and their graphic representation in knowledge graphs will enable pattern recognition and identification of drug combinations used to treat a specific type of cancer, improve drug efficacy and treatment of human disorders. Objective This paper aims to develop an automated, visual approach to discover knowledge about combination therapies from biomedical literature, especially from those studies with high-level evidence such as clinical trial reports and clinical practice guidelines. Methods Based on semantic predications, which consist of a triple structure of subject-predicate-object (SPO), we proposed an automated algorithm to discover knowledge of combination drug therapies using the following rules: 1) two or more semantic predications (S1-P-O and Si-P-O, i = 2, 3…) can be extracted from one conclusive claim (sentence) in the abstract of a given publication, and 2) these predications have an identical predicate (that closely relates to human disease treatment, eg, “treat”) and object (eg, disease name) but different subjects (eg, drug names). A customized knowledge graph organizes and visualizes these combinations, improving the traditional semantic triples. After automatic filtering of broad concepts such as “pharmacologic actions” and generic disease names, a set of combination drug therapies were identified and characterized through manual interpretation. Results We retrieved 22,263 clinical trial reports and 31 clinical practice guidelines from PubMed abstracts by searching “antineoplastic agents” for drug restriction (published between Jan 2009 and Oct 2019). There were 15,603 conclusive claims locally parsed using the search terms “conclusion*” and “conclude*” ready for semantic predications extraction by SemRep, and 325 candidate groups of semantic predications about combined medications were automatically discovered within 316 conclusive claims. Based on manual analysis, we determined that 255/316 claims (78.46%) were accurately identified as describing combination therapies and adopted these to construct the customized knowledge graph. We also identified two categories (and 4 subcategories) to characterize the inaccurate results: limitations of SemRep and limitations of proposal. We further learned the predominant patterns of drug combinations based on mechanism of action for new combined medication studies and discovered 4 obvious markers (“combin*,” “coadministration,” “co-administered,” and “regimen”) to identify potential combination therapies to enable development of a machine learning algorithm. Conclusions Semantic predications from conclusive claims in the biomedical literature can be used to support automated knowledge discovery and knowledge graph construction for combination therapies. A machine learning approach is warranted to take full advantage of the identified markers and other contextual features.

scholarly journals A Knowledge Graph of Combined Drug Therapies Using Semantic Predications From Biomedical Literature: Algorithm Development (Preprint)

2020 ◽  
Author(s):  
Jian Du ◽  
Xiaoying Li
Keyword(s):  
Machine Learning ◽  
Clinical Trial ◽  
Clinical Practice ◽  
Practice Guidelines ◽  
Biomedical Literature ◽  
Knowledge Graph ◽  
Combination Therapies ◽  
Combination Drug ◽  
Automated Knowledge ◽  
Semantic Predications

