The Voice of Chinese Health Consumers: A Text Mining Approach to Web-Based Physician Reviews.

J Med Internet Res. 2016;18(5):e108

Authors: Hao H, Zhang K

Abstract
BACKGROUND: Many Web-based health care platforms allow patients to evaluate physicians by posting open-end textual reviews based on their experiences. These reviews are helpful resources for other patients to choose high-quality doctors, especially in countries like China where no doctor referral systems exist. Analyzing such a large amount of user-generated content to understand the voice of health consumers has attracted much attention from health care providers and health care researchers.
OBJECTIVE: The aim of this paper is to automatically extract hidden topics from Web-based physician reviews using text-mining techniques to examine what Chinese patients have said about their doctors and whether these topics differ across various specialties. This knowledge will help health care consumers, providers, and researchers better understand this information.
METHODS: We conducted two-fold analyses on the data collected from the "Good Doctor Online" platform, the largest online health community in China. First, we explored all reviews from 2006-2014 using descriptive statistics. Second, we applied the well-known topic extraction algorithm Latent Dirichlet Allocation to more than 500,000 textual reviews from over 75,000 Chinese doctors across four major specialty areas to understand what Chinese health consumers said online about their doctor visits.
RESULTS: On the "Good Doctor Online" platform, 112,873 out of 314,624 doctors had been reviewed at least once by April 11, 2014. Among the 772,979 textual reviews, we chose to focus on four major specialty areas that received the most reviews: Internal Medicine, Surgery, Obstetrics/Gynecology and Pediatrics, and Chinese Traditional Medicine. Among the doctors who received reviews from those four medical specialties, two-thirds of them received more than two reviews and in a few extreme cases, some doctors received more than 500 reviews. Across the four major areas, the most popular topics reviewers found were the experience of finding doctors, doctors' technical skills and bedside manner, general appreciation from patients, and description of various symptoms.
CONCLUSIONS: To the best of our knowledge, our work is the first study using an automated text-mining approach to analyze a large amount of unstructured textual data of Web-based physician reviews in China. Based on our analysis, we found that Chinese reviewers mainly concentrate on a few popular topics. This is consistent with the goal of Chinese online health platforms and demonstrates the health care focus in China's health care system. Our text-mining approach reveals a new research area on how to use big data to help health care providers, health care administrators, and policy makers hear patient voices, target patient concerns, and improve the quality of care in this age of patient-centered care. Also, on the health care consumer side, our text mining technique helps patients make more informed decisions about which specialists to see without reading thousands of reviews, which is simply not feasible. In addition, our comparison analysis of Web-based physician reviews in China and the United States also indicates some cultural differences.

PMID: 27165558 [PubMed - in process]

Prioritizing Chemicals for Risk Assessment Using Chemoinformatics: Examples from the IARC Monographs on Pesticides.

Environ Health Perspect. 2016 May 10;

Authors: Guha N, Guyton KZ, Loomis D, Barupal DK

Abstract
BACKGROUND: Identifying cancer hazards is the first step towards cancer prevention. The IARC Monographs Programme, which has evaluated nearly 1000 agents for carcinogenic potential since 1971, typically selects agents for hazard identification on the basis of public nominations, expert advice, published data on carcinogenicity, and public health importance.
OBJECTIVES: Here we present a novel and complementary strategy for identifying agents for hazard evaluation using chemoinformatics, database integration and automated text mining.
DISCUSSION: To inform selection among a broad range of pesticides nominated for evaluation, we identified and screened nearly 6000 relevant chemical structures, thereafter systematically compiled information on 980 pesticides, creating chemical similarity network maps that allowed cluster visualization by chemical similarity, pesticide class, and publicly available information concerning cancer epidemiology, cancer bioassays, and carcinogenic mechanisms. For the IARC Monograph meetings that took place in March and June 2015, this approach supported high priority evaluation of glyphosate, malathion, parathion, tetrachlorvinphos, diazinon, DDT, lindane, and 2,4-D.
CONCLUSIONS: This systematic approach, accounting for chemical similarity and overlaying multiple data sources, can be used by risk assessors as well as researchers to systematize, inform and increase efficiency in selecting and prioritizing agents for hazard identification, risk assessment, regulation or further investigation. This approach could be extended to an array of outcomes and agents, including occupational carcinogens, drugs, and foods.