BACKGROUND Combination therapy plays an important role in the effective treatment of malignant neoplasms and precision medicine. Numerous clinical studies have been carried out to investigate combination drug therapies. Automated knowledge discovery of these combinations and their graphic representation in knowledge graphs will enable pattern recognition and identification of drug combinations used to treat a specific type of cancer, improve drug efficacy and treatment of human disorders. OBJECTIVE This paper aims to develop an automated, visual approach to discover knowledge about combination therapies from biomedical literature, especially from those studies with high-level evidence such as clinical trial reports and clinical practice guidelines. METHODS Based on semantic predications, which consist of a triple structure of subject-predicate-object (SPO), we proposed an automated algorithm to discover knowledge of combination drug therapies using the following rules: 1) two or more semantic predications (S<sub>1</sub>-P-O and S<sub>i</sub>-P-O, i = 2, 3…) can be extracted from one conclusive claim (sentence) in the abstract of a given publication, and 2) these predications have an identical predicate (that closely relates to human disease treatment, eg, “treat”) and object (eg, disease name) but different subjects (eg, drug names). A customized knowledge graph organizes and visualizes these combinations, improving the traditional semantic triples. After automatic filtering of broad concepts such as “pharmacologic actions” and generic disease names, a set of combination drug therapies were identified and characterized through manual interpretation. RESULTS We retrieved 22,263 clinical trial reports and 31 clinical practice guidelines from PubMed abstracts by searching “antineoplastic agents” for drug restriction (published between Jan 2009 and Oct 2019). There were 15,603 conclusive claims locally parsed using the search terms “conclusion*” and “conclude*” ready for semantic predications extraction by SemRep, and 325 candidate groups of semantic predications about combined medications were automatically discovered within 316 conclusive claims. Based on manual analysis, we determined that 255/316 claims (78.46%) were accurately identified as describing combination therapies and adopted these to construct the customized knowledge graph. We also identified two categories (and 4 subcategories) to characterize the inaccurate results: limitations of SemRep and limitations of proposal. We further learned the predominant patterns of drug combinations based on mechanism of action for new combined medication studies and discovered 4 obvious markers (“combin*,” “coadministration,” “co-administered,” and “regimen”) to identify potential combination therapies to enable development of a machine learning algorithm. CONCLUSIONS Semantic predications from conclusive claims in the biomedical literature can be used to support automated knowledge discovery and knowledge graph construction for combination therapies. A machine learning approach is warranted to take full advantage of the identified markers and other contextual features.

scholarly journals Text Classification in Clinical Practice Guidelines Using Machine-Learning Assisted Pattern-Based Approach

2021 ◽  
Vol 11 (8) ◽  
pp. 3296
Author(s):  
Musarrat Hussain ◽  
Jamil Hussain ◽  
Taqdir Ali ◽  
Syed Imran Ali ◽  
Hafiz Syed Muhammad Bilal ◽  
...  
Keyword(s):  
Machine Learning ◽  
Decision Making ◽  
Clinical Practice ◽  
Clinical Practice Guidelines ◽  
Practice Guidelines ◽  
Machine Learning Algorithms ◽  
Nominal Group ◽  
Specific Information ◽  
Matching Techniques ◽  
Disease Specific

Clinical Practice Guidelines (CPGs) aim to optimize patient care by assisting physicians during the decision-making process. However, guideline adherence is highly affected by its unstructured format and aggregation of background information with disease-specific information. The objective of our study is to extract disease-specific information from CPG for enhancing its adherence ratio. In this research, we propose a semi-automatic mechanism for extracting disease-specific information from CPGs using pattern-matching techniques. We apply supervised and unsupervised machine-learning algorithms on CPG to extract a list of salient terms contributing to distinguishing recommendation sentences (RS) from non-recommendation sentences (NRS). Simultaneously, a group of experts also analyzes the same CPG and extract the initial patterns “Heuristic Patterns” using a group decision-making method, nominal group technique (NGT). We provide the list of salient terms to the experts and ask them to refine their extracted patterns. The experts refine patterns considering the provided salient terms. The extracted heuristic patterns depend on specific terms and suffer from the specialization problem due to synonymy and polysemy. Therefore, we generalize the heuristic patterns to part-of-speech (POS) patterns and unified medical language system (UMLS) patterns, which make the proposed method generalize for all types of CPGs. We evaluated the initial extracted patterns on asthma, rhinosinusitis, and hypertension guidelines with the accuracy of 76.92%, 84.63%, and 89.16%, respectively. The accuracy increased to 78.89%, 85.32%, and 92.07% with refined machine-learning assistive patterns, respectively. Our system assists physicians by locating disease-specific information in the CPGs, which enhances the physicians’ performance and reduces CPG processing time. Additionally, it is beneficial in CPGs content annotation.

scholarly journals How Fragile Are Clinical Trial Outcomes That Support the CHEST Clinical Practice Guidelines for VTE?