PMID: 27164621 [PubMed - as supplied by publisher]

Convex biclustering.

Biometrics. 2016 May 10;

Authors: Chi EC, Allen GI, Baraniuk RG

Abstract
In the biclustering problem, we seek to simultaneously group observations and features. While biclustering has applications in a wide array of domains, ranging from text mining to collaborative filtering, the problem of identifying structure in high-dimensional genomic data motivates this work. In this context, biclustering enables us to identify subsets of genes that are co-expressed only within a subset of experimental conditions. We present a convex formulation of the biclustering problem that possesses a unique global minimizer and an iterative algorithm, COBRA, that is guaranteed to identify it. Our approach generates an entire solution path of possible biclusters as a single tuning parameter is varied. We also show how to reduce the problem of selecting this tuning parameter to solving a trivial modification of the convex biclustering problem. The key contributions of our work are its simplicity, interpretability, and algorithmic guarantees-features that arguably are lacking in the current alternative algorithms. We demonstrate the advantages of our approach, which includes stably and reproducibly identifying biclusterings, on simulated and real microarray data.

PMID: 27163413 [PubMed - as supplied by publisher]

BioCreative V CDR task corpus: a resource for chemical disease relation extraction.

Database (Oxford). 2016;2016

Authors: Li J, Sun Y, Johnson RJ, Sciaky D, Wei CH, Leaman R, Davis AP, Mattingly CJ, Wiegers TC, Lu Z

Abstract
Community-run, formal evaluations and manually annotated text corpora are critically important for advancing biomedical text-mining research. Recently in BioCreative V, a new challenge was organized for the tasks of disease named entity recognition (DNER) and chemical-induced disease (CID) relation extraction. Given the nature of both tasks, a test collection is required to contain both disease/chemical annotations and relation annotations in the same set of articles. Despite previous efforts in biomedical corpus construction, none was found to be sufficient for the task. Thus, we developed our own corpus called BC5CDR during the challenge by inviting a team of Medical Subject Headings (MeSH) indexers for disease/chemical entity annotation and Comparative Toxicogenomics Database (CTD) curators for CID relation annotation. To ensure high annotation quality and productivity, detailed annotation guidelines and automatic annotation tools were provided. The resulting BC5CDR corpus consists of 1500 PubMed articles with 4409 annotated chemicals, 5818 diseases and 3116 chemical-disease interactions. Each entity annotation includes both the mention text spans and normalized concept identifiers, using MeSH as the controlled vocabulary. To ensure accuracy, the entities were first captured independently by two annotators followed by a consensus annotation: The average inter-annotator agreement (IAA) scores were 87.49% and 96.05% for the disease and chemicals, respectively, in the test set according to the Jaccard similarity coefficient. Our corpus was successfully used for the BioCreative V challenge tasks and should serve as a valuable resource for the text-mining research community.Database URL: http://www.biocreative.org/tasks/biocreative-v/track-3-cdr/.

PMID: 27161011 [PubMed - in process]

Polypharmacology in Drug Development: A Minireview of Current Technologies.

ChemMedChem. 2016 May 6;

Authors: Tan Z, Chaudhai R, Zhang S

Abstract
Polypharmacology, the process in which a single drug is able to bind to multiple targets specifically and simultaneously, is an emerging paradigm in drug development. The potency of a given drug can be increased through the engagement of multiple targets involved in a certain disease. Polypharmacology may also help identify novel applications of existing drugs through drug repositioning. However, many problems and challenges remain in this field. Rather than covering all aspects of polypharmacology, this Minireview is focused primarily on recently reported techniques, from bioinformatics technologies to cheminformatics approaches as well as text-mining-based methods, all of which have made significant contributions to the research of polypharmacology.

PMID: 27154144 [PubMed - as supplied by publisher]

Building a glaucoma interaction network using a text mining approach.

BioData Min. 2016;9:17

Authors: Soliman M, Nasraoui O, Cooper NG

Abstract
BACKGROUND: The volume of biomedical literature and its underlying knowledge base is rapidly expanding, making it beyond the ability of a single human being to read through all the literature. Several automated methods have been developed to help make sense of this dilemma. The present study reports on the results of a text mining approach to extract gene interactions from the data warehouse of published experimental results which are then used to benchmark an interaction network associated with glaucoma. To the best of our knowledge, there is, as yet, no glaucoma interaction network derived solely from text mining approaches. The presence of such a network could provide a useful summative knowledge base to complement other forms of clinical information related to this disease.
RESULTS: A glaucoma corpus was constructed from PubMed Central and a text mining approach was applied to extract genes and their relations from this corpus. The extracted relations between genes were checked using reference interaction databases and classified generally as known or new relations. The extracted genes and relations were then used to construct a glaucoma interaction network. Analysis of the resulting network indicated that it bears the characteristics of a small world interaction network. Our analysis showed the presence of seven glaucoma linked genes that defined the network modularity. A web-based system for browsing and visualizing the extracted glaucoma related interaction networks is made available at http://neurogene.spd.louisville.edu/GlaucomaINViewer/Form1.aspx.
CONCLUSIONS: This study has reported the first version of a glaucoma interaction network using a text mining approach. The power of such an approach is in its ability to cover a wide range of glaucoma related studies published over many years. Hence, a bigger picture of the disease can be established. To the best of our knowledge, this is the first glaucoma interaction network to summarize the known literature. The major findings were a set of relations that could not be found in existing interaction databases and that were found to be new, in addition to a smaller subnetwork consisting of interconnected clusters of seven glaucoma genes. Future improvements can be applied towards obtaining a better version of this network.

PMID: 27152122 [PubMed]

Data-based Reconstruction of Gene Regulatory Networks of Fungal Pathogens.

Front Microbiol. 2016;7:570

Authors: Guthke R, Gerber S, Conrad T, Vlaic S, Durmuş S, Çakır T, Sevilgen FE, Shelest E, Linde J

Abstract
In the emerging field of systems biology of fungal infection, one of the central roles belongs to the modeling of gene regulatory networks (GRNs). Utilizing omics-data, GRNs can be predicted by mathematical modeling. Here, we review current advances of data-based reconstruction of both small-scale and large-scale GRNs for human pathogenic fungi. The advantage of large-scale genome-wide modeling is the possibility to predict central (hub) genes and thereby indicate potential biomarkers and drug targets. In contrast, small-scale GRN models provide hypotheses on the mode of gene regulatory interactions, which have to be validated experimentally. Due to the lack of sufficient quantity and quality of both experimental data and prior knowledge about regulator-target gene relations, the genome-wide modeling still remains problematic for fungal pathogens. While a first genome-wide GRN model has already been published for Candida albicans, the feasibility of such modeling for Aspergillus fumigatus is evaluated in the present article. Based on this evaluation, opinions are drawn on future directions of GRN modeling of fungal pathogens. The crucial point of genome-wide GRN modeling is the experimental evidence, both used for inferring the networks (omics 'first-hand' data as well as literature data used as prior knowledge) and for validation and evaluation of the inferred network models.

PMID: 27148247 [PubMed]

2024 Jul 31. LiverTox: Clinical and Research Information on Drug-Induced Liver Injury [Internet]. Bethesda (MD): National Institute of Diabetes and Digestive and Kidney Diseases; 2012–.

ABSTRACT

Macular degeneration is an age-related disease of the retina marked by progressive loss of central visual acuity. It is the major cause of visual loss above the age of 60 and risk factors include family history, smoking, being overweight or obese, and hypertension. Age-related macular degeneration occurs in two forms: “dry” and “wet”.