CHEST Journal ◽  
2018 ◽  
Vol 154 (3) ◽  
pp. 512-520 ◽  
Author(s):  
Elizabeth Edwards ◽  
Cole Wayant ◽  
Jonathan Besas ◽  
Justin Chronister ◽  
Matt Vassar
Keyword(s):  
Clinical Trial ◽  
Clinical Practice ◽  
Clinical Practice Guidelines ◽  
Practice Guidelines ◽  
Trial Outcomes

Primary Stroke Prevention in Nonvalvular Atrial Fibrillation: Implementing the Clinical Trial Findings

1997 ◽  
Vol 31 (10) ◽  
pp. 1187-1196 ◽  
Author(s):  
Patricia A Howard ◽  
Pamela W Duncan
Keyword(s):  
Atrial Fibrillation ◽  
Clinical Trial ◽  
Clinical Trials ◽  
Clinical Practice ◽  
Cost Effectiveness ◽  
Clinical Practice Guidelines ◽  
Practice Guidelines ◽  
Randomized Clinical Trials ◽  
Nonvalvular Atrial Fibrillation ◽  
Risk Patients

OBJECTIVE: To review the clinical trials evaluating warfarin for primary stroke prophylaxis in nonvalvular atrial fibrillation (NVAF), to discuss the relative benefits and risks of warfarin versus aspirin therapy, and to review the clinical practice guidelines and identify potential barriers to their implementation in clinical practice. DATA SOURCES: A MEDLINE literature search was performed to identify clinical trials of antithrombotic therapy for NVAF, clinical practice guidelines, studies evaluating physician practices and attitudes, cost-effectiveness studies, and pertinent review articles. Key search terms included atrial fibrillation, stroke, antithrombotic, warfarin, aspirin, and cost-effectiveness. DATA EXTRACTION: Prospective, randomized clinical trials were selected for analysis. Clinical practice guidelines from recognized panels of experts were reviewed. Comprehensive review articles were selected. DATA SYNTHESIS: NVAF is a common arrhythmia that is associated with a substantial risk for stroke. Seven prospective, randomized, clinical trials have conclusively demonstrated the efficacy of warfarin for stroke prevention. The greatest benefits are achieved in older patients and those with comorbidities that increase their risk for stroke. The potential benefits of preventing a devastating stroke, however, must be weighed against the potential for bleeding complications. Warfarin has been shown to be cost-effective in high-risk patients, provided the rate of complications is minimized. Nonetheless, many physicians remain hesitant to implement warfarin therapy in older, high-risk patients. The clinical data on aspirin are less consistent than those observed with warfarin. Aspirin appears to be most effective in younger individuals or those considered to be at low risk for stroke. CONCLUSIONS: In patients with NVAF, the personal, social, and economic consequences of stroke are often devastating. Clinical trials have provided definitive proof that the risks of stroke can be significantly reduced through the use of appropriate antithrombotic therapy. Despite this evidence and the recommendations of a number of clinical practice guidelines, variations in care exist that continue to place patients at risk. Additional outcomes research is needed to evaluate the impact of the clinical trial findings and practice guidelines on clinical practice and to develop methods for overcoming barriers to implementation.

scholarly journals Reporting Quality of Clinical Practice Guidelines Regarding Gout and Hyperuricemia According to the Right Checklist

2020 ◽  
Author(s):  
Can Wang ◽  
Xufei Luo ◽  
Maichao Li ◽  
Lingling Cui ◽  
Xinde Li ◽  
...  
Keyword(s):  
Clinical Practice ◽  
Clinical Practice Guidelines ◽  
Practice Guidelines ◽  
Guideline Development ◽  
Reporting Quality ◽  
Reporting Rate ◽  
Biomedical Literature ◽  
China National Knowledge Infrastructure ◽  
The Right