Dry macular degeneration accounts for 80% to 90% of cases and is caused by thinning of the retina, retinal cell loss, and subretinal accumulation of abnormal protein in clumps (drusen). In advanced forms there is patchy atrophy, referred to as geographic atrophy. Recently, agents have been developed for treatment of age-related dry macular degeneration with geographic atrophy. Approved agents include drugs that inhibit complement activation (avacincaptad pegol and pegcetacoplan). The agents are given by intravitreal injections every 1 to 2 months.

Wet macular degeneration accounts for 10% to 20% of cases and is due to neovascularization in the subretinal space with abnormal and leaky blood vessels. The vascularization is dependent, at least in part, on action of vascular endothelial growth factor (VEGF). Wet macular degeneration is treatable with agents that specifically target VEGF which, when given as intravitreal injections, slow the progression of (but do not cure) the neovascularization. These include monoclonal antibodies to VEGF (bevacizumab, brolucizumab, ranibizumab, faricimab), aptamers (small oligonucleotides that bind to VEGF: pegaptanib), and fusion VEGF receptor proteins that act as a decoy of the circulating growth factor (aflibercept). The agents are given as intravitreal injections every 4 to 8 weeks.

Most adverse events of these agents are ocular and relate to their local injection. Systemic exposure is limited and ex-ocular adverse events are rare. Some of the agents have been implicated in cardiovascular or cerebrovascular thromboembolic events, but these are uncommon. None of the drugs for macular degeneration have been implicated in causing hepatotoxicity, either serum enzyme elevations during treatment or clinically apparent liver injury, at least when administered by intravitreal injection. The lack of hepatotoxicity is probably due largely to the lack of significant systemic absorption and exposure. When given intravenously as therapy of neoplastic conditions, several have been linked to rare instances of liver injury.

PMID:31643677 | Bookshelf:NBK548356

2024 Jun 2. LiverTox: Clinical and Research Information on Drug-Induced Liver Injury [Internet]. Bethesda (MD): National Institute of Diabetes and Digestive and Kidney Diseases; 2012–.

ABSTRACT

Metamizole, also known as dipyrone, is an oral analgesic that is not available in the United States but is available over-the-counter in many countries of the world. Therapy with metamizole has been associated with rare severe bone marrow and liver adverse events including agranulocytosis, acute hepatitis, and acute liver failure.

PMID:38861630 | Bookshelf:NBK604194

2023 Jan 28. LiverTox: Clinical and Research Information on Drug-Induced Liver Injury [Internet]. Bethesda (MD): National Institute of Diabetes and Digestive and Kidney Diseases; 2012–.

ABSTRACT

Guarana is an extract of roasted and pulverized seeds of the plant Paullinia cupana which is indigenous to the Amazon Basin and whose major active components are caffeine and other xanthine alkaloids such as theophylline and theobromine. Guarana has been used as a stimulant and tonic to treat fatigue, decrease hunger and thirst and for headaches and dysmenorrhea. In conventional doses, guarana has few side effects and has not been linked to episodes of liver injury or jaundice.

PMID:36753599 | Bookshelf:NBK589113

2022 Feb 7. LiverTox: Clinical and Research Information on Drug-Induced Liver Injury [Internet]. Bethesda (MD): National Institute of Diabetes and Digestive and Kidney Diseases; 2012–.

ABSTRACT

The hepatitis C virus (HCV) is a small RNA virus belonging to the family flaviviridae and genus hepacivirus. The virion is approximately 50 nm in diameter and has an outer lipid associated envelop (E1 and E2) and inner nucleocapsid (Core). Within the nucleocapsid is a single molecule of single-stranded RNA of positive polarity approximately 9.5 kilobases in length. The RNA is transcribed into a large polyprotein that is subsequently cleaved into multiple polypeptides, labeled from the 5’ to 3’ end: core, envelope 1 and 2, and nonstructural proteins NS2, NS3, NS4 and NS5A and NS5B. The NS3 region encodes a viral helicase and protease. The NS5A region encodes a polypeptide that is essential for production and maintenance of the replicative complex. The NS5B region encodes a viral RNA dependent, RNA polymerase that is essential for replication. The NS3, NS5A and NS5B regions have been targeted with direct acting antiviral agents.