Abstract Objectives The Reporting Items for Practice Guidelines in Healthcare (RIGHT) checklist was used to assess the reporting quality of 2009–2019 Clinical Practice Guidelines (CPGs) regarding gout and hyperuricemia, aimed to improve the reporting quality of future guidelines.Methods We searched PubMed, the Chinese Biomedical Literature database, the Wan Fang Database, and the China National Knowledge Infrastructure from January 2009 to June 2019 for relevant guidelines. We also searched the websites of guideline development organizations (the Guidelines International Network, the National Institute for Health and Clinical Excellence, the American College of Rheumatology, and the European League Against Rheumatism) (EULAR). Furthermore, supplementary guidelines reported in included articles were systematically searched, as well as Medlive and Google Scholar. Results Seventeen guidelines were included, of which one was in Chinese and 16 were in English. The mean reporting rate of the 35 items specified was 14.9 (42.5%), only five CPGs (29.4%) had a reporting rate >50%. Of the 35 items, three were very frequently reported. The reporting proportion of the seven domains (Basic information, Background, Evidence, Recommendations, Review and quality assurance, Funding and declaration and management of interests, and Other information) were 64.7%, 36.8%, 50.6% 50.6%, 42.9%, 8.82%, 33.8%, and 31.4%, respectively.Conclusion The reporting quality of the present guidelines for gout and hyperuricemia is relatively poor. We suggest that the RIGHT reporting checklist should be used by CPG developers to ensure higher reporting quality of future guidelines.

scholarly journals Results of clinical studies PLATO and place of ticagrelor in the treatment of acute coronary syndrome

2015 ◽  
Vol 6 (1) ◽  
pp. 53-59
Author(s):  
A. D Erlih
Keyword(s):  
Clinical Trial ◽  
Acute Coronary Syndrome ◽  
Clinical Practice ◽  
Randomized Clinical Trial ◽  
Study Design ◽  
Clinical Practice Guidelines ◽  
Practice Guidelines ◽  
Clinical Studies ◽  
Efficacy And Safety ◽  
Coronary Syndrome

This article provides an analysis of the results of a large international randomized clinical trial PLATO, in which to learn a new antiplatelet ticagrelor compared with clopidogrel in addition to aspirin in patients with acute coronary syndrome (ACS). Material contains a description of the study design, the main results concerning the efficacy and safety of ticagrelor. In addition, the material presented those parts of modern clinical practice guidelines, which describes the location of ticagrelor in the treatment of ACS and which are modified according to a PLATO study.

scholarly journals Implementing the Fatigue Guidelines at One NCCN Member Institution: Process and Outcomes

2007 ◽  
Vol 5 (10) ◽  
pp. 1092-1101 ◽  
Author(s):  
Tami Borneman ◽  
Barbara F. Piper ◽  
Virginia Chih-Yi Sun ◽  
Marianna Koczywas ◽  
Gwen Uman ◽  
...  
Keyword(s):  
Clinical Trial ◽  
Clinical Practice ◽  
Usual Care ◽  
Clinical Practice Guidelines ◽  
Practice Guidelines ◽  
Phase 1 ◽  
Cancer Center ◽  
Chart Audit ◽  
Eligibility Criteria ◽  
Cancer Related Fatigue

Fatigue, despite being the most common and distressing symptom in cancer, is often unrelieved because of numerous patient, provider, and system barriers. The overall purpose of this 5-year prospective clinical trial is to translate the NCCN Cancer-Related Fatigue Clinical Practice Guidelines in Oncology and NCCN Adult Cancer Pain Clinical Practice Guidelines in Oncology into practice and develop a translational interventional model that can be replicated across settings. This article focuses on one NCCN member institution's experience related to the first phase of the NCCN Cancer-Related Fatigue Guidelines implementation, describing usual care compared with evidence-based guidelines. Phase 1 of this 3-phased clinical trial compared the usual care of fatigue with that administered according to the NCCN guidelines. Eligibility criteria included age 18 years or older; English-speaking; diagnosed with breast, lung, colon, or prostate cancer; and fatigue and/or pain ratings of 4 or more on a 0 to 10 screening scale. Research nurses screened all available subjects in a cancer center medical oncology clinic to identify those meeting these criteria. Instruments included the Piper Fatigue Scale, a Fatigue Barriers Scale, a Fatigue Knowledge Scale, and a Fatigue Chart Audit Tool. Descriptive and inferential statistics were used in data analysis. At baseline, 45 patients had fatigue only (≥ 4) and 24 had both fatigue and pain (≥ 4). This combined sample (N = 69) was predominantly Caucasian (65%), female (63%), an average of 60 years old, diagnosed with stage 3 or 4 breast cancer, and undergoing treatment (82%). The most common barriers noted were patients' belief that physicians would introduce the subject of fatigue if it was important (patient barrier); lack of fatigue documentation (professional barrier); and lack of supportive care referrals (system barrier). Findings showed several patient, professional, and system barriers that distinguish usual care from that recommended by the NCCN Cancer-Related Fatigue Guidelines. Phase 2, the intervention model, is designed to decrease these barriers and improve patient outcomes over time, and is in progress.

scholarly journals Quality appraisal of clinical practice guidelines for diabetes mellitus published in China between 2007 and 2017 using the AGREE II instrument

BMJ Open ◽  
2019 ◽  
Vol 9 (9) ◽  
pp. e022392
Author(s):  
Yuting Gao ◽  
Jinjing Wang ◽  
Xufei Luo ◽  
Xiaoyang Song ◽  
Lian Liu ◽  
...  
Keyword(s):  
Diabetes Mellitus ◽  
Clinical Practice ◽  
Correlation Coefficient ◽  
Clinical Practice Guidelines ◽  
Practice Guidelines ◽  
Biomedical Literature ◽  
Agree Ii ◽  
The Mean ◽  
Agree Ii Instrument

ObjectiveThe aim of this study was to systematically evaluate the quality of the clinical practice guidelines (CPGs) for diabetes mellitus published in China over the period of January 2007 to April 2017.MethodsWe searched the China National Knowledge Infrastructure, Chinese Biomedical Literature database, VIP database and WanFang databases and guideline websites for CPGs for diabetes mellitus published between January 2007 and April 2017 in China. Two reviewers independently screened the literature according to the inclusion and exclusion criteria and extracted data. We used the the Appraisal of Guidelines for Research and Evaluation II (AGREE II) tool (Canadian Institutes of Health Research, Ottawa, Canada) to evaluate the quality of the included guidelines, calculated the scores of each domain and evaluated the consistency among the assessors via use of the intragroup correlation coefficient. And then we compared the results with Chinese CPGs and international CPGs. We conducted a subgroup analysis based on different classification criteria and compared scores of each domain subgroup analyses.ResultsA total of 98 guidelines were identified. The correlation coefficient within the group was 0.93, suggesting that the consistency between the evaluators was good. The scores of the six domains of AGREE II were described in median (IQR) as follows: scope and purpose 53.7 (50.0–59.7), stakeholder involvement 31.5 (27.3–37.0), rigour of development 19.1 (15.3–22.2), clarity of presentation 59.3 (50.0–64.8), applicability 18.1 (13.9–25.7) and editorial independence 0.0 (0.0–0.0). The mean score in each domain of quality of Chinese diabetes CPGs was lower than that of CPGs published worldwide but higher than the mean score of Chinese guidelines of all topics. A funding source, the updated version, organisation and publishers of the guidelines and target fields are all the factors influencing the quality of CPGs to a certain degree.ConclusionsA large number of Chinese diabetes CPGs have been produced. Their quality remain unsatisfactorily low compared with CPGs worldwide, there is still room for improvement. Chinese guideline developers should pay more attention to the transparency of methodology, and use the AGREE II instrument to develop and report guidelines.

Sign in / Sign up

Export Citation Format

Share Document