The initial agents used to treat chronic hepatitis C were interferon alfa, peginterferon and ribavirin. The antiviral activity of interferon and peginterferon is based upon their ability to stimulate interferon stimulated genes (ISGs) that have endogenous antiviral activities. Ribavirin is a nucleoside analogue that potentiates the effects of interferon against hepatitis C by as yet undefined mechanisms. Until 2010, the standard therapy of chronic hepatitis C was the combination of peginterferon and ribavirin given for 24 or 48 weeks. This combination led to sustained clearance of HCV and remission in disease in 40% to 50% of patients. Response rates were higher with certain HCV genotypes, so that response rates in patients with genotypes 2 and 3 were as high as 70% to 80%. Importantly, these remissions in disease have been shown to represent cure of the chronic viral infection, in that long term follow up demonstrated lack of HCV replication and resolution of disease activity in over 98% of patients. The shortcomings of peginterferon-ribavirin therapy were significant, most importantly the poor tolerance and side effects of this regimen. Thus, a high proportion of patients was intolerant or had contraindications to treatment. In 2010, three HCV-specific protease inhibitors were approved for use and introduced into practice: boceprevir, telaprevir and simeprevir. All three of these were specific to genotype 1 HCV and had little or no activity against genotypes 2 or 3 or the lesser common genotypes 4, 5 and 6. Triple therapy with peginterferon, ribavirin and a HCV-specific protease inhibitor (boceprevir, telaprevir or simeprevir) increased the response rate in patients with chronic hepatitis C, genotype 1 from 40%-45% to 65%-75%. A persistent difficulty, however, was the continued need to combine these agents with peginterferon and the considerable side effects which were worsened by these protease inhibitors.

An important advance in therapy of hepatitis C came in 2013 with the approval of an HCV specific RNA polymerase inhibitor, sofosbuvir. Sofosbuvir not only increased the response rate when combined with peginterferon and ribavirin, but also allowed for interferon-free treatment when combined with ribavirin, HCV protease inhibitors or a new class of agents that antagonized HCV NS5A activity. In 2014, all-oral HCV specific antiviral regimens were approved that yielded response rates in excess of 95% in patients with genotype 1. Furthermore, successful therapy required only 8 to 12 weeks of treatment in most patients and were extremely well tolerated. These all-oral regimens revolutionized therapy of hepatitis C, allowing treatment of virtually all patients regardless of severity of illness or co-morbid conditions with few side effects and durations of therapy of 8, 12 or 24 weeks. Other all oral regimens, including treatments for the less common genotypes of hepatitis C began to become available in 2015, 2016 and 2017. The several classes of agents that are combined in either a two-, three- or four-drug regimens include HCV RNA polymerase inhibitors (nucleoside and nonnucleoside), HCV NS5A antagonists and the HCV protease inhibitors. Several of these drug combinations have been formulated as single tablet or co-packaged regimens. These combination products made therapy easier to apply, but also resulted in the withdrawal of less successful agents, including boceprevir, telaprevir, daclatasvir, simeprevir and the four-drug combination of ombitasvir, dasabuvir, paritaprevir and ritonavir (Viekira Pak). Most currently used regimens are given for 8 to 12 weeks and yield response rates of 98% or more (Epclusa, Mavyret and Zepatier). Widespread application of these therapies to patients with chronic hepatitis C will likely decrease the morbidity and mortality of this disease and make significant inroads into decreasing the burden of chronic liver disease, cirrhosis, and hepatocellular carcinoma worldwide.

PMID:31644192 | Bookshelf:NBK548885

2021 Oct 13. LiverTox: Clinical and Research Information on Drug-Induced Liver Injury [Internet]. Bethesda (MD): National Institute of Diabetes and Digestive and Kidney Diseases; 2012–.

ABSTRACT

Diuretics constitute a large family of medications that increase urine flow and induce urinary sodium loss and are widely used for therapy of hypertension, congestive heart failure, and edematous states. Diuretics in current use (and the year of their approval for use in the United States) include chlorothiazide (1958), hydrochlorothiazide (1959), bendroflumethiazide (1959), spironolactone (1960), chlorthalidone (1960), methyclothiazide (1961), polythiazide (1961), triamterene (1964), furosemide (1966), ethacrynic acid (1967), metolazone (1973), bumetanide (1983), indapamide (1983), amiloride (1986), acetazolamide (1986), torsemide (1993), and eplerenone (2002). Diuretics are typically classified as thiazide diuretics (bendroflumethiazide, chlorothiazide, chlorthalidone, hydrochlorothiazide, indapamide, metolazone and polythiazide), loop diuretics (bumetanide, ethacrynic acid, furosemide, and torsemide), and potassium-sparing agents (amiloride, eplerenone, spironolactone, and triamterene). The carbonic anhydrase blockers acetazolamide (1986) and methazolamide (1959) are also diuretics, but are more commonly used for the therapy of glaucoma.

Diuretics are some of the most frequently used medications in medicine and are usually well tolerated. Common side effects are those that are caused by the diuresis and mineral loss such as weakness, dizziness, electrolyte imbalance, low sodium and potassium. Diuretics have not been associated with an appreciable increased rate of serum aminotransferase elevations and have rarely been associated with clinically apparent liver injury. Isolated case reports of idiosyncratic hepatotoxicity due to diuretics have been published, but there have been virtually no case series on individual diuretics or even whole class of drugs. The case reports that have been published provide only a very general pattern of injury that has not provided a clear clinical signature or suggestion that hepatotoxicity is a class effect among the thiazides and the loop diuretics. Switching from one diuretic to another has not been reported in any systematic fashion. Most information on hepatotoxicity is available on the commonly used diuretics which include (and the number of prescriptions filled in 2007 for each): hydrochlorothiazide (45 million), furosemide (37 million), triamterene (21 million), spironolactone (8 million), and metolazone, bumetanide, indapamide and torsemide (1 to 2 million each). Diuretics implicated in rare cases of drug induced liver injury include hydrochlorothiazide, acetazolamide, amiloride, spironolactone and triamterene.

The thiazide and loop diuretics are discussed as a class; the other diuretics as individual agents. Selected references are given together at the end of this introductory section.

PMID:31644115 | Bookshelf:NBK548808

2021 May 3. LiverTox: Clinical and Research Information on Drug-Induced Liver Injury [Internet]. Bethesda (MD): National Institute of Diabetes and Digestive and Kidney Diseases; 2012–.

ABSTRACT

The Severe Acute Respiratory Syndrome Coronavirus type 2 (SARS-CoV-2) is the cause of the pandemic of coronavirus disease (COVID-19) that was first detected in December 2019 in Wuhan, China and subsequently spread globally. By March 2020, COVID-19 was declared a global pandemic and within a year it accounted for more than 100 million cases and 2 million deaths. Also, within a year of its detection, vaccines against SARS-CoV-2 were developed using several methodologies including mRNA-, adenoviral vector- and recombinant DNA-technology. Several of these vaccines have been evaluated in large, placebo-controlled trials and found to be both safe and effective. Adverse events have been mild-to-moderate local reactions and transient systemic symptoms such as fatigue, nausea and headache. No hepatic specific adverse events have been described, although rare reports of thrombotic thrombocytopenia have occurred with the adenoviral based vaccines that sometimes involve portal or hepatic vein thromboses and some degree of liver dysfunction.

PMID:34014629 | Bookshelf:NBK570468